terraform/internal/command/init_test.go

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

6401 lines
202 KiB
Go
Raw Permalink Normal View History

2014-09-26 19:03:39 -04:00
package command
import (
"bytes"
"context"
2019-05-23 18:21:52 -04:00
"encoding/json"
"fmt"
"io"
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
"log"
"maps"
"net/http"
"net/http/httptest"
"net/url"
2014-09-26 19:30:49 -04:00
"os"
"path/filepath"
"regexp"
"slices"
2017-01-18 23:50:45 -05:00
"strings"
2014-09-26 19:03:39 -04:00
"testing"
"github.com/davecgh/go-spew/spew"
"github.com/google/go-cmp/cmp"
"github.com/hashicorp/cli"
2024-03-26 04:02:09 -04:00
version "github.com/hashicorp/go-version"
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
tfaddr "github.com/hashicorp/terraform-registry-address"
"github.com/zclconf/go-cty/cty"
"github.com/hashicorp/terraform/internal/addrs"
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
"github.com/hashicorp/terraform/internal/backend"
backendInit "github.com/hashicorp/terraform/internal/backend/init"
httpBackend "github.com/hashicorp/terraform/internal/backend/remote-state/http"
"github.com/hashicorp/terraform/internal/backend/remote-state/inmem"
"github.com/hashicorp/terraform/internal/cloud"
"github.com/hashicorp/terraform/internal/command/arguments"
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
"github.com/hashicorp/terraform/internal/command/clistate"
"github.com/hashicorp/terraform/internal/command/views"
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
"github.com/hashicorp/terraform/internal/command/workdir"
"github.com/hashicorp/terraform/internal/configs"
"github.com/hashicorp/terraform/internal/configs/configschema"
"github.com/hashicorp/terraform/internal/depsfile"
"github.com/hashicorp/terraform/internal/getproviders"
"github.com/hashicorp/terraform/internal/providercache"
"github.com/hashicorp/terraform/internal/providers"
testing_provider "github.com/hashicorp/terraform/internal/providers/testing"
"github.com/hashicorp/terraform/internal/states"
"github.com/hashicorp/terraform/internal/states/statefile"
"github.com/hashicorp/terraform/internal/states/statemgr"
2014-09-26 19:03:39 -04:00
)
// cleanString removes newlines, and redundant spaces.
func cleanString(s string) string {
// Replace newlines with a single space.
s = strings.ReplaceAll(s, "\n", " ")
// Remove other special characters like \r, \t
s = strings.ReplaceAll(s, "\r", "")
s = strings.ReplaceAll(s, "\t", "")
// Replace multiple spaces with a single space.
spaceRegex := regexp.MustCompile(`\s+`)
s = spaceRegex.ReplaceAllString(s, " ")
// Trim any leading or trailing spaces.
s = strings.TrimSpace(s)
return s
}
2017-01-18 23:50:45 -05:00
func TestInit_empty(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
2017-01-18 23:50:45 -05:00
os.MkdirAll(td, 0755)
t.Chdir(td)
2017-01-18 23:50:45 -05:00
2014-09-26 19:03:39 -04:00
ui := new(cli.MockUi)
view, done := testView(t)
2014-09-26 19:03:39 -04:00
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
2014-09-26 19:03:39 -04:00
},
}
2017-01-18 23:50:45 -05:00
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
2014-09-26 19:03:39 -04:00
}
exp := views.MessageRegistry[views.OutputInitEmptyMessage].JSONValue
actual := cleanString(done(t).All())
if !strings.Contains(actual, cleanString(exp)) {
t.Fatalf("expected output to be %q\n, got %q", exp, actual)
}
}
func TestInit_only_test_files(t *testing.T) {
// Create a temporary working directory that has only test files and no tf configuration
td := t.TempDir()
os.MkdirAll(td, 0755)
t.Chdir(td)
if _, err := os.Create("main.tftest.hcl"); err != nil {
t.Fatalf("err: %s", err)
}
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
}
exp := views.MessageRegistry[views.OutputInitSuccessCLIMessage].JSONValue
actual := cleanString(done(t).All())
if !strings.Contains(actual, cleanString(exp)) {
t.Fatalf("expected output to be %q\n, got %q", exp, actual)
}
2014-09-26 19:03:39 -04:00
}
PSS: Add alternative, experimental version of `init` command that downloads providers in two stages (#37350) * Add forked version of `run` logic that's only used if experiments are enabled * Reorder actions in experimental init - load in full config before configuring the backend. * Add getProvidersFromConfig method, initially as an exact copy of getProviders * Make getProvidersFromConfig not use state to get providers * Add `appendLockedDependencies` method to `Meta` to allow multi-phase saving to the dep locks file * Update experimental init to use new getProvidersFromConfig method * Add new getProvidersFromState method that only accepts state information as input for getting providers. Use in experimental init and append values to existing deps lock file * Update messages sent to view about provider download phases * Change init to save updates to the deps lock file only once * Make Terraform output report that a lock file _will_ be made after providers are determined from config * Remove use of `ProviderDownloadOutcome`s * Move repeated code into separate method * Change provider download approach: determine if locks changed at point of attempting to update the lockfile, keep record of incomplete providers inside init command struct * Refactor `mergeLockedDependencies` and update test * Add comments to provider download methods * Fix issue where incorrect message ouput to view when downloading providers * Update `mergeLockedDependencies` method to be more generic * Update `getProvidersFromState` method to receive in-progress config locks and merge those with any locks on file. This allows re-use of providers downloaded by `getProvidersFromConfig` in the same init command * Fix config for `TestInit_stateStoreBlockIsExperimental` * Improve testing of mergeLockedDependencies; state locks are always missing version constraints * Add tests for 2 phase provider download * Add test case to cover use of the `-upgrade` flag * Change the message shown when a provider is reused during the second provider download step. When downloading providers described only in the state then the provider may already be downloaded from a previous init (i.e. is recorded in the deps lock file) or downloaded during step 1 of provider download. The message here needs to cover both potential scenarios. * Update mergeLockedDependencies comment * fix: completely remove use of upgrade flag in getProvidersFromState * Fix: avoid nil pointer errors by returning an empty collection of locks when there is no state * Fix: use state store data only in diagnostic * Change how we make PSS experimental - avoid relying on a package level variable that causes tests to interact. * Remove full-stop in view message, update tests * Update span names to be unique * Re-add lost early returns * Remove unused view messages * Add comments to new view messages
2025-08-18 06:20:18 -04:00
func TestInit_two_step_provider_download(t *testing.T) {
cases := map[string]struct {
workDirPath string
flags []string
expectedDownloadMsgs []string
}{
"providers required by only the state file": {
// TODO - should the output indicate that no providers were found in config?
workDirPath: "init-provider-download/state-file-only",
expectedDownloadMsgs: []string{
views.MessageRegistry[views.OutputInitSuccessCLIMessage].JSONValue,
`Initializing provider plugins found in the configuration...
Initializing the backend...`, // No providers found in the configuration so next output is backend-related
`Initializing provider plugins found in the state...
- Finding latest version of hashicorp/random...
- Installing hashicorp/random v9.9.9...`, // The latest version is expected, as state has no version constraints
},
},
"different providers required by config and state": {
workDirPath: "init-provider-download/config-and-state-different-providers",
expectedDownloadMsgs: []string{
views.MessageRegistry[views.OutputInitSuccessCLIMessage].JSONValue,
// Config - this provider is affected by a version constraint
`Initializing provider plugins found in the configuration...
- Finding hashicorp/null versions matching "< 9.0.0"...
- Installing hashicorp/null v1.0.0...
- Installed hashicorp/null v1.0.0`,
// State - the latest version of this provider is expected, as state has no version constraints
`Initializing provider plugins found in the state...
- Finding latest version of hashicorp/random...
- Installing hashicorp/random v9.9.9...`,
},
},
"does not re-download providers that are present in both config and state": {
workDirPath: "init-provider-download/config-and-state-same-providers",
expectedDownloadMsgs: []string{
// Config
`Initializing provider plugins found in the configuration...
- Finding hashicorp/random versions matching "< 9.0.0"...
- Installing hashicorp/random v1.0.0...
- Installed hashicorp/random v1.0.0`,
// State
`Initializing provider plugins found in the state...
- Reusing previous version of hashicorp/random
- Using previously-installed hashicorp/random v1.0.0`,
},
},
"reuses providers already represented in a dependency lock file": {
workDirPath: "init-provider-download/config-state-file-and-lockfile",
expectedDownloadMsgs: []string{
// Config
`Initializing provider plugins found in the configuration...
- Reusing previous version of hashicorp/random from the dependency lock file
- Installing hashicorp/random v1.0.0...
- Installed hashicorp/random v1.0.0`,
// State
`Initializing provider plugins found in the state...
- Reusing previous version of hashicorp/random
- Using previously-installed hashicorp/random v1.0.0`,
},
},
"using the -upgrade flag causes provider download to ignore the lock file": {
workDirPath: "init-provider-download/config-state-file-and-lockfile",
flags: []string{"-upgrade"},
expectedDownloadMsgs: []string{
// Config - lock file is not mentioned due to the -upgrade flag
`Initializing provider plugins found in the configuration...
- Finding hashicorp/random versions matching "< 9.0.0"...
- Installing hashicorp/random v1.0.0...
- Installed hashicorp/random v1.0.0`,
// State - reuses the provider download from the config
`Initializing provider plugins found in the state...
- Reusing previous version of hashicorp/random
- Using previously-installed hashicorp/random v1.0.0`,
},
},
}
for tn, tc := range cases {
t.Run(tn, func(t *testing.T) {
// Create a temporary working directory no tf configuration but has state
td := t.TempDir()
testCopyDir(t, testFixturePath(tc.workDirPath), td)
os.MkdirAll(td, 0755)
t.Chdir(td)
// A provider source containing the random and null providers
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/random": {"1.0.0", "9.9.9"},
"hashicorp/null": {"1.0.0", "9.9.9"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
AllowExperimentalFeatures: true, // Needed to test init changes for PSS project
},
}
args := append(tc.flags, "-enable-pluggable-state-storage-experiment") // Needed to test init changes for PSS project
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
}
actual := cleanString(done(t).All())
for _, downloadMsg := range tc.expectedDownloadMsgs {
if !strings.Contains(cleanString(actual), cleanString(downloadMsg)) {
t.Fatalf("expected output to contain %q\n, got %q", cleanString(downloadMsg), cleanString(actual))
}
}
})
}
}
// Test that an error is returned if users provide the removed directory argument, which was replaced with -chdir
// See: https://github.com/hashicorp/terraform/commit/ca23a096d8c48544b9bfc6dbf13c66488f9b6964
2017-01-18 23:50:45 -05:00
func TestInit_multipleArgs(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
os.MkdirAll(td, 0755)
t.Chdir(td)
2014-09-26 19:03:39 -04:00
ui := new(cli.MockUi)
view, done := testView(t)
2014-09-26 19:03:39 -04:00
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
2014-09-26 19:03:39 -04:00
},
}
2017-01-18 23:50:45 -05:00
args := []string{
"bad",
"bad",
}
2014-09-26 19:03:39 -04:00
if code := c.Run(args); code != 1 {
t.Fatalf("bad: \n%s", done(t).All())
2014-09-26 19:03:39 -04:00
}
expectedMsg := "Did you mean to use -chdir?"
if !strings.Contains(done(t).All(), expectedMsg) {
t.Fatalf("expected the error message to include %q as part of protecting against deprecated additional arguments.",
expectedMsg,
)
}
2014-09-26 19:03:39 -04:00
}
2014-11-04 15:19:39 -05:00
func TestInit_migrateStateAndJSON(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
os.MkdirAll(td, 0755)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{
"-migrate-state=true",
"-json=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 1 {
t.Fatalf("error, -migrate-state and -json should be exclusive: \n%s", testOutput.All())
}
// Check output
checkGoldenReference(t, testOutput, "init-migrate-state-with-json")
}
func TestInit_fromModule_cwdDest(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
os.MkdirAll(td, os.ModePerm)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{
"-from-module=" + testFixturePath("init"),
}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
}
if _, err := os.Stat(filepath.Join(td, "hello.tf")); err != nil {
t.Fatalf("err: %s", err)
}
}
// Regression test to check that Terraform doesn't recursively copy
// a directory when the source module includes the current directory.
// See: https://github.com/hashicorp/terraform/issues/518
func TestInit_fromModule_dstInSrc(t *testing.T) {
// Change to a temporary directory
td := t.TempDir()
t.Chdir(td)
// Create contents
// .
// ├── issue518.tf
// └── foo/
// └── (empty)
if err := os.Mkdir("foo", os.ModePerm); err != nil {
t.Fatal(err)
}
if _, err := os.Create("issue518.tf"); err != nil {
t.Fatalf("err: %s", err)
}
// Instead of using the -chdir flag, we change directory into the directory foo.
// .
// ├── issue518.tf
// └── foo/ << current directory
// └── (empty)
if err := os.Chdir("foo"); err != nil {
t.Fatalf("err: %s", err)
}
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
// The path ./.. includes the current directory foo.
args := []string{
"-from-module=./..",
}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
}
// Assert this outcome
// .
// ├── issue518.tf
// └── foo/ << current directory
// ├── issue518.tf
// └── foo/
// └── (empty)
if _, err := os.Stat(filepath.Join(td, "foo", "issue518.tf")); err != nil {
t.Fatalf("err: %s", err)
}
if _, err := os.Stat(filepath.Join(td, "foo", "foo")); err != nil {
// Note: originally foo was never copied into itself in this scenario,
// but behavior changed sometime around when -chdir replaced legacy positional
// path arguments. We may want to revert to the original behavior in a
// future major release.
// See: https://github.com/hashicorp/terraform/pull/38059
t.Fatalf("err: %s", err)
}
// We don't expect foo to be copied into itself multiple times
_, err := os.Stat(filepath.Join(td, "foo", "foo", "foo"))
if err == nil {
t.Fatal("expected directory ./foo/foo/foo to not exist, but it does")
}
if _, ok := err.(*os.PathError); !ok {
t.Fatalf("unexpected err: %s", err)
}
}
2017-01-18 23:50:45 -05:00
func TestInit_get(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get"), td)
t.Chdir(td)
2017-01-18 23:50:45 -05:00
ui := new(cli.MockUi)
view, done := testView(t)
2017-01-18 23:50:45 -05:00
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
2017-01-18 23:50:45 -05:00
},
}
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
2017-01-18 23:50:45 -05:00
}
// Check output
output := done(t).Stdout()
if !strings.Contains(output, "foo in foo") {
t.Fatalf("doesn't look like we installed module 'foo': %s", output)
2017-01-18 23:50:45 -05:00
}
}
2024-03-26 04:02:09 -04:00
func TestInit_json(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
2024-03-26 04:02:09 -04:00
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get"), td)
t.Chdir(td)
2024-03-26 04:02:09 -04:00
ui := new(cli.MockUi)
view, done := testView(t)
2024-03-26 04:02:09 -04:00
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-json"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
2024-03-26 04:02:09 -04:00
}
// Check output
output := done(t)
checkGoldenReference(t, output, "init-get")
2024-03-26 04:02:09 -04:00
}
func TestInit_getUpgradeModules(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get"), td)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{
"-get=true",
"-upgrade",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("command did not complete successfully:\n%s", testOutput.Stderr())
}
// Check output
if !strings.Contains(testOutput.Stdout(), "Upgrading modules...") {
t.Fatalf("doesn't look like get upgrade: %s", testOutput.Stdout())
}
}
// Test initializing a backend from config (new working directory with no pre-existing backend state file).
func TestInit_backend_initFromConfig(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend"), td)
t.Chdir(td)
2017-01-18 23:50:45 -05:00
ui := new(cli.MockUi)
view, done := testView(t)
2017-01-18 23:50:45 -05:00
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
2017-01-18 23:50:45 -05:00
},
}
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
2017-01-18 23:50:45 -05:00
}
if _, err := os.Stat(filepath.Join(DefaultDataDir, DefaultStateFilename)); err != nil {
t.Fatalf("err: %s", err)
}
}
// Test init when the -backend=false flag is present (backend state file is used instead of the config).
func TestInit_backend_initFromState(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-file-change-to-s3"), td)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{
"-backend=false",
}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
}
// Double check that the successful init above was due to ignoring the config.
// When we don't provide -backend=false there should be an error due to a config change being detected;
// the config specifies an s3 backend instead of local.
args = []string{}
view, done = testView(t)
c.View = view
if code := c.Run(args); code != 1 {
t.Fatalf("bad, expected a 'Backend configuration changed' error but command succeeded : \n%s", done(t).All())
}
}
// regression test for https://github.com/hashicorp/terraform/issues/38027
func TestInit_backend_migration_stateMgr_error(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
t.Chdir(td)
{
// create some state in (implied) local backend
outputCfg := `output "test" { value = "test" }
`
if err := os.WriteFile("output.tf", []byte(outputCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := new(cli.MockUi)
applyView, done := testView(t)
applyCmd := &ApplyCommand{
Meta: Meta{
Ui: ui,
View: applyView,
},
}
code := applyCmd.Run([]string{"-auto-approve"})
testOut := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOut.All())
}
if _, err := os.Stat(DefaultStateFilename); err != nil {
t.Fatalf("err: %s", err)
}
}
{
// attempt to migrate the state to a broken backend
testBackend := new(httpBackend.TestHTTPBackend)
testBackend.SetMethodFunc("GET", func(w http.ResponseWriter, r *http.Request) {
// simulate "broken backend" in the way described in #38027
// i.e. access denied
w.WriteHeader(403)
})
ts := httptest.NewServer(http.HandlerFunc(testBackend.Handle))
t.Cleanup(ts.Close)
backendCfg := fmt.Sprintf(`terraform {
backend "http" {
address = %q
}
}
`, ts.URL)
if err := os.WriteFile("backend.tf", []byte(backendCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := new(cli.MockUi)
initView, done := testView(t)
initCmd := &InitCommand{
Meta: Meta{
Ui: ui,
View: initView,
},
}
code := initCmd.Run([]string{"-migrate-state"})
out := done(t)
if code == 0 {
t.Fatalf("expected migration to fail (gracefully): %s", out.Stdout())
}
expectedErrMsg := "HTTP remote state endpoint invalid auth"
if !strings.Contains(out.Stderr(), expectedErrMsg) {
t.Fatalf("expected error %q, given: %s", expectedErrMsg, out.Stderr())
}
getCalled := testBackend.CallCount("GET")
if getCalled != 1 {
t.Fatalf("expected GET to be called exactly %d, called %d times", 1, getCalled)
}
}
}
func TestInit_backendUnset(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend"), td)
t.Chdir(td)
{
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
log.Printf("[TRACE] TestInit_backendUnset: beginning first init")
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
// Init
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
log.Printf("[TRACE] TestInit_backendUnset: first init complete")
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if _, err := os.Stat(filepath.Join(DefaultDataDir, DefaultStateFilename)); err != nil {
t.Fatalf("err: %s", err)
}
}
{
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
log.Printf("[TRACE] TestInit_backendUnset: beginning second init")
// Unset
if err := os.WriteFile("main.tf", []byte(""), 0644); err != nil {
t.Fatalf("err: %s", err)
}
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-force-copy"}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
log.Printf("[TRACE] TestInit_backendUnset: second init complete")
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if !s.Backend.Empty() {
t.Fatal("should not have backend config")
}
}
}
2017-01-18 23:50:45 -05:00
func TestInit_backendConfigFile(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-file"), td)
t.Chdir(td)
2017-01-18 23:50:45 -05:00
t.Run("good-config-file", func(t *testing.T) {
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "input.config"}
if code := c.Run(args); code != 0 {
2024-03-26 04:02:09 -04:00
t.Fatalf("bad: \n%s", done(t).All())
}
2017-01-18 23:50:45 -05:00
// Read our saved backend config and verify we have our settings
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"hello","workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
})
// the backend config file must not be a full terraform block
t.Run("full-backend-config-file", func(t *testing.T) {
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "backend.config"}
if code := c.Run(args); code != 1 {
t.Fatalf("expected error, got success\n")
}
2024-03-26 04:02:09 -04:00
if !strings.Contains(done(t).All(), "Unsupported block type") {
t.Fatalf("wrong error: %s", done(t).Stderr())
}
})
// the backend config file must match the schema for the backend
t.Run("invalid-config-file", func(t *testing.T) {
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "invalid.config"}
if code := c.Run(args); code != 1 {
t.Fatalf("expected error, got success\n")
}
2024-03-26 04:02:09 -04:00
if !strings.Contains(done(t).All(), "Unsupported argument") {
t.Fatalf("wrong error: %s", done(t).Stderr())
}
})
// missing file is an error
t.Run("missing-config-file", func(t *testing.T) {
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "missing.config"}
if code := c.Run(args); code != 1 {
t.Fatalf("expected error, got success\n")
}
2024-03-26 04:02:09 -04:00
if !strings.Contains(done(t).All(), "Failed to read file") {
t.Fatalf("wrong error: %s", done(t).Stderr())
}
})
// blank filename clears the backend config
t.Run("blank-config-file", func(t *testing.T) {
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config=", "-migrate-state"}
if code := c.Run(args); code != 0 {
2024-03-26 04:02:09 -04:00
t.Fatalf("bad: \n%s", done(t).All())
}
// Read our saved backend config and verify the backend config is empty
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":null,"workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
})
// simulate the local backend having a required field which is not
// specified in the override file
t.Run("required-argument", func(t *testing.T) {
c := &InitCommand{}
schema := &configschema.Block{
Attributes: map[string]*configschema.Attribute{
"path": {
Type: cty.String,
Optional: true,
},
"workspace_dir": {
Type: cty.String,
Required: true,
},
},
}
flagConfigExtra := arguments.NewFlagNameValueSlice("-backend-config")
flagConfigExtra.Set("input.config")
_, diags := c.backendConfigOverrideBody(flagConfigExtra, schema)
if len(diags) != 0 {
t.Errorf("expected no diags, got: %s", diags.Err())
}
})
}
func TestInit_backendConfigFilePowershellConfusion(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-file"), td)
t.Chdir(td)
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
// SUBTLE: when using -flag=value with Powershell, unquoted values are
// broken into separate arguments. This results in the init command
// interpreting the flags as an empty backend-config setting (which is
// semantically valid!) followed by a custom configuration path.
//
// Adding the "=" here forces this codepath to be checked, and it should
// result in an early exit with a diagnostic that the provided
// configuration file is not a diretory.
args := []string{"-backend-config=", "./input.config"}
2024-03-26 04:02:09 -04:00
code := c.Run(args)
output := done(t)
if code != 1 {
t.Fatalf("got exit status %d; want 1\nstderr:\n%s\n\nstdout:\n%s", code, output.Stderr(), output.Stdout())
}
2024-03-26 04:02:09 -04:00
if got, want := output.Stderr(), `Too many command line arguments`; !strings.Contains(got, want) {
t.Fatalf("wrong output\ngot:\n%s\n\nwant: message containing %q", got, want)
}
}
func TestInit_backendReconfigure(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend"), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
ProviderSource: providerSource,
Ui: ui,
View: view,
},
}
// create some state, so the backend has something to migrate.
f, err := os.Create("foo") // this is the path" in the backend config
if err != nil {
t.Fatalf("err: %s", err)
}
err = writeStateForTesting(testState(), f)
f.Close()
if err != nil {
t.Fatalf("err: %s", err)
}
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// now run init again, changing the path.
// The -reconfigure flag prevents init from migrating
// Without -reconfigure, the test fails since the backend asks for input on migrating state
args = []string{"-reconfigure", "-backend-config", "path=changed"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
}
func TestInit_backendConfigFileChange(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-file-change"), td)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "input.config", "-migrate-state"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// Read our saved backend config and verify we have our settings
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"hello","workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
}
func TestInit_backendMigrateWhileLocked(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-migrate-while-locked"), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
ProviderSource: providerSource,
Ui: ui,
View: view,
},
}
// Create some state, so the backend has something to migrate from
f, err := os.Create("local-state.tfstate")
if err != nil {
t.Fatalf("err: %s", err)
}
err = writeStateForTesting(testState(), f)
f.Close()
if err != nil {
t.Fatalf("err: %s", err)
}
// Lock the source state
unlock, err := testLockState(t, testDataDir, "local-state.tfstate")
if err != nil {
t.Fatal(err)
}
defer unlock()
// Attempt to migrate
args := []string{"-backend-config", "input.config", "-migrate-state", "-force-copy"}
if code := c.Run(args); code == 0 {
t.Fatalf("expected nonzero exit code: %s", done(t).Stdout())
}
// Disabling locking should work
args = []string{"-backend-config", "input.config", "-migrate-state", "-force-copy", "-lock=false"}
if code := c.Run(args); code != 0 {
t.Fatalf("expected zero exit code, got %d: %s", code, done(t).Stderr())
}
}
func TestInit_backendConfigFileChangeWithExistingState(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-file-change-migrate-existing"), td)
t.Chdir(td)
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, _ := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
2024-03-26 04:02:09 -04:00
View: view,
},
}
oldState := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
// we deliberately do not provide the answer for backend-migrate-copy-to-empty to trigger error
args := []string{"-migrate-state", "-backend-config", "input.config", "-input=true"}
if code := c.Run(args); code == 0 {
t.Fatal("expected error")
}
// Read our backend config and verify new settings are not saved
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"local-state.tfstate"}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
// without changing config, hash should not change
if oldState.Backend.Hash != state.Backend.Hash {
t.Errorf("backend hash should not have changed\ngot: %d\nwant: %d", state.Backend.Hash, oldState.Backend.Hash)
}
}
func TestInit_backendConfigKV(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-kv"), td)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "path=hello"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
2017-01-18 23:50:45 -05:00
}
// Read our saved backend config and verify we have our settings
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"hello","workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
2017-01-18 23:50:45 -05:00
}
}
2019-05-23 18:21:52 -04:00
func TestInit_backendConfigKVReInit(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-config-kv"), td)
t.Chdir(td)
2019-05-23 18:21:52 -04:00
ui := new(cli.MockUi)
view, done := testView(t)
2019-05-23 18:21:52 -04:00
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
2019-05-23 18:21:52 -04:00
},
}
args := []string{"-backend-config", "path=test"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
2019-05-23 18:21:52 -04:00
}
ui = new(cli.MockUi)
c = &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
2019-05-23 18:21:52 -04:00
},
}
// a second init should require no changes, nor should it change the backend.
args = []string{"-input=false"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
2019-05-23 18:21:52 -04:00
}
// make sure the backend is configured how we expect
configState := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
cfg := map[string]interface{}{}
if err := json.Unmarshal(configState.Backend.ConfigRaw, &cfg); err != nil {
t.Fatal(err)
}
if cfg["path"] != "test" {
t.Fatalf(`expected backend path="test", got path="%v"`, cfg["path"])
}
// override the -backend-config options by settings
args = []string{"-input=false", "-backend-config", "", "-migrate-state"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// make sure the backend is configured how we expect
configState = testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
cfg = map[string]interface{}{}
if err := json.Unmarshal(configState.Backend.ConfigRaw, &cfg); err != nil {
t.Fatal(err)
}
if cfg["path"] != nil {
t.Fatalf(`expected backend path="<nil>", got path="%v"`, cfg["path"])
}
2019-05-23 18:21:52 -04:00
}
func TestInit_backendConfigKVReInitWithConfigDiff(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend"), td)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-input=false"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
ui = new(cli.MockUi)
c = &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
// a second init with identical config should require no changes, nor
// should it change the backend.
args = []string{"-input=false", "-backend-config", "path=foo"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// make sure the backend is configured how we expect
configState := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
cfg := map[string]interface{}{}
if err := json.Unmarshal(configState.Backend.ConfigRaw, &cfg); err != nil {
t.Fatal(err)
}
if cfg["path"] != "foo" {
t.Fatalf(`expected backend path="foo", got path="%v"`, cfg["foo"])
}
}
func TestInit_backendCli_no_config_block(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init"), td)
t.Chdir(td)
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "path=test"}
if code := c.Run(args); code != 0 {
t.Fatalf("got exit status %d; want 0\nstderr:\n%s\n\nstdout:\n%s", code, done(t).Stderr(), done(t).Stdout())
}
2024-03-26 04:02:09 -04:00
errMsg := done(t).All()
if !strings.Contains(errMsg, "Warning: Missing backend configuration") {
t.Fatal("expected missing backend block warning, got", errMsg)
}
}
func TestInit_backendReinitWithExtra(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend-empty"), td)
t.Chdir(td)
m := testMetaBackend(t, nil)
opts := &BackendOpts{
ConfigOverride: configs.SynthBody("synth", map[string]cty.Value{
"path": cty.StringVal("hello"),
}),
Init: true,
}
_, cHash, err := m.backendConfig(opts)
if err != nil {
t.Fatal(err)
}
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-backend-config", "path=hello"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// Read our saved backend config and verify we have our settings
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"hello","workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
2018-12-18 19:06:49 -05:00
if state.Backend.Hash != uint64(cHash) {
t.Fatal("mismatched state and config backend hashes")
}
// init again and make sure nothing changes
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
state = testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"hello","workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
2018-12-18 19:06:49 -05:00
if state.Backend.Hash != uint64(cHash) {
t.Fatal("mismatched state and config backend hashes")
}
}
// move option from config to -backend-config args
func TestInit_backendReinitConfigToExtra(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend"), td)
t.Chdir(td)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
if code := c.Run([]string{"-input=false"}); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// Read our saved backend config and verify we have our settings
state := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"foo","workspace_dir":null}`; got != want {
t.Errorf("wrong config\ngot: %s\nwant: %s", got, want)
}
backendHash := state.Backend.Hash
// init again but remove the path option from the config
cfg := "terraform {\n backend \"local\" {}\n}\n"
if err := os.WriteFile("main.tf", []byte(cfg), 0644); err != nil {
t.Fatal(err)
}
// We need a fresh InitCommand here because the old one now has our configuration
// file cached inside it, so it won't re-read the modification we just made.
c = &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-input=false", "-backend-config=path=foo"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
state = testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if got, want := normalizeJSON(t, state.Backend.ConfigRaw), `{"path":"foo","workspace_dir":null}`; got != want {
t.Errorf("wrong config after moving to arg\ngot: %s\nwant: %s", got, want)
}
if state.Backend.Hash == backendHash {
t.Fatal("state.Backend.Hash was not updated")
}
}
func TestInit_backendCloudInvalidOptions(t *testing.T) {
// There are various "terraform init" options that are only for
// traditional backends and not applicable to HCP Terraform mode.
// For those, we want to return an explicit error rather than
// just silently ignoring them, so that users will be aware that
// Cloud mode has more of an expected "happy path" than the
// less-vertically-integrated backends do, and to avoid these
// unapplicable options becoming compatibility constraints for
// future evolution of Cloud mode.
// We use the same starting fixture for all of these tests, but some
// of them will customize it a bit as part of their work.
setupTempDir := func(t *testing.T) {
t.Helper()
td := t.TempDir()
testCopyDir(t, testFixturePath("init-cloud-simple"), td)
t.Chdir(td)
}
// Some of the tests need a non-empty placeholder state file to work
// with.
fakeState := states.BuildState(func(cb *states.SyncState) {
// Having a root module output value should be enough for this
// state file to be considered "non-empty" and thus a candidate
// for migration.
cb.SetOutputValue(
addrs.OutputValue{Name: "a"}.Absolute(addrs.RootModuleInstance),
cty.True,
false,
)
})
fakeStateFile := &statefile.File{
Lineage: "boop",
Serial: 4,
TerraformVersion: version.Must(version.NewVersion("1.0.0")),
State: fakeState,
}
var fakeStateBuf bytes.Buffer
err := statefile.WriteForTest(fakeStateFile, &fakeStateBuf)
if err != nil {
t.Error(err)
}
fakeStateBytes := fakeStateBuf.Bytes()
t.Run("-backend-config", func(t *testing.T) {
setupTempDir(t)
// We have -backend-config as a pragmatic way to dynamically set
// certain settings of backends that tend to vary depending on
// where Terraform is running, such as AWS authentication profiles
// that are naturally local only to the machine where Terraform is
// running. Those needs don't apply to HCP Terraform, because
// the remote workspace encapsulates all of the details of how
// operations and state work in that case, and so the Cloud
// configuration is only about which workspaces we'll be working
// with.
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-backend-config=anything"}
if code := c.Run(args); code == 0 {
t.Fatalf("unexpected success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
wantStderr := `
Error: Invalid command-line option
The -backend-config=... command line option is only for state backends, and
is not applicable to HCP Terraform-based configurations.
To change the set of workspaces associated with this configuration, edit the
Cloud configuration block in the root module.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
t.Run("-reconfigure", func(t *testing.T) {
setupTempDir(t)
// The -reconfigure option was originally imagined as a way to force
// skipping state migration when migrating between backends, but it
// has a historical flaw that it doesn't work properly when the
// initial situation is the implicit local backend with a state file
// present. The HCP Terraform migration path has some additional
// steps to take care of more details automatically, and so
// -reconfigure doesn't really make sense in that context, particularly
// with its design bug with the handling of the implicit local backend.
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-reconfigure"}
if code := c.Run(args); code == 0 {
t.Fatalf("unexpected success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
wantStderr := `
Error: Invalid command-line option
The -reconfigure option is for in-place reconfiguration of state backends
only, and is not needed when changing HCP Terraform settings.
When using HCP Terraform, initialization automatically activates any new
Cloud configuration settings.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
t.Run("-reconfigure when migrating in", func(t *testing.T) {
setupTempDir(t)
// We have a slightly different error message for the case where we
// seem to be trying to migrate to HCP Terraform with existing
// state or explicit backend already present.
if err := os.WriteFile("terraform.tfstate", fakeStateBytes, 0644); err != nil {
t.Fatal(err)
}
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-reconfigure"}
if code := c.Run(args); code == 0 {
t.Fatalf("unexpected success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
wantStderr := `
Error: Invalid command-line option
The -reconfigure option is unsupported when migrating to HCP Terraform,
because activating HCP Terraform involves some additional steps.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
t.Run("-migrate-state", func(t *testing.T) {
setupTempDir(t)
// In Cloud mode, migrating in or out always proposes migrating state
// and changing configuration while staying in cloud mode never migrates
// state, so this special option isn't relevant.
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-migrate-state"}
if code := c.Run(args); code == 0 {
t.Fatalf("unexpected success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
wantStderr := `
Error: Invalid command-line option
The -migrate-state option is for migration between state backends only, and
is not applicable when using HCP Terraform.
State storage is handled automatically by HCP Terraform and so the state
storage location is not configurable.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
t.Run("-migrate-state when migrating in", func(t *testing.T) {
setupTempDir(t)
// We have a slightly different error message for the case where we
// seem to be trying to migrate to HCP Terraform with existing
// state or explicit backend already present.
if err := os.WriteFile("terraform.tfstate", fakeStateBytes, 0644); err != nil {
t.Fatal(err)
}
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-migrate-state"}
if code := c.Run(args); code == 0 {
t.Fatalf("unexpected success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
wantStderr := `
Error: Invalid command-line option
The -migrate-state option is for migration between state backends only, and
is not applicable when using HCP Terraform.
HCP Terraform migrations have additional steps, configured by interactive
prompts.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
t.Run("-force-copy", func(t *testing.T) {
setupTempDir(t)
// In Cloud mode, migrating in or out always proposes migrating state
// and changing configuration while staying in cloud mode never migrates
// state, so this special option isn't relevant.
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-force-copy"}
if code := c.Run(args); code == 0 {
t.Fatalf("unexpected success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
wantStderr := `
Error: Invalid command-line option
The -force-copy option is for migration between state backends only, and is
not applicable when using HCP Terraform.
State storage is handled automatically by HCP Terraform and so the state
storage location is not configurable.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
t.Run("-force-copy when migrating in", func(t *testing.T) {
setupTempDir(t)
// We have a slightly different error message for the case where we
// seem to be trying to migrate to HCP Terraform with existing
// state or explicit backend already present.
if err := os.WriteFile("terraform.tfstate", fakeStateBytes, 0644); err != nil {
t.Fatal(err)
}
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{"-force-copy"}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("unexpected success\n%s", testOutput.Stdout())
}
gotStderr := testOutput.Stderr()
wantStderr := `
Error: Invalid command-line option
The -force-copy option is for migration between state backends only, and is
not applicable when using HCP Terraform.
HCP Terraform migrations have additional steps, configured by interactive
prompts.
`
if diff := cmp.Diff(wantStderr, gotStderr); diff != "" {
t.Errorf("wrong error output\n%s", diff)
}
})
}
2026-02-11 08:44:12 -05:00
func TestInit_cloudConfigColorTokensProcessed(t *testing.T) {
// This test verifies that when the error
// diagnostic detail contains color formatting tokens like [bold] and
// [reset], they are properly processed
// by the diagnostic formatter and do not appear as literal text in the
// output.
td := t.TempDir()
testCopyDir(t, testFixturePath("init-cloud-no-workspaces"), td)
t.Chdir(td)
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
},
}
args := []string{}
code := c.Run(args)
if code == 0 {
t.Fatalf("expected error, got success\n%s", done(t).Stdout())
}
gotStderr := done(t).Stderr()
expected := `
Error: failed to create backend alias to target "". The hostname is not in the correct format.
Error: Invalid workspaces configuration
on main.tf line 7, in terraform:
7: cloud {
Missing workspace mapping strategy. Either workspace "tags" or "name" is
required.
The 'workspaces' block configures how Terraform CLI maps its workspaces for
this single
configuration to workspaces within an HCP Terraform or Terraform Enterprise
organization. Two strategies are available:
tags - A set of tags used to select remote HCP Terraform or Terraform
Enterprise workspaces to be used for this single
configuration. New workspaces will automatically be tagged with these tag
values. Generally, this
is the primary and recommended strategy to use. This option conflicts with
"name".
name - The name of a single HCP Terraform or Terraform Enterprise workspace
to be used with this configuration.
When configured, only the specified workspace can be used. This option
conflicts with "tags"
and with the TF_WORKSPACE environment variable.
`
if diff := cmp.Diff(gotStderr, expected); diff != "" {
t.Errorf("unexpected output (-got +expected):\n%s", diff)
}
}
// make sure inputFalse stops execution on migrate
func TestInit_inputFalse(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-backend"), td)
t.Chdir(td)
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{"-input=false", "-backend-config=path=foo"}
2020-12-01 12:34:50 -05:00
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// write different states for foo and bar
fooState := states.BuildState(func(s *states.SyncState) {
s.SetOutputValue(
addrs.OutputValue{Name: "foo"}.Absolute(addrs.RootModuleInstance),
cty.StringVal("foo"),
false, // not sensitive
)
})
if err := statemgr.NewFilesystem("foo").WriteState(fooState); err != nil {
t.Fatal(err)
}
barState := states.BuildState(func(s *states.SyncState) {
s.SetOutputValue(
addrs.OutputValue{Name: "bar"}.Absolute(addrs.RootModuleInstance),
cty.StringVal("bar"),
false, // not sensitive
)
})
if err := statemgr.NewFilesystem("bar").WriteState(barState); err != nil {
t.Fatal(err)
}
ui = new(cli.MockUi)
c = &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args = []string{"-input=false", "-backend-config=path=bar", "-migrate-state"}
if code := c.Run(args); code == 0 {
t.Fatal("init should have failed", done(t).Stdout())
}
2024-03-26 04:02:09 -04:00
errMsg := done(t).All()
Cloud integration requires input for migrations We cannot programmatically migrate workspaces to Terraform Cloud without prompts, so `-input=false` should not be allowed in those cases. There are 4 scenarios where we need input from a user to complete migrating workspaces to Terraform Cloud. 1.) Migrate from a single local workspace to Terraform Cloud * Terraform config for a local backend. Implicit local (no backend specified) is fine. * `terraform init` and `terraform apply` * Change the Terraform config to use the cloud block * `terraform init -input=false` * You should now see an error message 2.) Migrate from a remote backend with a prefix to Terraform Cloud with tags * Create a workspace in Terraform Cloud manually. The name should include a prefix, like "app-one" * Have the terraform config use `backend "remote"` with a prefix set to "app-" * `terraform init` and `terraform apply` * Update the Terraform config to use a cloud block with `tags = ["app"]`. There should not be a prefix defined in the config now. * `terraform init -input=false` * You should now see an error message 3.) Migrate from multiple local workspaces to a single Terraform Cloud workspace * Create one or many local workspaces * `terraform init` and `terraform apply` in each * Change the Terraform config to use the cloud block * `terraform init -input=false` * You should now see an error message 4.) Migrate to Terraform Cloud and ask for a workspace name * Create several local workspaces * `terraform init` and `terraform apply` in each * Change the Terraform config to use the cloud block with tags * `terraform init -input=false` * You should now see an error message
2021-11-22 14:17:04 -05:00
if !strings.Contains(errMsg, "interactive input is disabled") {
t.Fatal("expected input disabled error, got", errMsg)
}
ui = new(cli.MockUi)
c = &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
// A missing input=false should abort rather than loop infinitely
args = []string{"-backend-config=path=baz"}
if code := c.Run(args); code == 0 {
t.Fatal("init should have failed", done(t).Stdout())
}
}
func TestInit_getProvider(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-providers"), td)
t.Chdir(td)
overrides := metaOverridesForProvider(testProvider())
ui := new(cli.MockUi)
view, done := testView(t)
providerSource, close := newMockProviderSource(t, map[string][]string{
// looking for an exact version
"exact": {"1.2.3"},
// config requires >= 2.3.3
"greater-than": {"2.3.4", "2.3.3", "2.3.0"},
// config specifies
"between": {"3.4.5", "2.3.4", "1.2.3"},
})
defer close()
m := Meta{
testingOverrides: overrides,
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{
"-backend=false", // should be possible to install plugins without backend init
}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// check that we got the providers for our config
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
exactPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/hashicorp/exact/1.2.3/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(exactPath); os.IsNotExist(err) {
t.Fatal("provider 'exact' not downloaded")
}
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
greaterThanPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/hashicorp/greater-than/2.3.4/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(greaterThanPath); os.IsNotExist(err) {
t.Fatal("provider 'greater-than' not downloaded")
}
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
betweenPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/hashicorp/between/2.3.4/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(betweenPath); os.IsNotExist(err) {
t.Fatal("provider 'between' not downloaded")
}
t.Run("future-state", func(t *testing.T) {
// getting providers should fail if a state from a newer version of
// terraform exists, since InitCommand.getProviders needs to inspect that
// state.
2020-10-05 08:33:49 -04:00
f, err := os.Create(DefaultStateFilename)
if err != nil {
t.Fatalf("err: %s", err)
}
defer f.Close()
// Construct a mock state file from the far future
type FutureState struct {
Version uint `json:"version"`
Lineage string `json:"lineage"`
TerraformVersion string `json:"terraform_version"`
Outputs map[string]interface{} `json:"outputs"`
Resources []map[string]interface{} `json:"resources"`
2020-10-05 08:33:49 -04:00
}
fs := &FutureState{
Version: 999,
Lineage: "123-456-789",
TerraformVersion: "999.0.0",
2020-12-01 12:34:50 -05:00
Outputs: make(map[string]interface{}),
Resources: make([]map[string]interface{}, 0),
}
src, err := json.MarshalIndent(fs, "", " ")
if err != nil {
t.Fatalf("failed to marshal future state: %s", err)
}
src = append(src, '\n')
_, err = f.Write(src)
2020-12-01 12:34:50 -05:00
if err != nil {
t.Fatal(err)
}
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m.Ui = ui
m.View = view
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code == 0 {
t.Fatal("expected error, got:", testOutput.Stdout())
}
errMsg := testOutput.Stderr()
if !strings.Contains(errMsg, "Unsupported state file format") {
t.Fatal("unexpected error:", errMsg)
}
})
}
func TestInit_getProviderSource(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-provider-source"), td)
t.Chdir(td)
overrides := metaOverridesForProvider(testProvider())
ui := new(cli.MockUi)
view, done := testView(t)
providerSource, close := newMockProviderSource(t, map[string][]string{
// looking for an exact version
"acme/alpha": {"1.2.3"},
// config doesn't specify versions for other providers
"registry.example.com/acme/beta": {"1.0.0"},
"gamma": {"2.0.0"},
})
defer close()
m := Meta{
testingOverrides: overrides,
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{
"-backend=false", // should be possible to install plugins without backend init
}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
// check that we got the providers for our config
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
exactPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/acme/alpha/1.2.3/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(exactPath); os.IsNotExist(err) {
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
t.Error("provider 'alpha' not downloaded")
}
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
greaterThanPath := fmt.Sprintf(".terraform/providers/registry.example.com/acme/beta/1.0.0/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(greaterThanPath); os.IsNotExist(err) {
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
t.Error("provider 'beta' not downloaded")
}
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
betweenPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/hashicorp/gamma/2.0.0/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(betweenPath); os.IsNotExist(err) {
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
t.Error("provider 'gamma' not downloaded")
}
}
func TestInit_getProviderLegacyFromState(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-provider-legacy-from-state"), td)
t.Chdir(td)
overrides := metaOverridesForProvider(testProvider())
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
providerSource, close := newMockProviderSource(t, map[string][]string{
"acme/alpha": {"1.2.3"},
})
defer close()
m := Meta{
testingOverrides: overrides,
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code != 1 {
t.Fatalf("got exit status %d; want 1\nstderr:\n%s\n\nstdout:\n%s", code, testOutput.Stderr(), testOutput.Stdout())
}
// Expect this diagnostic output
wants := []string{
"Invalid legacy provider address",
"You must complete the Terraform 0.13 upgrade process",
}
got := testOutput.All()
for _, want := range wants {
if !strings.Contains(got, want) {
t.Fatalf("expected output to contain %q, got:\n\n%s", want, got)
}
}
}
func TestInit_getProviderInvalidPackage(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-provider-invalid-package"), td)
t.Chdir(td)
overrides := metaOverridesForProvider(testProvider())
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
// create a provider source which allows installing an invalid package
addr := addrs.MustParseProviderSourceString("invalid/package")
version := getproviders.MustParseVersion("1.0.0")
meta, close, err := getproviders.FakeInstallablePackageMeta(
addr,
version,
getproviders.VersionList{getproviders.MustParseVersion("5.0")},
getproviders.CurrentPlatform,
"terraform-package", // should be "terraform-provider-package"
)
defer close()
if err != nil {
t.Fatalf("failed to prepare fake package for %s %s: %s", addr.ForDisplay(), version, err)
}
providerSource := getproviders.NewMockSource([]getproviders.PackageMeta{meta}, nil)
m := Meta{
testingOverrides: overrides,
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{
"-backend=false", // should be possible to install plugins without backend init
}
code := c.Run(args)
testOutput := done(t)
if code != 1 {
t.Fatalf("got exit status %d; want 1\nstderr:\n%s\n\nstdout:\n%s", code, testOutput.Stderr(), testOutput.Stdout())
}
// invalid provider should be installed
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
packagePath := fmt.Sprintf(".terraform/providers/registry.terraform.io/invalid/package/1.0.0/%s/terraform-package", getproviders.CurrentPlatform)
if _, err := os.Stat(packagePath); os.IsNotExist(err) {
t.Fatal("provider 'invalid/package' not downloaded")
}
wantErrors := []string{
"Failed to install provider",
"could not find executable file starting with terraform-provider-package",
}
got := testOutput.All()
for _, wantError := range wantErrors {
if !strings.Contains(got, wantError) {
t.Fatalf("missing error:\nwant: %q\ngot:\n%s", wantError, got)
}
}
}
func TestInit_getProviderDetectedLegacy(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-provider-detected-legacy"), td)
t.Chdir(td)
// We need to construct a multisource with a mock source and a registry
// source: the mock source will return ErrRegistryProviderNotKnown for an
// unknown provider, and the registry source will allow us to look up the
// appropriate namespace if possible.
providerSource, psClose := newMockProviderSource(t, map[string][]string{
"hashicorp/foo": {"1.2.3"},
"terraform-providers/baz": {"2.3.4"}, // this will not be installed
})
defer psClose()
registrySource, rsClose := testRegistrySource(t)
defer rsClose()
multiSource := getproviders.MultiSource{
{Source: providerSource},
{Source: registrySource},
}
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
Ui: ui,
View: view,
ProviderSource: multiSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{
"-backend=false", // should be possible to install plugins without backend init
}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("expected error, got output: \n%s", testOutput.Stdout())
}
// foo should be installed
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
fooPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/hashicorp/foo/1.2.3/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(fooPath); os.IsNotExist(err) {
t.Error("provider 'foo' not installed")
}
// baz should not be installed
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
bazPath := fmt.Sprintf(".terraform/providers/registry.terraform.io/terraform-providers/baz/2.3.4/%s", getproviders.CurrentPlatform)
if _, err := os.Stat(bazPath); !os.IsNotExist(err) {
t.Error("provider 'baz' installed, but should not be")
}
// error output is the main focus of this test
errOutput := testOutput.All()
errors := []string{
"Failed to query available provider packages",
"Could not retrieve the list of available versions",
"registry.terraform.io/hashicorp/baz",
"registry.terraform.io/hashicorp/frob",
}
for _, want := range errors {
if !strings.Contains(errOutput, want) {
t.Fatalf("expected error %q: %s", want, errOutput)
}
}
}
func TestInit_providerSource(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-required-providers"), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
"test": {"1.2.3", "1.2.4"},
"test-beta": {"1.2.4"},
"source": {"1.2.2", "1.2.3", "1.2.1"},
})
defer close()
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
if strings.Contains(testOutput.Stdout(), "Terraform has initialized, but configuration upgrades may be needed") {
t.Fatalf("unexpected \"configuration upgrade\" warning in output")
}
cacheDir := m.providerLocalCacheDir()
gotPackages := cacheDir.AllAvailablePackages()
wantPackages := map[addrs.Provider][]providercache.CachedProvider{
addrs.NewDefaultProvider("test"): {
{
Provider: addrs.NewDefaultProvider("test"),
Version: getproviders.MustParseVersion("1.2.3"),
PackageDir: expectedPackageInstallPath("test", "1.2.3", false),
},
},
addrs.NewDefaultProvider("test-beta"): {
{
Provider: addrs.NewDefaultProvider("test-beta"),
Version: getproviders.MustParseVersion("1.2.4"),
PackageDir: expectedPackageInstallPath("test-beta", "1.2.4", false),
},
},
addrs.NewDefaultProvider("source"): {
{
Provider: addrs.NewDefaultProvider("source"),
Version: getproviders.MustParseVersion("1.2.3"),
PackageDir: expectedPackageInstallPath("source", "1.2.3", false),
},
},
}
if diff := cmp.Diff(wantPackages, gotPackages); diff != "" {
t.Errorf("wrong cache directory contents after upgrade\n%s", diff)
}
locks, err := m.lockedDependencies()
if err != nil {
t.Fatalf("failed to get locked dependencies: %s", err)
}
gotProviderLocks := locks.AllProviders()
wantProviderLocks := map[addrs.Provider]*depsfile.ProviderLock{
addrs.NewDefaultProvider("test-beta"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("test-beta"),
getproviders.MustParseVersion("1.2.4"),
getproviders.MustParseVersionConstraints("= 1.2.4"),
[]getproviders.Hash{
getproviders.HashScheme1.New("see6W06w09Ea+AobFJ+mbvPTie6ASqZAAdlFZbs8BSM="),
},
),
addrs.NewDefaultProvider("test"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("test"),
getproviders.MustParseVersion("1.2.3"),
getproviders.MustParseVersionConstraints("= 1.2.3"),
[]getproviders.Hash{
getproviders.HashScheme1.New("wlbEC2mChQZ2hhgUhl6SeVLPP7fMqOFUZAQhQ9GIIno="),
},
),
addrs.NewDefaultProvider("source"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("source"),
getproviders.MustParseVersion("1.2.3"),
getproviders.MustParseVersionConstraints("= 1.2.3"),
[]getproviders.Hash{
getproviders.HashScheme1.New("myS3qb3px3tRBq1ZWRYJeUH+kySWpBc0Yy8rw6W7/p4="),
},
),
}
if diff := cmp.Diff(gotProviderLocks, wantProviderLocks, depsfile.ProviderLockComparer); diff != "" {
t.Errorf("wrong version selections after upgrade\n%s", diff)
}
if got, want := testOutput.Stdout(), "Installed hashicorp/test v1.2.3 (verified checksum)"; !strings.Contains(got, want) {
t.Fatalf("unexpected output: %s\nexpected to include %q", got, want)
}
if got, want := testOutput.All(), "\n - hashicorp/source\n - hashicorp/test\n - hashicorp/test-beta"; !strings.Contains(got, want) {
t.Fatalf("wrong error message\nshould contain: %s\ngot:\n%s", want, got)
2020-04-08 16:22:07 -04:00
}
}
func TestInit_cancelModules(t *testing.T) {
// This test runs `terraform init` as if SIGINT (or similar on other
// platforms) were sent to it, testing that it is interruptible.
td := t.TempDir()
testCopyDir(t, testFixturePath("init-registry-module"), td)
t.Chdir(td)
// Our shutdown channel is pre-closed so init will exit as soon as it
// starts a cancelable portion of the process.
shutdownCh := make(chan struct{})
close(shutdownCh)
ui := cli.NewMockUi()
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ShutdownCh: shutdownCh,
}
c := &InitCommand{
Meta: m,
}
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded; wanted error\n%s", testOutput.Stdout())
}
if got, want := testOutput.Stderr(), `Module installation was canceled by an interrupt signal`; !strings.Contains(got, want) {
t.Fatalf("wrong error message\nshould contain: %s\ngot:\n%s", want, got)
}
}
func TestInit_cancelProviders(t *testing.T) {
// This test runs `terraform init` as if SIGINT (or similar on other
// platforms) were sent to it, testing that it is interruptible.
td := t.TempDir()
testCopyDir(t, testFixturePath("init-required-providers"), td)
t.Chdir(td)
// Use a provider source implementation which is designed to hang indefinitely,
// to avoid a race between the closed shutdown channel and the provider source
// operations.
providerSource := &getproviders.HangingSource{}
// Our shutdown channel is pre-closed so init will exit as soon as it
// starts a cancelable portion of the process.
shutdownCh := make(chan struct{})
close(shutdownCh)
ui := cli.NewMockUi()
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
ShutdownCh: shutdownCh,
}
c := &InitCommand{
Meta: m,
}
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded; wanted error\n%s", testOutput.All())
}
// Currently the first operation that is cancelable is provider
// installation, so our error message comes from there. If we
// make the earlier steps cancelable in future then it'd be
// expected for this particular message to change.
if got, want := testOutput.Stderr(), `Provider installation was canceled by an interrupt signal`; !strings.Contains(got, want) {
t.Fatalf("wrong error message\nshould contain: %s\ngot:\n%s", want, got)
}
}
func TestInit_getUpgradePlugins(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-providers"), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
// looking for an exact version
"exact": {"1.2.3"},
// config requires >= 2.3.3
"greater-than": {"2.3.4", "2.3.3", "2.3.0"},
// config specifies > 1.0.0 , < 3.0.0
"between": {"3.4.5", "2.3.4", "1.2.3"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
installFakeProviderPackages(t, &m, map[string][]string{
"exact": {"0.0.1"},
"greater-than": {"2.3.3"},
})
c := &InitCommand{
Meta: m,
}
args := []string{
"-upgrade=true",
}
if code := c.Run(args); code != 0 {
t.Fatalf("command did not complete successfully:\n%s", done(t).All())
}
cacheDir := m.providerLocalCacheDir()
gotPackages := cacheDir.AllAvailablePackages()
wantPackages := map[addrs.Provider][]providercache.CachedProvider{
// "between" wasn't previously installed at all, so we installed
// the newest available version that matched the version constraints.
addrs.NewDefaultProvider("between"): {
{
Provider: addrs.NewDefaultProvider("between"),
Version: getproviders.MustParseVersion("2.3.4"),
PackageDir: expectedPackageInstallPath("between", "2.3.4", false),
},
},
// The existing version of "exact" did not match the version constraints,
// so we installed what the configuration selected as well.
addrs.NewDefaultProvider("exact"): {
{
Provider: addrs.NewDefaultProvider("exact"),
Version: getproviders.MustParseVersion("1.2.3"),
PackageDir: expectedPackageInstallPath("exact", "1.2.3", false),
},
// Previous version is still there, but not selected
{
Provider: addrs.NewDefaultProvider("exact"),
Version: getproviders.MustParseVersion("0.0.1"),
PackageDir: expectedPackageInstallPath("exact", "0.0.1", false),
},
},
// The existing version of "greater-than" _did_ match the constraints,
// but a newer version was available and the user specified
// -upgrade and so we upgraded it anyway.
addrs.NewDefaultProvider("greater-than"): {
{
Provider: addrs.NewDefaultProvider("greater-than"),
Version: getproviders.MustParseVersion("2.3.4"),
PackageDir: expectedPackageInstallPath("greater-than", "2.3.4", false),
},
// Previous version is still there, but not selected
{
Provider: addrs.NewDefaultProvider("greater-than"),
Version: getproviders.MustParseVersion("2.3.3"),
PackageDir: expectedPackageInstallPath("greater-than", "2.3.3", false),
},
},
}
if diff := cmp.Diff(wantPackages, gotPackages); diff != "" {
t.Errorf("wrong cache directory contents after upgrade\n%s", diff)
}
locks, err := m.lockedDependencies()
if err != nil {
t.Fatalf("failed to get locked dependencies: %s", err)
}
gotProviderLocks := locks.AllProviders()
wantProviderLocks := map[addrs.Provider]*depsfile.ProviderLock{
addrs.NewDefaultProvider("between"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("between"),
getproviders.MustParseVersion("2.3.4"),
getproviders.MustParseVersionConstraints("> 1.0.0, < 3.0.0"),
[]getproviders.Hash{
getproviders.HashScheme1.New("JVqAvZz88A+hS2wHVtTWQkHaxoA/LrUAz0H3jPBWPIA="),
},
),
addrs.NewDefaultProvider("exact"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("exact"),
getproviders.MustParseVersion("1.2.3"),
getproviders.MustParseVersionConstraints("= 1.2.3"),
[]getproviders.Hash{
getproviders.HashScheme1.New("H1TxWF8LyhBb6B4iUdKhLc/S9sC/jdcrCykpkbGcfbg="),
},
),
addrs.NewDefaultProvider("greater-than"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("greater-than"),
getproviders.MustParseVersion("2.3.4"),
getproviders.MustParseVersionConstraints(">= 2.3.3"),
[]getproviders.Hash{
getproviders.HashScheme1.New("SJPpXx/yoFE/W+7eCipjJ+G21xbdnTBD7lWodZ8hWkU="),
},
),
}
if diff := cmp.Diff(gotProviderLocks, wantProviderLocks, depsfile.ProviderLockComparer); diff != "" {
t.Errorf("wrong version selections after upgrade\n%s", diff)
}
}
func TestInit_getProviderMissing(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-providers"), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
// looking for exact version 1.2.3
"exact": {"1.2.4"},
// config requires >= 2.3.3
"greater-than": {"2.3.4", "2.3.3", "2.3.0"},
// config specifies
"between": {"3.4.5", "2.3.4", "1.2.3"},
})
defer close()
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("expected error, got output: \n%s", testOutput.Stdout())
}
if !strings.Contains(testOutput.All(), "no available releases match") {
t.Fatalf("unexpected error output: %s", testOutput.Stderr())
}
}
func TestInit_checkRequiredVersion(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-check-required-version"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{}
if code := c.Run(args); code != 1 {
t.Fatalf("got exit status %d; want 1\nstderr:\n%s\n\nstdout:\n%s", code, done(t).Stderr(), done(t).Stdout())
}
2024-03-26 04:02:09 -04:00
errStr := done(t).All()
if !strings.Contains(errStr, `required_version = "~> 0.9.0"`) {
t.Fatalf("output should point to unmet version constraint, but is:\n\n%s", errStr)
}
if strings.Contains(errStr, `required_version = ">= 0.13.0"`) {
t.Fatalf("output should not point to met version constraint, but is:\n\n%s", errStr)
}
}
// Verify that init will error out with an invalid version constraint, even if
// there are other invalid configuration constructs.
func TestInit_checkRequiredVersionFirst(t *testing.T) {
t.Run("root_module", func(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-check-required-version-first"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{}
if code := c.Run(args); code != 1 {
t.Fatalf("got exit status %d; want 1\nstderr:\n%s\n\nstdout:\n%s", code, done(t).Stderr(), done(t).Stdout())
}
2024-03-26 04:02:09 -04:00
errStr := done(t).All()
if !strings.Contains(errStr, `Unsupported Terraform Core version`) {
t.Fatalf("output should point to unmet version constraint, but is:\n\n%s", errStr)
}
})
t.Run("sub_module", func(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-check-required-version-first-module"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
},
}
args := []string{}
if code := c.Run(args); code != 1 {
t.Fatalf("got exit status %d; want 1\nstderr:\n%s\n\nstdout:\n%s", code, done(t).Stderr(), done(t).Stdout())
}
2024-03-26 04:02:09 -04:00
errStr := done(t).All()
if !strings.Contains(errStr, `Unsupported Terraform Core version`) {
t.Fatalf("output should point to unmet version constraint, but is:\n\n%s", errStr)
}
})
}
func TestInit_providerLockFile(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-provider-lock-file"), td)
// The temporary directory does not have write permission (dr-xr-xr-x) after the copy
defer os.Chmod(td, os.ModePerm)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
"test": {"1.2.3"},
})
defer close()
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
lockFile := ".terraform.lock.hcl"
buf, err := os.ReadFile(lockFile)
if err != nil {
t.Fatalf("failed to read dependency lock file %s: %s", lockFile, err)
}
buf = bytes.TrimSpace(buf)
// The hash in here is for the fake package that newMockProviderSource produces
// (so it'll change if newMockProviderSource starts producing different contents)
wantLockFile := strings.TrimSpace(`
# This file is maintained automatically by "terraform init".
# Manual edits may be lost in future updates.
provider "registry.terraform.io/hashicorp/test" {
version = "1.2.3"
constraints = "1.2.3"
hashes = [
"h1:wlbEC2mChQZ2hhgUhl6SeVLPP7fMqOFUZAQhQ9GIIno=",
]
}
`)
if diff := cmp.Diff(wantLockFile, string(buf)); diff != "" {
t.Errorf("wrong dependency lock file contents\n%s", diff)
}
// Make the local directory read-only, and verify that rerunning init
// succeeds, to ensure that we don't try to rewrite an unchanged lock file
os.Chmod(".", 0555)
if code := c.Run(args); code != 0 {
2024-03-26 04:02:09 -04:00
t.Fatalf("bad: \n%s", done(t).All())
}
}
func TestInit_providerLockFileReadonly(t *testing.T) {
// The hash in here is for the fake package that newMockProviderSource produces
// (so it'll change if newMockProviderSource starts producing different contents)
inputLockFile := strings.TrimSpace(`
# This file is maintained automatically by "terraform init".
# Manual edits may be lost in future updates.
provider "registry.terraform.io/hashicorp/test" {
version = "1.2.3"
constraints = "1.2.3"
hashes = [
"zh:e919b507a91e23a00da5c2c4d0b64bcc7900b68d43b3951ac0f6e5d80387fbdc",
]
}
`)
badLockFile := strings.TrimSpace(`
# This file is maintained automatically by "terraform init".
# Manual edits may be lost in future updates.
provider "registry.terraform.io/hashicorp/test" {
version = "1.2.3"
constraints = "1.2.3"
hashes = [
"zh:0000000000000000000000000000000000000000000000000000000000000000",
]
}
`)
updatedLockFile := strings.TrimSpace(`
# This file is maintained automatically by "terraform init".
# Manual edits may be lost in future updates.
provider "registry.terraform.io/hashicorp/test" {
version = "1.2.3"
constraints = "1.2.3"
hashes = [
"h1:wlbEC2mChQZ2hhgUhl6SeVLPP7fMqOFUZAQhQ9GIIno=",
"zh:e919b507a91e23a00da5c2c4d0b64bcc7900b68d43b3951ac0f6e5d80387fbdc",
]
}
`)
emptyUpdatedLockFile := strings.TrimSpace(`
# This file is maintained automatically by "terraform init".
# Manual edits may be lost in future updates.
`)
cases := []struct {
desc string
fixture string
providers map[string][]string
input string
args []string
ok bool
want string
}{
{
desc: "default",
fixture: "init-provider-lock-file",
providers: map[string][]string{"test": {"1.2.3"}},
input: inputLockFile,
args: []string{},
ok: true,
want: updatedLockFile,
},
{
desc: "unused provider",
fixture: "init-provider-now-unused",
providers: map[string][]string{"test": {"1.2.3"}},
input: inputLockFile,
args: []string{},
ok: true,
want: emptyUpdatedLockFile,
},
{
desc: "readonly",
fixture: "init-provider-lock-file",
providers: map[string][]string{"test": {"1.2.3"}},
input: inputLockFile,
args: []string{"-lockfile=readonly"},
ok: true,
want: inputLockFile,
},
{
desc: "unused provider readonly",
fixture: "init-provider-now-unused",
providers: map[string][]string{"test": {"1.2.3"}},
input: inputLockFile,
args: []string{"-lockfile=readonly"},
ok: false,
want: inputLockFile,
},
{
desc: "conflict",
fixture: "init-provider-lock-file",
providers: map[string][]string{"test": {"1.2.3"}},
input: inputLockFile,
args: []string{"-lockfile=readonly", "-upgrade"},
ok: false,
want: inputLockFile,
},
{
desc: "checksum mismatch",
fixture: "init-provider-lock-file",
providers: map[string][]string{"test": {"1.2.3"}},
input: badLockFile,
args: []string{"-lockfile=readonly"},
ok: false,
want: badLockFile,
},
{
desc: "reject to change required provider dependences",
fixture: "init-provider-lock-file-readonly-add",
providers: map[string][]string{
"test": {"1.2.3"},
"foo": {"1.0.0"},
},
input: inputLockFile,
args: []string{"-lockfile=readonly"},
ok: false,
want: inputLockFile,
},
}
for _, tc := range cases {
t.Run(tc.desc, func(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath(tc.fixture), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, tc.providers)
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
2024-03-26 04:02:09 -04:00
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
// write input lockfile
lockFile := ".terraform.lock.hcl"
if err := os.WriteFile(lockFile, []byte(tc.input), 0644); err != nil {
t.Fatalf("failed to write input lockfile: %s", err)
}
code := c.Run(tc.args)
if tc.ok && code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
if !tc.ok && code == 0 {
t.Fatalf("expected error, got output: \n%s", done(t).Stdout())
}
buf, err := os.ReadFile(lockFile)
if err != nil {
t.Fatalf("failed to read dependency lock file %s: %s", lockFile, err)
}
buf = bytes.TrimSpace(buf)
if diff := cmp.Diff(tc.want, string(buf)); diff != "" {
t.Errorf("wrong dependency lock file contents\n%s", diff)
}
})
}
}
func TestInit_pluginDirReset(t *testing.T) {
td := testTempDir(t)
defer os.RemoveAll(td)
t.Chdir(td)
// An empty provider source
providerSource, close := newMockProviderSource(t, nil)
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
// make our vendor paths
pluginPath := []string{"a", "b", "c"}
for _, p := range pluginPath {
if err := os.MkdirAll(p, 0755); err != nil {
t.Fatal(err)
}
}
// run once and save the -plugin-dir
args := []string{"-plugin-dir", "a"}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
pluginDirs, err := c.loadPluginPath()
if err != nil {
t.Fatal(err)
}
if len(pluginDirs) != 1 || pluginDirs[0] != "a" {
t.Fatalf(`expected plugin dir ["a"], got %q`, pluginDirs)
}
ui = new(cli.MockUi)
c = &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource, // still empty
},
}
// make sure we remove the plugin-dir record
args = []string{"-plugin-dir="}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
pluginDirs, err = c.loadPluginPath()
if err != nil {
t.Fatal(err)
}
if len(pluginDirs) != 0 {
t.Fatalf("expected no plugin dirs got %q", pluginDirs)
}
}
// Test user-supplied -plugin-dir
func TestInit_pluginDirProviders(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-providers"), td)
t.Chdir(td)
// An empty provider source
providerSource, close := newMockProviderSource(t, nil)
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
// make our vendor paths
pluginPath := []string{"a", "b", "c"}
for _, p := range pluginPath {
if err := os.MkdirAll(p, 0755); err != nil {
t.Fatal(err)
}
}
// We'll put some providers in our plugin dirs. To do this, we'll pretend
// for a moment that they are provider cache directories just because that
// allows us to lean on our existing test helper functions to do this.
for i, def := range [][]string{
{"exact", "1.2.3"},
{"greater-than", "2.3.4"},
{"between", "2.3.4"},
} {
name, version := def[0], def[1]
dir := providercache.NewDir(pluginPath[i])
installFakeProviderPackagesElsewhere(t, dir, map[string][]string{
name: {version},
})
}
args := []string{
"-plugin-dir", "a",
"-plugin-dir", "b",
"-plugin-dir", "c",
}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
locks, err := m.lockedDependencies()
if err != nil {
t.Fatalf("failed to get locked dependencies: %s", err)
}
gotProviderLocks := locks.AllProviders()
wantProviderLocks := map[addrs.Provider]*depsfile.ProviderLock{
addrs.NewDefaultProvider("between"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("between"),
getproviders.MustParseVersion("2.3.4"),
getproviders.MustParseVersionConstraints("> 1.0.0, < 3.0.0"),
[]getproviders.Hash{
getproviders.HashScheme1.New("JVqAvZz88A+hS2wHVtTWQkHaxoA/LrUAz0H3jPBWPIA="),
},
),
addrs.NewDefaultProvider("exact"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("exact"),
getproviders.MustParseVersion("1.2.3"),
getproviders.MustParseVersionConstraints("= 1.2.3"),
[]getproviders.Hash{
getproviders.HashScheme1.New("H1TxWF8LyhBb6B4iUdKhLc/S9sC/jdcrCykpkbGcfbg="),
},
),
addrs.NewDefaultProvider("greater-than"): depsfile.NewProviderLock(
addrs.NewDefaultProvider("greater-than"),
getproviders.MustParseVersion("2.3.4"),
getproviders.MustParseVersionConstraints(">= 2.3.3"),
[]getproviders.Hash{
getproviders.HashScheme1.New("SJPpXx/yoFE/W+7eCipjJ+G21xbdnTBD7lWodZ8hWkU="),
},
),
}
if diff := cmp.Diff(gotProviderLocks, wantProviderLocks, depsfile.ProviderLockComparer); diff != "" {
t.Errorf("wrong version selections after upgrade\n%s", diff)
}
// -plugin-dir overrides the normal provider source, so it should not have
// seen any calls at all.
if calls := providerSource.CallLog(); len(calls) > 0 {
t.Errorf("unexpected provider source calls (want none)\n%s", spew.Sdump(calls))
}
}
// Test user-supplied -plugin-dir doesn't allow auto-install
func TestInit_pluginDirProvidersDoesNotGet(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-get-providers"), td)
t.Chdir(td)
// Our provider source has a suitable package for "between" available,
// but we should ignore it because -plugin-dir is set and thus this
// source is temporarily overridden during install.
providerSource, close := newMockProviderSource(t, map[string][]string{
"between": {"2.3.4"},
})
defer close()
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
// make our vendor paths
pluginPath := []string{"a", "b"}
for _, p := range pluginPath {
if err := os.MkdirAll(p, 0755); err != nil {
t.Fatal(err)
}
}
// We'll put some providers in our plugin dirs. To do this, we'll pretend
// for a moment that they are provider cache directories just because that
// allows us to lean on our existing test helper functions to do this.
for i, def := range [][]string{
{"exact", "1.2.3"},
{"greater-than", "2.3.4"},
} {
name, version := def[0], def[1]
dir := providercache.NewDir(pluginPath[i])
installFakeProviderPackagesElsewhere(t, dir, map[string][]string{
name: {version},
})
}
args := []string{
"-plugin-dir", "a",
"-plugin-dir", "b",
}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
// should have been an error
t.Fatalf("succeeded; want error\nstdout:\n%s\nstderr\n%s", testOutput.Stdout(), testOutput.Stderr())
}
// The error output should mention the "between" provider but should not
// mention either the "exact" or "greater-than" provider, because the
// latter two are available via the -plugin-dir directories.
errStr := testOutput.Stderr()
if subStr := "hashicorp/between"; !strings.Contains(errStr, subStr) {
t.Errorf("error output should mention the 'between' provider\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "hashicorp/exact"; strings.Contains(errStr, subStr) {
t.Errorf("error output should not mention the 'exact' provider\ndo not want substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "hashicorp/greater-than"; strings.Contains(errStr, subStr) {
t.Errorf("error output should not mention the 'greater-than' provider\ndo not want substr: %s\ngot:\n%s", subStr, errStr)
}
if calls := providerSource.CallLog(); len(calls) > 0 {
t.Errorf("unexpected provider source calls (want none)\n%s", spew.Sdump(calls))
}
}
// Verify that plugin-dir doesn't prevent discovery of internal providers
func TestInit_pluginDirWithBuiltIn(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-internal"), td)
t.Chdir(td)
// An empty provider source
providerSource, close := newMockProviderSource(t, nil)
defer close()
ui := cli.NewMockUi()
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
args := []string{"-plugin-dir", "./"}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("error: %s", testOutput.Stderr())
}
outputStr := testOutput.Stdout()
if subStr := "terraform.io/builtin/terraform is built in to Terraform"; !strings.Contains(outputStr, subStr) {
t.Errorf("output should mention the terraform provider\nwant substr: %s\ngot:\n%s", subStr, outputStr)
}
}
func TestInit_invalidBuiltInProviders(t *testing.T) {
// This test fixture includes two invalid provider dependencies:
// - an implied dependency on terraform.io/builtin/terraform with an
// explicit version number, which is not allowed because it's builtin.
// - an explicit dependency on terraform.io/builtin/nonexist, which does
// not exist at all.
td := t.TempDir()
testCopyDir(t, testFixturePath("init-internal-invalid"), td)
t.Chdir(td)
// An empty provider source
providerSource, close := newMockProviderSource(t, nil)
defer close()
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
testingOverrides: metaOverridesForProvider(testProvider()),
Ui: ui,
View: view,
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded, but was expecting error\nstdout:\n%s\nstderr:\n%s", testOutput.Stdout(), testOutput.Stderr())
}
errStr := testOutput.Stderr()
if subStr := "Cannot use terraform.io/builtin/terraform: built-in"; !strings.Contains(errStr, subStr) {
t.Errorf("error output should mention the terraform provider\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Cannot use terraform.io/builtin/nonexist: this Terraform release"; !strings.Contains(errStr, subStr) {
t.Errorf("error output should mention the 'nonexist' provider\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
}
command: "terraform init" can partially initialize for 0.12upgrade There are a few constructs from 0.11 and prior that cause 0.12 parsing to fail altogether, which previously created a chicken/egg problem because we need to install the providers in order to run "terraform 0.12upgrade" and thus fix the problem. This changes "terraform init" to use the new "early configuration" loader for module and provider installation. This is built on the more permissive parser in the terraform-config-inspect package, and so it allows us to read out the top-level blocks from the configuration while accepting legacy HCL syntax. In the long run this will let us do version compatibility detection before attempting a "real" config load, giving us better error messages for any future syntax additions, but in the short term the key thing is that it allows us to install the dependencies even if the configuration isn't fully valid. Because backend init still requires full configuration, this introduces a new mode of terraform init where it detects heuristically if it seems like we need to do a configuration upgrade and does a partial init if so, before finally directing the user to run "terraform 0.12upgrade" before running any other commands. The heuristic here is based on two assumptions: - If the "early" loader finds no errors but the normal loader does, the configuration is likely to be valid for Terraform 0.11 but not 0.12. - If there's already a version constraint in the configuration that excludes Terraform versions prior to v0.12 then the configuration is probably _already_ upgraded and so it's just a normal syntax error, even if the early loader didn't detect it. Once the upgrade process is removed in 0.13.0 (users will be required to go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can be simplified to remove that special mode, but the idea of doing the dependency version checks against the liberal parser will remain valuable to increase our chances of reporting version-based incompatibilities rather than syntax errors as we add new features in future.
2019-01-14 14:11:00 -05:00
func TestInit_invalidSyntaxNoBackend(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-syntax-invalid-no-backend"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
Ui: ui,
View: view,
}
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded, but was expecting error\nstdout:\n%s\nstderr:\n%s", testOutput.Stdout(), testOutput.Stderr())
}
errStr := testOutput.Stderr()
if subStr := "Terraform encountered problems during initialisation, including problems\nwith the configuration, described below."; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should include preamble\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Error: Unsupported block type"; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should mention the syntax problem\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
}
func TestInit_invalidSyntaxWithBackend(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-syntax-invalid-with-backend"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
Ui: ui,
View: view,
}
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded, but was expecting error\nstdout:\n%s\nstderr:\n%s", testOutput.Stdout(), testOutput.Stderr())
}
errStr := testOutput.Stderr()
if subStr := "Terraform encountered problems during initialisation, including problems\nwith the configuration, described below."; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should include preamble\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Error: Unsupported block type"; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should mention the syntax problem\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
}
func TestInit_invalidSyntaxInvalidBackend(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-syntax-invalid-backend-invalid"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
Ui: ui,
View: view,
}
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded, but was expecting error\nstdout:\n%s\nstderr:\n%s", testOutput.Stdout(), testOutput.Stderr())
}
errStr := testOutput.Stderr()
if subStr := "Terraform encountered problems during initialisation, including problems\nwith the configuration, described below."; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should include preamble\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Error: Unsupported block type"; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should mention syntax errors\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Error: Unsupported backend type"; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should mention the invalid backend\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
}
func TestInit_invalidSyntaxBackendAttribute(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-syntax-invalid-backend-attribute-invalid"), td)
t.Chdir(td)
ui := cli.NewMockUi()
2024-03-26 04:02:09 -04:00
view, done := testView(t)
m := Meta{
Ui: ui,
View: view,
}
c := &InitCommand{
Meta: m,
}
code := c.Run(nil)
testOutput := done(t)
if code == 0 {
t.Fatalf("succeeded, but was expecting error\nstdout:\n%s\nstderr:\n%s", testOutput.Stdout(), testOutput.Stderr())
}
errStr := testOutput.All()
if subStr := "Terraform encountered problems during initialisation, including problems\nwith the configuration, described below."; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should include preamble\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Error: Invalid character"; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should mention the invalid character\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
if subStr := "Error: Invalid expression"; !strings.Contains(errStr, subStr) {
t.Errorf("Error output should mention the invalid expression\nwant substr: %s\ngot:\n%s", subStr, errStr)
}
}
func TestInit_testsWithExternalProviders(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-tests-external-providers"), td)
t.Chdir(td)
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/testing": {"1.0.0"},
"testing/configure": {"1.0.0"},
})
defer close()
hashicorpTestingProviderAddress := addrs.NewDefaultProvider("testing")
hashicorpTestingProvider := new(testing_provider.MockProvider)
testingConfigureProviderAddress := addrs.NewProvider(addrs.DefaultProviderRegistryHost, "testing", "configure")
testingConfigureProvider := new(testing_provider.MockProvider)
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
hashicorpTestingProviderAddress: providers.FactoryFixed(hashicorpTestingProvider),
testingConfigureProviderAddress: providers.FactoryFixed(testingConfigureProvider),
},
},
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
var args []string
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).All())
}
}
func TestInit_tests(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-tests"), td)
t.Chdir(td)
provider := applyFixtureProvider() // We just want the types from this provider.
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(provider),
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
args := []string{}
if code := c.Run(args); code != 0 {
t.Fatalf("bad: \n%s", done(t).Stderr())
}
}
func TestInit_testsWithProvider(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-tests-with-provider"), td)
t.Chdir(td)
provider := applyFixtureProvider() // We just want the types from this provider.
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
2024-03-26 04:02:09 -04:00
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(provider),
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("expected failure but got: \n%s", testOutput.All())
}
got := testOutput.Stderr()
want := `
Error: Failed to query available provider packages
Could not retrieve the list of available versions for provider
hashicorp/test: no available releases match the given constraints 1.0.1,
1.0.2
To see which modules are currently depending on hashicorp/test and what
versions are specified, run the following command:
terraform providers
`
if diff := cmp.Diff(got, want); len(diff) > 0 {
t.Fatalf("wrong error message: \ngot:\n%s\nwant:\n%s\ndiff:\n%s", got, want, diff)
}
}
func TestInit_testsWithOverriddenInvalidRequiredProviders(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-overrides-and-duplicates"), td)
t.Chdir(td)
provider := applyFixtureProvider() // We just want the types from this provider.
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(provider),
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
args := []string{}
code := c.Run(args) // just make sure it doesn't crash.
if code != 1 {
t.Fatalf("expected failure but got: \n%s", done(t).All())
}
}
func TestInit_testsWithInvalidRequiredProviders(t *testing.T) {
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-duplicates"), td)
t.Chdir(td)
provider := applyFixtureProvider() // We just want the types from this provider.
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(provider),
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
args := []string{}
code := c.Run(args) // just make sure it doesn't crash.
if code != 1 {
t.Fatalf("expected failure but got: \n%s", done(t).All())
}
}
func TestInit_testsWithModule(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-tests-with-module"), td)
t.Chdir(td)
provider := applyFixtureProvider() // We just want the types from this provider.
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: metaOverridesForProvider(provider),
Ui: ui,
View: view,
ProviderSource: providerSource,
},
}
args := []string{}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
// Check output
output := testOutput.Stdout()
if !strings.Contains(output, "test.main.setup in setup") {
t.Fatalf("doesn't look like we installed the test module': %s", output)
}
}
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
// Testing init's behaviors with `state_store` when run in an empty working directory
func TestInit_stateStore_newWorkingDir(t *testing.T) {
t.Run("temporary: test showing use of HTTP server in mock provider source", func(t *testing.T) {
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
// Mock provider still needs to be supplied via testingOverrides despite the mock HTTP source
mockProvider := mockPluggableStateStorageProvider()
mockProviderVersion := getproviders.MustParseVersion("1.2.3")
mockProviderAddress := addrs.NewDefaultProvider("test")
// Set up mock provider source that mocks out downloading hashicorp/test v1.2.3 via HTTP.
// This stops Terraform auto-approving the provider installation.
source := newMockProviderSourceUsingTestHttpServer(t, mockProviderAddress, mockProviderVersion)
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: source,
}
c := &InitCommand{
Meta: meta,
}
args := []string{"-enable-pluggable-state-storage-experiment=true"}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutputs := []string{
"Initializing the state store...",
"Terraform created an empty state file for the default workspace",
"Terraform has been successfully initialized!",
}
for _, expected := range expectedOutputs {
if !strings.Contains(output, expected) {
t.Fatalf("expected output to include %q, but got':\n %s", expected, output)
}
}
})
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
t.Run("the init command creates a backend state file, and creates the default workspace by default", func(t *testing.T) {
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// The test fixture config has no version constraints, so the latest version will
// be used; below is the 'latest' version in the test world.
"hashicorp/test": {"1.2.3"},
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{"-enable-pluggable-state-storage-experiment=true"}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutputs := []string{
"Initializing the state store...",
"Terraform created an empty state file for the default workspace",
"Terraform has been successfully initialized!",
}
for _, expected := range expectedOutputs {
if !strings.Contains(output, expected) {
t.Fatalf("expected output to include %q, but got':\n %s", expected, output)
}
}
// Assert the default workspace was created
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; !exists {
t.Fatal("expected the default workspace to be created during init, but it is missing")
}
// Assert contents of the backend state file
statePath := filepath.Join(meta.DataDir(), DefaultStateFilename)
sMgr := &clistate.LocalState{Path: statePath}
if err := sMgr.RefreshState(); err != nil {
t.Fatal("Failed to load state:", err)
}
s := sMgr.State()
if s == nil {
t.Fatal("expected backend state file to be created, but there isn't one")
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
v1_2_3, _ := version.NewVersion("1.2.3")
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
expectedState := &workdir.StateStoreConfigState{
Type: "test_store",
ConfigRaw: []byte("{\n \"value\": \"foobar\"\n }"),
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
Hash: uint64(4158988729),
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
Provider: &workdir.ProviderConfigState{
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
Version: v1_2_3,
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
Source: &tfaddr.Provider{
Hostname: tfaddr.DefaultProviderRegistryHost,
Namespace: "hashicorp",
Type: "test",
},
ConfigRaw: []byte("{\n \"region\": null\n }"),
},
}
if diff := cmp.Diff(s.StateStore, expectedState); diff != "" {
t.Fatalf("unexpected diff in backend state file's description of state store:\n%s", diff)
}
})
t.Run("an init command with the flag -create-default-workspace=false will not make the default workspace by default", func(t *testing.T) {
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
},
}
args := []string{"-enable-pluggable-state-storage-experiment=true", "-create-default-workspace=false"}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutput := `Terraform has been configured to skip creation of the default workspace`
if !strings.Contains(output, expectedOutput) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedOutput, output)
}
// Assert the default workspace was created
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; exists {
t.Fatal("expected Terraform to skip creating the default workspace, but it has been created")
}
})
t.Run("an init command with TF_SKIP_CREATE_DEFAULT_WORKSPACE set will not make the default workspace by default", func(t *testing.T) {
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
},
}
t.Setenv("TF_SKIP_CREATE_DEFAULT_WORKSPACE", "1") // any value
args := []string{"-enable-pluggable-state-storage-experiment=true"}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutput := `Terraform has been configured to skip creation of the default workspace`
if !strings.Contains(output, expectedOutput) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedOutput, output)
}
// Assert the default workspace was created
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; exists {
t.Fatal("expected Terraform to skip creating the default workspace, but it has been created")
}
})
// This scenario would be rare, but protecting against it is easy and avoids assumptions.
t.Run("if a custom workspace is selected but no workspaces exist an error is returned", func(t *testing.T) {
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
// Select a custom workspace (which will not exist)
customWorkspace := "my-custom-workspace"
t.Setenv(WorkspaceNameEnvVar, customWorkspace)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{"-enable-pluggable-state-storage-experiment=true"}
code := c.Run(args)
testOutput := done(t)
if code != 1 {
t.Fatalf("expected code 1 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutputs := []string{
fmt.Sprintf("Workspace %q has not been created yet", customWorkspace),
fmt.Sprintf("To create the custom workspace %q use the command `terraform workspace new %s`", customWorkspace, customWorkspace),
}
for _, expected := range expectedOutputs {
if !strings.Contains(cleanString(output), expected) {
t.Fatalf("expected output to include %q, but got':\n %s", expected, cleanString(output))
}
}
// Assert no workspaces exist
if len(mockProvider.MockStates) != 0 {
t.Fatalf("expected no workspaces, but got: %#v", mockProvider.MockStates)
}
// Assert no backend state file made due to the error
statePath := filepath.Join(meta.DataDir(), DefaultStateFilename)
_, err := os.Stat(statePath)
if pathErr, ok := err.(*os.PathError); !ok || !os.IsNotExist(pathErr.Err) {
t.Fatalf("expected backend state file to not be created, but it exists")
}
})
// Test what happens when the selected workspace doesn't exist, but there are other workspaces available.
//
// When input is disabled (in automation, etc) Terraform cannot prompts the user to select an alternative.
// Instead, an error is returned.
t.Run("init: returns an error when input is disabled and the selected workspace doesn't exist and other custom workspaces do exist.", func(t *testing.T) {
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProvider.GetStatesResponse = &providers.GetStatesResponse{
States: []string{
"foobar1",
"foobar2",
// Force provider to report workspaces exist
// But default workspace doesn't exist
},
}
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
// If input is disabled users receive an error about the missing workspace
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-input=false",
}
code := c.Run(args)
testOutput := done(t)
if code != 1 {
t.Fatalf("expected code 1 exit code, got %d, output: \n%s", code, testOutput.All())
}
output := testOutput.All()
expectedOutput := "Failed to select a workspace: Currently selected workspace \"default\" does not exist"
if !strings.Contains(cleanString(output), expectedOutput) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedOutput, cleanString(output))
}
statePath := filepath.Join(meta.DataDir(), DefaultStateFilename)
_, err := os.Stat(statePath)
if _, ok := err.(*os.PathError); !ok {
if err == nil {
t.Fatalf("expected backend state file to not be created, but it exists")
}
t.Fatalf("unexpected error: %s", err)
}
})
// Test what happens when the selected workspace doesn't exist, but there are other workspaces available.
//
// When input is enabled Terraform prompts the user to select an alternative.
t.Run("init: prompts user to select a workspace if the selected workspace doesn't exist and other custom workspaces do exist.", func(t *testing.T) {
// Create a temporary, uninitialized working directory with configuration including a state store
td := t.TempDir()
testCopyDir(t, testFixturePath("init-with-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProvider.GetStatesResponse = &providers.GetStatesResponse{
States: []string{
"foobar1",
"foobar2",
// Force provider to report workspaces exist
// But default workspace doesn't exist
},
}
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.0.0"},
})
defer close()
// Allow the test to respond to the prompt to pick an
// existing workspace, given the selected one doesn't exist.
_ = testInputMap(t, map[string]string{
"select-workspace": "1", // foobar1 in numbered list
})
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// The init command should have caused the selected workspace to change, based on the input
// provided by the user.
currentWorkspace, err := c.Meta.Workspace()
if err != nil {
t.Fatal(err)
}
if currentWorkspace != "foobar1" {
t.Fatalf("expected init command to alter the selected workspace from 'default' to 'foobar1', but got: %s", currentWorkspace)
}
})
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
// TODO(SarahFrench/radeksimko): Add test cases below:
// 1) "during a non-init command, the command ends in with an error telling the user to run an init command"
// >>> Currently this is handled at a lower level in `internal/command/meta_backend_test.go`
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Testing init's behaviors with `state_store` when run in a working directory where the configuration
// doesn't match the backend state file.
func TestInit_stateStore_configUnchanged(t *testing.T) {
// This matches the backend state test fixture in "state-store-unchanged"
v1_2_3, _ := version.NewVersion("1.2.3")
expectedState := &workdir.StateStoreConfigState{
Type: "test_store",
ConfigRaw: []byte("{\n \"value\": \"foobar\"\n }"),
Hash: uint64(4158988729),
Provider: &workdir.ProviderConfigState{
Version: v1_2_3,
Source: &tfaddr.Provider{
Hostname: tfaddr.DefaultProviderRegistryHost,
Namespace: "hashicorp",
Type: "test",
},
ConfigRaw: []byte("{\n \"region\": null\n }"),
},
}
t.Run("init is successful when the configuration and backend state match", func(t *testing.T) {
// Create a temporary working directory with state store configuration
// that matches the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-unchanged"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// If the working directory was previously initialized successfully then at least
// one workspace is guaranteed to exist when a user is re-running init with no config
// changes since last init. So this test says `default` exists.
mockProvider.GetStatesResponse = &providers.GetStatesResponse{
States: []string{"default"},
}
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
// Before running init, confirm the contents of the backend state file before
statePath := filepath.Join(meta.DataDir(), DefaultStateFilename)
sMgr := &clistate.LocalState{Path: statePath}
if err := sMgr.RefreshState(); err != nil {
t.Fatal("Failed to load state:", err)
}
s := sMgr.State()
if s == nil {
t.Fatal("expected backend state file to be present, but there isn't one")
}
if diff := cmp.Diff(s.StateStore, expectedState); diff != "" {
t.Fatalf("unexpected diff in backend state file's description of state store:\n%s", diff)
}
// Run init command
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutputs := []string{
"Initializing the state store...",
"Terraform has been successfully initialized!",
}
for _, expected := range expectedOutputs {
if !strings.Contains(output, expected) {
t.Fatalf("expected output to include %q, but got':\n %s", expected, output)
}
}
// Confirm init was a no-op and backend state is unchanged afterwards
if err := sMgr.RefreshState(); err != nil {
t.Fatal("Failed to load state:", err)
}
s = sMgr.State()
if diff := cmp.Diff(s.StateStore, expectedState); diff != "" {
t.Fatalf("unexpected diff in backend state file's description of state store:\n%s", diff)
}
})
}
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
// Testing init's behaviors with `state_store` when run in a working directory where the configuration
// doesn't match the backend state file.
func TestInit_stateStore_configChanges(t *testing.T) {
t.Run("the -reconfigure flag makes Terraform ignore the backend state file during initialization", func(t *testing.T) {
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
testCopyDir(t, testFixturePath("state-store-changed/store-config"), td)
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// The previous init implied by this test scenario would have created this.
mockProvider.GetStatesResponse = &providers.GetStatesResponse{States: []string{"default"}}
PSS: Add tests showing `state` subcommands being used with PSS (#37891) * test: Add E2E tests for `state list` and `state show` commands * test: Update `mockPluggableStateStorageProvider` to log a warning during tests where the values in `MockStates` aren't compatible with the `ReadStateBytesFn` default function. Make existing test set an appropriate value in `MockStates`. * test: Update `mockPluggableStateStorageProvider` helper to include a resource schema * test: Add command-level test for `state list` showing integration with pluggable state storage code. * test: Add command-level test for `state show` showing integration with pluggable state storage code. * test: Add command-level test for `state pull` showing integration with pluggable state storage code. * test: Add command-level test for `state identities` showing integration with pluggable state storage code. * test: Add command-level test for `state rm` showing integration with pluggable state storage code. * test: Add command-level test for `state mv` showing integration with pluggable state storage code. * test: Add command-level test for `state push` showing integration with pluggable state storage code. * test: Add command-level test for `state replace-provider` showing integration with pluggable state storage code. * test: Change shared test fixture to not be named after a specific command under test. This test fixure is reused across tests that need the config to define a state store but otherwise rely on the mock provider to set up the test scenario. * test: Update test to use shared test fixture * test: Remove redundant test fixture The internal/command/testdata/state-commands-state-store and internal/command/testdata/state-store-unchanged test fixtures are the same. * fix: Re-add logic for setting chunk size in the context of E2E tests using grpcwrap package This was removed, incorrectly, in https://github.com/hashicorp/terraform/pull/37899 * refactor: Let panic happen if there's incompatibility between mock returned from `mockPluggableStateStorageProvider` and the `MockStates` that's been set in the mock * test: Refactor to contain paths in reused variable, remove unnecessary .gitkeep * test: Remove unneeded test code
2025-12-09 12:24:38 -05:00
mockProvider.MockStates = map[string]interface{}{"default": []byte(`{"version": 4,"terraform_version":"1.15.0","serial": 1,"lineage": "","outputs": {},"resources": [],"checks":[]}`)}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-reconfigure",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutputs := []string{
"Initializing the state store...",
"Terraform has been successfully initialized!",
}
for _, expected := range expectedOutputs {
if !strings.Contains(output, expected) {
t.Fatalf("expected output to include %q, but got':\n %s", expected, output)
}
}
// Assert contents of the backend state file
statePath := filepath.Join(meta.DataDir(), DefaultStateFilename)
sMgr := &clistate.LocalState{Path: statePath}
if err := sMgr.RefreshState(); err != nil {
t.Fatal("Failed to load state:", err)
}
s := sMgr.State()
if s == nil {
t.Fatal("expected backend state file to be created, but there isn't one")
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
v1_2_3, _ := version.NewVersion("1.2.3")
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
expectedState := &workdir.StateStoreConfigState{
Type: "test_store",
ConfigRaw: []byte("{\n \"value\": \"changed-value\"\n }"),
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
Hash: uint64(1157855489), // The new hash after reconfiguring; this doesn't match the backend state test fixture
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
Provider: &workdir.ProviderConfigState{
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
Version: v1_2_3,
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
Source: &tfaddr.Provider{
Hostname: tfaddr.DefaultProviderRegistryHost,
Namespace: "hashicorp",
Type: "test",
},
ConfigRaw: []byte("{\n \"region\": null\n }"),
},
}
if diff := cmp.Diff(s.StateStore, expectedState); diff != "" {
t.Fatalf("unexpected diff in backend state file's description of state store:\n%s", diff)
}
})
t.Run("the -backend=false flag makes Terraform ignore config and use only the the backend state file during initialization", func(t *testing.T) {
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/store-config"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// The previous init implied by this test scenario would have created this.
mockProvider.GetStatesResponse = &providers.GetStatesResponse{States: []string{"default"}}
mockProvider.MockStates = map[string]interface{}{"default": []byte(`{"version": 4,"terraform_version":"1.15.0","serial": 1,"lineage": "","outputs": {},"resources": [],"checks":[]}`)}
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-backend=false",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected code 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedOutput := "Terraform has been successfully initialized!"
if !strings.Contains(output, expectedOutput) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedOutput, output)
}
// When -backend=false the backend/state store isn't initialized, so we don't expect this
// output if the flag has the expected effect on Terraform.
unexpectedOutput := "Initializing the state store..."
if strings.Contains(output, unexpectedOutput) {
t.Fatalf("output included %q, which is unexpected if -backend=false is behaving correctly':\n %s", unexpectedOutput, output)
}
})
t.Run("handling changed state store config without -migrate-state flag", func(t *testing.T) {
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/store-config"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProvider.GetStatesResponse = &providers.GetStatesResponse{States: []string{"default"}} // The previous init implied by this test scenario would have created the default workspace.
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
// missing -migrate-state flag
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("expected non-zero exit code, got %d, output: \n%s", code, testOutput.All())
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
// Check output
output := testOutput.All()
expectedMsg := "Error: State store configuration changed"
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
if !strings.Contains(output, expectedMsg) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedMsg, output)
}
})
t.Run("handling changed state store config", func(t *testing.T) {
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/store-config"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// The previous init implied by this test scenario would have created the default workspace.
mockProvider.MockStates = map[string]any{
backend.DefaultStateName: []byte(`{"version":4,"terraform_version":"1.15.0","serial":1,"lineage":"91adaece-23b3-7bce-0695-5aea537d2fef","outputs":{"test":{"value":"test","type":"string"}},"resources":[],"check_results":null}`),
}
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected 0 exit code, got %d, output: \n%s", code, testOutput.All())
}
// Check output
output := testOutput.All()
expectedMsg := "Terraform has been successfully initialized!"
if !strings.Contains(output, expectedMsg) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedMsg, output)
}
expectedReason := "State store \"test_store\" (hashicorp/test) configuration changed"
if !strings.Contains(output, expectedReason) {
t.Fatalf("expected output to include reason %q, but got':\n %s", expectedReason, output)
}
// check state remains accessible after migration
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; !exists {
t.Fatalf("expected the default workspace to exist after migration, but it is missing: %#v", mockProvider.MockStates)
}
})
t.Run("handling changed state store provider config", func(t *testing.T) {
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/provider-config"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// The previous init implied by this test scenario would have created the default workspace.
mockProvider.MockStates = map[string]any{
backend.DefaultStateName: []byte(`{"version":4,"terraform_version":"1.15.0","serial":1,"lineage":"91adaece-23b3-7bce-0695-5aea537d2fef","outputs":{"test":{"value":"test","type":"string"}},"resources":[],"check_results":null}`),
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected 0 exit code, got %d, output: \n%s", code, testOutput.All())
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
// Check output
output := testOutput.All()
expectedMsg := "Terraform has been successfully initialized!"
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
if !strings.Contains(output, expectedMsg) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedMsg, output)
}
expectedReason := "State store provider \"test\" (hashicorp/test) configuration changed"
if !strings.Contains(output, expectedReason) {
t.Fatalf("expected output to include reason %q, but got':\n %s", expectedReason, output)
}
// check state remains accessible after migration
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; !exists {
t.Fatal("expected the default workspace to exist after migration, but it is missing")
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
})
t.Run("handling changed state store type in the same provider", func(t *testing.T) {
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/state-store-type"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// The previous init implied by this test scenario would have created the default workspace.
mockProvider.MockStates = map[string]any{
backend.DefaultStateName: []byte(`{"version":4,"terraform_version":"1.15.0","serial":1,"lineage":"91adaece-23b3-7bce-0695-5aea537d2fef","outputs":{"test":{"value":"test","type":"string"}},"resources":[],"check_results":null}`),
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
storeName := "test_store"
otherStoreName := "test_otherstore"
// Make the provider report that it contains a 2nd storage implementation with the above name
mockProvider.GetProviderSchemaResponse.StateStores[otherStoreName] = mockProvider.GetProviderSchemaResponse.StateStores[storeName]
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected 0 exit code, got %d, output: \n%s", code, testOutput.All())
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
// Check output
output := testOutput.All()
expectedMsg := "Terraform has been successfully initialized!"
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
if !strings.Contains(output, expectedMsg) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedMsg, output)
}
expectedReason := "State store type changed from \"test_store\" to \"test_otherstore\""
if !strings.Contains(output, expectedReason) {
t.Fatalf("expected output to include reason %q, but got':\n %s", expectedReason, output)
}
// check state remains accessible after migration
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; !exists {
t.Fatal("expected the default workspace to exist after migration, but it is missing")
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
})
t.Run("handling changing the provider used for state storage", func(t *testing.T) {
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/provider-used"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// The previous init implied by this test scenario would have created the default workspace.
mockProvider.MockStates = map[string]any{
backend.DefaultStateName: []byte(`{"version":4,"terraform_version":"1.15.0","serial":1,"lineage":"91adaece-23b3-7bce-0695-5aea537d2fef","outputs":{"test":{"value":"test","type":"string"}},"resources":[],"check_results":null}`),
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Make a mock that implies its name is test2 based on returned schemas
mockProvider2 := mockPluggableStateStorageProvider()
mockProvider2.GetProviderSchemaResponse.StateStores["test2_store"] = mockProvider.GetProviderSchemaResponse.StateStores["test_store"]
delete(mockProvider2.GetProviderSchemaResponse.StateStores, "test_store")
mockProviderAddress := addrs.NewDefaultProvider("test")
mockProviderAddress2 := addrs.NewDefaultProvider("test2")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Provider in backend state file fixture
"hashicorp/test2": {"1.2.3"}, // Provider now used in config
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider), // test provider
mockProviderAddress2: providers.FactoryFixed(mockProvider2), // test2 provider
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected 0 exit code, got %d, output: \n%s", code, testOutput.All())
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
// Check output
output := testOutput.All()
expectedMsg := "Terraform has been successfully initialized!"
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
if !strings.Contains(output, expectedMsg) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedMsg, output)
}
expectedReason := "State store provider changed from hashicorp/test to hashicorp/test2"
if !strings.Contains(output, expectedReason) {
t.Fatalf("expected output to include reason %q, but got':\n %s", expectedReason, output)
}
// check state remains accessible after migration
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; !exists {
t.Fatal("expected the default workspace to exist after migration, but it is missing")
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
})
}
// Testing init's behaviors with `state_store` when the provider used for state storage in a previous init
// command is updated.
//
// TODO: Add a test case showing that downgrading provider version is ok as long as the schema version hasn't
// changed. We should also have a test demonstrating that downgrades when the schema version HAS changed will fail.
func TestInit_stateStore_providerUpgrade(t *testing.T) {
t.Run("handling upgrading the provider used for state storage", func(t *testing.T) {
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
// Create a temporary working directory with state store configuration
// that doesn't match the backend state file
td := t.TempDir()
testCopyDir(t, testFixturePath("state-store-changed/provider-upgraded"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// The previous init implied by this test scenario would have created the default workspace.
mockProvider.MockStates = map[string]any{
backend.DefaultStateName: []byte(`{"version":4,"terraform_version":"1.15.0","serial":1,"lineage":"91adaece-23b3-7bce-0695-5aea537d2fef","outputs":{"test":{"value":"test","type":"string"}},"resources":[],"check_results":null}`),
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3", "9.9.9"}, // 1.2.3 is the version used in the backend state file, 9.9.9 is the version being upgraded to
})
defer close()
ui := new(cli.MockUi)
view, done := testView(t)
meta := Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
}
c := &InitCommand{
Meta: meta,
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-migrate-state=true",
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
"-upgrade",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("expected 0 exit code, got %d, output: \n%s", code, testOutput.All())
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
}
// Check output
output := testOutput.All()
expectedMsg := "Terraform has been successfully initialized!"
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
if !strings.Contains(output, expectedMsg) {
t.Fatalf("expected output to include %q, but got':\n %s", expectedMsg, output)
}
expectedReason := "State store provider \"test\" (hashicorp/test) version changed from 1.2.3 to 9.9.9"
if !strings.Contains(output, expectedReason) {
t.Fatalf("expected output to include reason %q, but got':\n %s", expectedReason, output)
}
// check state remains accessible after migration
if _, exists := mockProvider.MockStates[backend.DefaultStateName]; !exists {
t.Fatal("expected the default workspace to exist after migration, but it is missing")
}
PSS: Let the `init` command recognise when there are no changes in configuration. (#37777) * Pull determining of PSS provider's version from current locks into a separate method * Add code for identifying when config and provider version match existing backend state (i.e. no changes) * Update test - locks are now needed before it hits expected error diag return * Add test showing successful init when no config changes are detected. * Update `getStateStorageProviderVersion` to return nil versions for builtin and re-attached providers. This makes comparison easier when determining if config has changed since last init. * Add test coverage for `getStateStorageProviderVersion` * Move testing fixtures around, preparing for different types of changed state_store config changes being tested * Add test showing that changing the state_store config is detected as a change, but handling this scenario isn't implemented yet * Update hashes in test fixture backend state file to be accurate Previously dummy values were fine, but as tests using hashes to identify changes these values need to be accurate! * Update existing test cases so that Terraform uses the same test provider version as described in the backend state file fixture for the test. * Add test showing that changing the PSS provider's config is detected as a change, but handling this scenario isn't implemented yet * Add test showing that swapping to a different state storage implementation in the same provider is detected as a change, but handling this scenario isn't implemented yet * Add test showing that changing the provider used for PSS is detected as a change, but handling this scenario isn't implemented yet * Add test showing that upgrading a provider is detected as a change, but handling this scenario isn't implemented yet * Update test to use v1.2.3 for consistency with other tests Just to avoid any confusion if copy-pasting happens in future. * More corrections to existing test fixtures - unset config should be null, and replace dummy hash values with correct values. * Fix test for using -reconfigure with state_store; the default workspace would already exist in this scenario * Update TestInit_stateStore_configUnchanged to assert that init was a no-op for backend state * Remove unused fixture * Remove test that's replaced by new tests in command/init_test.go * Replace old references to deleted "state-store-changed" test fixture & update test to not expect a value for region attr in provider config * Make test fixture coupling a little more understandable * Refactor detection of no need to migrate into a function * Add TODO about more involved provider version change tests We will allow downgrades to succeed as long as the schema version number is unchanged * Update (configs.StateStore)Hash method to return a single hash that's impacted by: state store config, provider config, state store type, provider source * Update calling code and test helper code to reflect that the nested provider block no longer has its own hash * Remove test; there is now a single hash that SHOULD be affected by the provider block! * Also use provider name, from config, in hash * Update tests to reflect changes in how hashes are made * Remove unused `stateStoreConfigNeedsMigration` function * Remove duplicate isProviderReattached function. * Fixes to affected tests * Allow provider version to impact the state storage hash, update impacted tests and test fixtures * Update tests that now require locks data to be present in test setup * Update comment for accuracy * Fixes to other test fixtures - remove excess hash field, set hash to 0 to indicate they're not set accurately. * Make upgrade test actually use upgrade code path * Add lock files to test fixture directories that represent a project that's had a successful prior init using PSS
2025-10-20 11:51:45 -04:00
})
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
}
// Test a scenario where the configuration changes but the -backend-config CLI flags compensate for those changes
// to result in the state store being configured in the same way. In this scenario the user isn't prompted to migrate
// state but the backend state file is updated with a new hash to reflect the new configuration's values.
func TestInit_stateStore_backendConfigFlagNoMigrate(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
// Make the state store's value attribute optional in this test.
mockProvider.GetProviderSchemaResponse.StateStores["test_store"].Body.Attributes["value"].Required = false
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
t.Cleanup(close)
var originalStateStoreConfigHash uint64
{
log.Printf("[TRACE] TestInit_stateStore_unset: beginning first init")
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
// Init
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_unset: first init complete")
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if s.StateStore.Empty() {
t.Fatal("should have StateStore config")
}
// Store this for comparison after the 2nd init
originalStateStoreConfigHash = s.StateStore.Hash
}
{
log.Printf("[TRACE] TestInit_stateStore_unset: beginning second init with changed config but compensating CLI flags")
// Remove `value` attribute from config
cfg := `terraform {
state_store "test_store" {
provider "test" {}
# value attr removed here
}
}`
if err := os.WriteFile("main.tf", []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-backend-config=value=foobar", // value = foobar, matches the line removed from config
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("Terraform either experienced an unexpected error, or suggested a state migration when this test scenario should not include migrations: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_unset: second init complete")
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if s.StateStore.Empty() {
t.Fatal("should have StateStore config")
}
if s.StateStore.Hash == originalStateStoreConfigHash {
t.Fatal("expected second init to update the state_store config hash in the backend state file, but it did not")
}
}
}
func TestInit_stateStore_unset(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
storeName := "test_store"
otherStoreName := "test_otherstore"
// Make the provider report that it contains a 2nd storage implementation with the above name
mockProvider.GetProviderSchemaResponse.StateStores[otherStoreName] = mockProvider.GetProviderSchemaResponse.StateStores[storeName]
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
{
log.Printf("[TRACE] TestInit_stateStore_unset: beginning first init")
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
// Init
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_unset: first init complete")
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if _, err := os.Stat(filepath.Join(DefaultDataDir, DefaultStateFilename)); err != nil {
t.Fatalf("err: %s", err)
}
}
{
log.Printf("[TRACE] TestInit_stateStore_unset: beginning second init")
// Unset
if err := os.WriteFile("main.tf", []byte(""), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_unset: second init complete")
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if !s.StateStore.Empty() {
t.Fatal("should not have StateStore config")
}
if !s.Backend.Empty() {
t.Fatalf("expected empty Backend config after unsetting state store, found: %#v", s.Backend)
}
}
}
func TestInit_stateStore_unset_withoutProviderRequirements(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
storeName := "test_store"
otherStoreName := "test_otherstore"
// Make the provider report that it contains a 2nd storage implementation with the above name
mockProvider.GetProviderSchemaResponse.StateStores[otherStoreName] = mockProvider.GetProviderSchemaResponse.StateStores[storeName]
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
{
log.Printf("[TRACE] TestInit_stateStore_unset_withoutProviderRequirements: beginning first init")
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
// Init
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_unset_withoutProviderRequirements: first init complete")
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if _, err := os.Stat(filepath.Join(DefaultDataDir, DefaultStateFilename)); err != nil {
t.Fatalf("err: %s", err)
}
}
{
log.Printf("[TRACE] TestInit_stateStore_unset_withoutProviderRequirements: beginning second init")
// Unset state store and provider requirements
if err := os.WriteFile("main.tf", []byte(""), 0644); err != nil {
t.Fatalf("err: %s", err)
}
if err := os.WriteFile("providers.tf", []byte(""), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_unset_withoutProviderRequirements: second init complete")
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if !s.StateStore.Empty() {
t.Fatal("should not have StateStore config")
}
if !s.Backend.Empty() {
t.Fatalf("expected empty Backend config after unsetting state store, found: %#v", s.Backend)
}
}
}
func TestInit_stateStore_to_backend(t *testing.T) {
// Create a temporary working directory and copy in test fixtures
td := t.TempDir()
testCopyDir(t, testFixturePath("init-state-store"), td)
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"}, // Matches provider version in backend state file fixture
})
defer close()
tOverrides := &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
}
{
log.Printf("[TRACE] TestInit_stateStore_to_backend: beginning first init")
// Init
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_to_backend: first init complete")
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if _, err := os.Stat(filepath.Join(DefaultDataDir, DefaultStateFilename)); err != nil {
t.Fatalf("err: %s", err)
}
}
{
// run apply to ensure state isn't empty
// to bypass edge case handling which causes empty state to stop migration
log.Printf("[TRACE] TestInit_stateStore_to_backend: beginning apply")
ui := cli.NewMockUi()
aView, aDone := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: aView,
AllowExperimentalFeatures: true,
},
}
aCode := cApply.Run([]string{"-auto-approve"})
aTestOutput := aDone(t)
if aCode != 0 {
t.Fatalf("bad: \n%s", aTestOutput.All())
}
t.Logf("Apply output:\n%s", aTestOutput.Stdout())
t.Logf("Apply errors:\n%s", aTestOutput.Stderr())
}
{
log.Printf("[TRACE] TestInit_stateStore_to_backend: beginning uninitialised apply")
backendCfg := []byte(`terraform {
backend "http" {
address = "https://example.com"
}
}
`)
if err := os.WriteFile("main.tf", backendCfg, 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
code := cApply.Run([]string{"-auto-approve"})
testOutput := done(t)
if code == 0 {
t.Fatalf("expected apply to fail: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_to_backend: apply complete")
expectedErr := "Backend initialization required"
if !strings.Contains(testOutput.Stderr(), expectedErr) {
t.Fatalf("unexpected error, expected %q, given: %q", expectedErr, testOutput.Stderr())
}
log.Printf("[TRACE] TestInit_stateStore_to_backend: uninitialised apply complete")
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if _, err := os.Stat(filepath.Join(DefaultDataDir, DefaultStateFilename)); err != nil {
t.Fatalf("err: %s", err)
}
}
{
log.Printf("[TRACE] TestInit_stateStore_to_backend: beginning second init")
testBackend := new(httpBackend.TestHTTPBackend)
ts := httptest.NewServer(http.HandlerFunc(testBackend.Handle))
t.Cleanup(ts.Close)
// Override state store to backend
backendCfg := fmt.Sprintf(`terraform {
backend "http" {
address = %q
}
}
`, ts.URL)
if err := os.WriteFile("main.tf", []byte(backendCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
},
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-migrate-state",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_to_backend: second init complete")
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if !s.StateStore.Empty() {
t.Fatal("should not have StateStore config")
}
if s.Backend.Empty() {
t.Fatalf("expected backend to not be empty")
}
data, err := statefile.Read(bytes.NewBuffer(testBackend.Data))
if err != nil {
t.Fatal(err)
}
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data: %s", diff)
}
expectedGetCalls := 6
if testBackend.CallCount("GET") != expectedGetCalls {
t.Fatalf("expected %d GET calls, got %d", expectedGetCalls, testBackend.CallCount("GET"))
}
expectedPostCalls := 1
if testBackend.CallCount("POST") != expectedPostCalls {
t.Fatalf("expected %d POST calls, got %d", expectedPostCalls, testBackend.CallCount("POST"))
}
}
}
// Test that users are shown actionable errors if they try to use a state store in a non-init command
// before running an init operation to download the state storage provider and record it in the dependency lock file.
func TestInit_uninitialized_stateStore(t *testing.T) {
t.Run("error if working directory isn't initialized before apply", func(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
cfg := `terraform {
required_providers {
test = {
source = "hashicorp/test"
}
}
state_store "test_store" {
provider "test" {}
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
t.Chdir(td)
ui := cli.NewMockUi()
view, done := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
code := cApply.Run([]string{"-no-color"})
testOutput := done(t)
if code == 0 {
t.Fatalf("expected apply to fail: \n%s", testOutput.All())
}
log.Printf("[TRACE] TestInit_stateStore_to_backend: uninitialised apply with state store complete")
expectedErrMsgs := []string{
"The provider dependency used for state storage is missing from the lock file despite being present in the current configuration",
`provider registry.terraform.io/hashicorp/test: required by this configuration but no version is selected`,
}
for _, expectedErr := range expectedErrMsgs {
if !strings.Contains(cleanString(testOutput.Stderr()), expectedErr) {
t.Fatalf("unexpected error, expected %q, given: %s", expectedErr, testOutput.Stderr())
}
}
})
t.Run("the error isn't shown if the provider is supplied through reattach config", func(t *testing.T) {
t.Skip("This is implemented as an E2E test: TestPrimary_stateStore_unmanaged_separatePlan")
})
}
func TestInit_backend_to_stateStore_singleWorkspace(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
testBackend := new(httpBackend.TestHTTPBackend)
ts := httptest.NewServer(http.HandlerFunc(testBackend.Handle))
t.Cleanup(ts.Close)
cfg := fmt.Sprintf(`terraform {
backend "http" {
address = %q
}
}
`, ts.URL)
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
tOverrides := &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
}
{
log.Printf("[TRACE] %s: beginning first init with backend", t.Name())
// Init
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] %s: first init complete", t.Name())
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if testBackend.CallCount("POST") != 0 {
t.Fatalf("expected 0 POST calls after init, got %d", testBackend.CallCount("POST"))
}
if testBackend.CallCount("GET") != 2 {
t.Fatalf("expected 2 GET calls after init, got %d", testBackend.CallCount("GET"))
}
}
{
// run apply to ensure state isn't empty
// to bypass edge case handling which causes empty state to stop migration
log.Printf("[TRACE] %s: beginning apply with backend", t.Name())
outputCfg := `output "test" {
value = "test"
}
`
if err := os.WriteFile(filepath.Join(td, "output.tf"), []byte(outputCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
aView, aDone := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
Ui: ui,
View: aView,
AllowExperimentalFeatures: true,
},
}
aCode := cApply.Run([]string{"-auto-approve"})
aTestOutput := aDone(t)
if aCode != 0 {
t.Fatalf("bad: \n%s", aTestOutput.All())
}
t.Logf("Apply output:\n%s", aTestOutput.Stdout())
t.Logf("Apply errors:\n%s", aTestOutput.Stderr())
if testBackend.CallCount("POST") != 1 {
t.Fatalf("expected 1 POST call after apply, got %d", testBackend.CallCount("POST"))
}
if testBackend.CallCount("GET") != 5 {
t.Fatalf("expected 5 GET calls after apply, got %d", testBackend.CallCount("GET"))
}
data, err := statefile.Read(bytes.NewBuffer(testBackend.Data))
if err != nil {
t.Fatal(err)
}
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data after apply: %s", diff)
}
}
{
log.Printf("[TRACE] %s: beginning second init with state store", t.Name())
ssCfg := `terraform {
required_providers {
test = {
source = "hashicorp/test"
}
}
state_store "test_store" {
provider "test" {}
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(ssCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] %s: second init with state store complete", t.Name())
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if s.StateStore.Empty() {
t.Fatal("should have StateStore config")
}
if !s.Backend.Empty() {
t.Fatalf("expected backend to be empty")
}
rawData, ok := mockProvider.MockStates[backend.DefaultStateName].([]byte)
if !ok {
t.Fatalf("expected %q state to exist in %s: %#v",
backend.DefaultStateName,
mockProviderAddress,
mockProvider.MockStates)
}
data, err := statefile.Read(bytes.NewBuffer(rawData))
if err != nil {
t.Fatal(err)
}
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data: %s", diff)
}
}
}
// TestInit_backend_to_stateStore_noState tests that given no state
// in the source backend, no state is created in the destination state store
// as a result of the migration.
func TestInit_backend_to_stateStore_noState(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
testBackend := new(httpBackend.TestHTTPBackend)
ts := httptest.NewServer(http.HandlerFunc(testBackend.Handle))
t.Cleanup(ts.Close)
cfg := fmt.Sprintf(`terraform {
backend "http" {
address = %q
}
}
`, ts.URL)
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
tOverrides := &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
}
{
log.Printf("[TRACE] %s: beginning first init with backend", t.Name())
// Init
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("first init exited with non-zero code %d:\n%s", code, testOutput.Stderr())
}
log.Printf("[TRACE] %s: first init complete", t.Name())
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
if testBackend.CallCount("POST") != 0 {
t.Fatalf("expected 0 POST calls after init, got %d", testBackend.CallCount("POST"))
}
if testBackend.CallCount("GET") != 2 {
t.Fatalf("expected 2 GET calls after init, got %d", testBackend.CallCount("GET"))
}
}
{
log.Printf("[TRACE] %s: beginning second init with state store", t.Name())
ssCfg := `terraform {
required_providers {
test = {
source = "hashicorp/test"
}
}
state_store "test_store" {
provider "test" {}
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(ssCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("second init exited with non-zero code %d:\n%s", code, testOutput.Stderr())
}
log.Printf("[TRACE] %s: second init with state store complete", t.Name())
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if s.StateStore.Empty() {
t.Fatal("should have StateStore config")
}
if !s.Backend.Empty() {
t.Fatalf("expected backend to be empty")
}
if len(mockProvider.MockStates) != 0 {
t.Fatalf("expected no state to exist in %s: %#v",
mockProviderAddress,
mockProvider.MockStates)
}
}
}
func TestInit_localBackend_to_stateStore(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
cfg := `terraform {
backend "local" {}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
tOverrides := &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
}
{
log.Printf("[TRACE] %s: beginning first init with local backend", t.Name())
// Init
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("first init exited with non-zero code %d:\n%s", code, testOutput.Stderr())
}
log.Printf("[TRACE] %s: first init complete", t.Name())
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
}
{
// run apply to ensure state isn't empty
// to bypass edge case handling which causes empty state to stop migration
log.Printf("[TRACE] %s: beginning apply with backend", t.Name())
outputCfg := `output "test" {
value = "test"
}
`
if err := os.WriteFile(filepath.Join(td, "output.tf"), []byte(outputCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
aView, aDone := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
Ui: ui,
View: aView,
AllowExperimentalFeatures: true,
},
}
aCode := cApply.Run([]string{"-auto-approve"})
aTestOutput := aDone(t)
if aCode != 0 {
t.Fatalf("bad: \n%s", aTestOutput.All())
}
t.Logf("Apply output:\n%s", aTestOutput.Stdout())
t.Logf("Apply errors:\n%s", aTestOutput.Stderr())
b, err := os.ReadFile(DefaultStateFilename)
if err != nil {
t.Fatalf("unable to read state file: %s", err)
}
data, err := statefile.Read(bytes.NewBuffer(b))
if err != nil {
t.Fatal(err)
}
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data after apply: %s", diff)
}
}
{
log.Printf("[TRACE] %s: beginning second init with state store", t.Name())
ssCfg := `terraform {
required_providers {
test = {
source = "hashicorp/test"
}
}
state_store "test_store" {
provider "test" {}
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(ssCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("second init exited with non-zero code %d:\n%s", code, testOutput.Stderr())
}
log.Printf("[TRACE] %s: second init with state store complete", t.Name())
t.Logf("Second run output:\n%s", testOutput.Stdout())
t.Logf("Second run errors:\n%s", testOutput.Stderr())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if s.StateStore.Empty() {
t.Fatal("should have StateStore config")
}
if !s.Backend.Empty() {
t.Fatalf("expected backend to be empty")
}
rawData, ok := mockProvider.MockStates[backend.DefaultStateName].([]byte)
if !ok {
t.Fatalf("expected %q state to exist in %s: %#v",
backend.DefaultStateName,
mockProviderAddress,
mockProvider.MockStates)
}
data, err := statefile.Read(bytes.NewBuffer(rawData))
if err != nil {
t.Fatal(err)
}
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data: %s", diff)
}
if f, err := os.Stat(DefaultStateFilename); err == nil && f.Size() > 0 {
t.Fatalf("expected state file to have been removed at %q. Has size %d bytes.", DefaultStateFilename, f.Size())
}
}
}
func TestInit_backend_to_stateStore_multipleWorkspaces(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
cfg := `terraform {
backend "inmem" {}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
tOverrides := &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
}
{
log.Printf("[TRACE] %s: beginning first init with backend", t.Name())
// Init
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] %s: first init complete", t.Name())
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
}
{
// run apply to ensure state isn't empty
// to bypass edge case handling which causes empty state to stop migration
log.Printf("[TRACE] %s: beginning first apply to default workspace with backend", t.Name())
outputCfg := `output "test" {
value = "test"
}
`
if err := os.WriteFile(filepath.Join(td, "output.tf"), []byte(outputCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
aView, aDone := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
Ui: ui,
View: aView,
AllowExperimentalFeatures: true,
},
}
aCode := cApply.Run([]string{"-auto-approve"})
aTestOutput := aDone(t)
if aCode != 0 {
t.Fatalf("bad: \n%s", aTestOutput.All())
}
t.Logf("Apply output:\n%s", aTestOutput.Stdout())
t.Logf("Apply errors:\n%s", aTestOutput.Stderr())
data := inmem.ReadState(t, backend.DefaultStateName)
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.RootOutputValues); diff != "" {
t.Fatalf("unexpected data after apply: %s", diff)
}
}
{
ui := cli.NewMockUi()
aView, aDone := testView(t)
cSelect := &WorkspaceSelectCommand{
Meta: Meta{
Ui: ui,
View: aView,
AllowExperimentalFeatures: true,
},
}
sCode := cSelect.Run([]string{"-or-create", "second"})
aTestOutput := aDone(t)
if sCode != 0 {
t.Fatalf("unable to select workspace: \n%s", aTestOutput.All())
}
t.Logf("Select workspace output:\n%s", aTestOutput.All())
}
{
ui := cli.NewMockUi()
aView, aDone := testView(t)
cApply := &ApplyCommand{
Meta: Meta{
Ui: ui,
View: aView,
AllowExperimentalFeatures: true,
},
}
aCode := cApply.Run([]string{"-auto-approve"})
aTestOutput := aDone(t)
if aCode != 0 {
t.Fatalf("bad: \n%s", aTestOutput.All())
}
t.Logf("Apply output:\n%s", aTestOutput.Stdout())
t.Logf("Apply errors:\n%s", aTestOutput.Stderr())
data := inmem.ReadState(t, "second")
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
if diff := cmp.Diff(expectedOutputs, data.RootOutputValues); diff != "" {
t.Fatalf("unexpected data after apply: %s", diff)
}
}
{
log.Printf("[TRACE] %s: beginning second init with state store", t.Name())
ssCfg := `terraform {
required_providers {
test = {
source = "hashicorp/test"
}
}
state_store "test_store" {
provider "test" {}
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(ssCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
"-force-copy",
"-migrate-state",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("second init failed: \n%s", testOutput.All())
}
log.Printf("[TRACE] %s: second init with state store complete", t.Name())
t.Logf("Second init output:\n%s", testOutput.All())
s := testDataStateRead(t, filepath.Join(DefaultDataDir, DefaultStateFilename))
if s.StateStore.Empty() {
t.Fatal("should have StateStore config")
}
if !s.Backend.Empty() {
t.Fatalf("expected backend to be empty")
}
expectedOutputs := map[string]*states.OutputValue{
"test": {
Addr: addrs.AbsOutputValue{
OutputValue: addrs.OutputValue{
Name: "test",
},
},
Value: cty.StringVal("test"),
},
}
expectedWorkspaces := []string{"default", "second"}
ws := slices.Sorted(maps.Keys(mockProvider.MockStates))
if diff := cmp.Diff(expectedWorkspaces, ws); diff != "" {
t.Fatalf("unexpected workspaces: %s", diff)
}
// check default workspace first
rawData, ok := mockProvider.MockStates[backend.DefaultStateName].([]byte)
if !ok {
t.Fatalf("expected %q state to exist in %s: %#v",
backend.DefaultStateName,
mockProviderAddress,
mockProvider.MockStates)
}
stateData, err := statefile.Read(bytes.NewBuffer(rawData))
if err != nil {
t.Fatal(err)
}
if diff := cmp.Diff(expectedOutputs, stateData.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data: %s", diff)
}
// check second workspace
rawData2, ok := mockProvider.MockStates["second"].([]byte)
if !ok {
t.Fatalf("expected %q state to exist in %s: %#v",
backend.DefaultStateName,
mockProviderAddress,
mockProvider.MockStates)
}
stateData2, err := statefile.Read(bytes.NewBuffer(rawData2))
if err != nil {
t.Fatal(err)
}
if diff := cmp.Diff(expectedOutputs, stateData2.State.RootOutputValues); diff != "" {
t.Fatalf("unexpected data: %s", diff)
}
}
}
func TestInit_cloud_to_stateStore(t *testing.T) {
// Create a temporary working directory that is empty
td := t.TempDir()
ts := cloud.TestServerWithHandlers(t, map[string]func(http.ResponseWriter, *http.Request){
"/api/v2/organizations/hashicorp/workspaces/test": func(w http.ResponseWriter, r *http.Request) {
if r.Method == "GET" {
w.Write([]byte(`{"data":{"id":"ws-TEST","type":"workspaces","attributes":{"allow-destroy-plan":true,"auto-apply":false,"auto-apply-run-trigger":false,"auto-destroy-activity-duration":null,"auto-destroy-at":null,"auto-destroy-status":null,"inherits-project-auto-destroy":true,"created-at":"2022-06-22T14:24:13.836Z","environment":"default","locked":false,"name":"test","queue-all-runs":false,"speculative-enabled":true,"structured-run-output-enabled":true,"terraform-version":"1.10.0","working-directory":null,"global-remote-state":false,"updated-at":"2026-01-29T15:09:18.075Z","resource-count":0,"apply-duration-average":2000,"plan-duration-average":4000,"policy-check-failures":0,"run-failures":0,"workspace-kpis-runs-count":1,"unarchived-workspace-change-requests-count":0,"latest-change-at":"2026-01-29T15:09:17.200Z","operations":true,"execution-mode":"remote","vcs-repo":null,"vcs-repo-identifier":null,"permissions":{"can-update":true,"can-destroy":true,"can-queue-run":true,"can-read-run":true,"can-read-variable":true,"can-update-variable":true,"can-read-state-versions":true,"can-read-state-outputs":true,"can-create-state-versions":true,"can-queue-apply":true,"can-lock":true,"can-unlock":true,"can-force-unlock":true,"can-read-settings":true,"can-manage-tags":true,"can-manage-run-tasks":true,"can-force-delete":true,"can-manage-assessments":true,"can-manage-ephemeral-workspaces":false,"can-read-assessment-results":true,"can-read-change-requests":false,"can-update-change-requests":false,"can-queue-destroy":true},"actions":{"is-destroyable":true},"description":null,"file-triggers-enabled":true,"trigger-prefixes":[],"trigger-patterns":[],"assessments-enabled":false,"last-assessment-result-at":null,"locked-reason":"","source":"terraform","source-name":null,"source-url":null,"tag-names":[],"setting-overwrites":{"execution-mode":true,"agent-pool":true}},"relationships":{"organization":{"data":{"id":"hashicorp","type":"organizations"}},"current-run":{"data":{"id":"run-TEST","type":"runs"},"links":{"related":"/api/v2/runs/run-TEST"}},"latest-run":{"data":{"id":"run-TEST","type":"runs"},"links":{"related":"/api/v2/runs/run-TEST"}},"outputs":{"data":[{"id":"wsout-TEST","type":"workspace-outputs"}],"links":{"related":"/api/v2/workspaces/ws-TEST/current-state-version-outputs"}},"remote-state-consumers":{"links":{"related":"/api/v2/workspaces/ws-TEST/relationships/remote-state-consumers"}},"current-state-version":{"data":{"id":"sv-TEST","type":"state-versions"},"links":{"related":"/api/v2/workspaces/ws-TEST/current-state-version"}},"current-configuration-version":{"data":{"id":"cv-TEST","type":"configuration-versions"},"links":{"related":"/api/v2/configuration-versions/cv-TEST"}},"agent-pool":{"data":null},"readme":{"data":null},"project":{"data":{"id":"prj-TEST","type":"projects"}},"current-assessment-result":{"data":null},"vars":{"data":[]}},"links":{"self":"/api/v2/organizations/hashicorp/workspaces/test","self-html":"/app/hashicorp/workspaces/test"}}}`))
w.WriteHeader(http.StatusOK)
return
}
},
"/api/v2/workspaces/ws-TEST/current-state-version": func(w http.ResponseWriter, r *http.Request) {
hostname := r.URL.Hostname()
w.Write(fmt.Appendf([]byte{}, `{"data":{"id":"sv-TEST","type":"state-versions","attributes":{"created-at":"2026-01-29T15:09:17.200Z","size":651,"hosted-state-download-url":"%s/api/state-versions/sv-TEST/hosted_state","hosted-json-state-download-url":"%s/api/state-versions/sv-TEST/hosted_json_state","modules":{},"providers":{},"resources-processed":true,"serial":1,"state-version":4,"status":"finalized","terraform-version":"1.10.0","vcs-commit-url":null,"vcs-commit-sha":null,"resources":[],"billable-rum-count":0},"relationships":{"run":{"data":{"id":"run-TEST","type":"runs"}},"rollback-state-version":{"data":null},"created-by":{"data":{"id":"user-TEST","type":"users"},"links":{"self":"/api/v2/users/user-TEST","related":"/api/v2/runs/run-TEST/created-by"}},"workspace":{"data":{"id":"ws-TEST","type":"workspaces"}},"outputs":{"data":[{"id":"wsout-TEST","type":"state-version-outputs"}],"links":{"related":"/api/v2/state-versions/sv-TEST/outputs"}}},"links":{"self":"/api/v2/state-versions/sv-TEST"}}}`, hostname, hostname))
w.WriteHeader(http.StatusOK)
},
"/api/state-versions/sv-TEST/hosted_state": func(w http.ResponseWriter, r *http.Request) {
w.Write([]byte(`{"version":4,"terraform_version":"1.15.0","serial":1,"lineage":"91adaece-23b3-7bce-0695-5aea537d2fef","outputs":{"test":{"value":"test","type":"string"}},"resources":[],"check_results":null}`))
w.WriteHeader(http.StatusOK)
},
})
t.Cleanup(ts.Close)
mockURL, err := url.Parse(ts.URL)
if err != nil {
t.Fatal(err)
}
backendInit.Init(testDisco(ts))
t.Cleanup(func() { backendInit.Init(nil) })
cfg := fmt.Sprintf(`terraform {
cloud {
hostname = %q
organization = "hashicorp"
token = "test-token"
workspaces {
name = "test"
}
}
}
`, mockURL.Host)
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
t.Chdir(td)
mockProvider := mockPluggableStateStorageProvider()
mockProviderAddress := addrs.NewDefaultProvider("test")
providerSource, close := newMockProviderSource(t, map[string][]string{
"hashicorp/test": {"1.2.3"},
})
defer close()
tOverrides := &testingOverrides{
Providers: map[addrs.Provider]providers.Factory{
mockProviderAddress: providers.FactoryFixed(mockProvider),
},
}
{
log.Printf("[TRACE] %s: beginning first init with backend", t.Name())
// Init
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Services: testDisco(ts),
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code != 0 {
t.Fatalf("bad: \n%s", testOutput.All())
}
log.Printf("[TRACE] %s: first init complete", t.Name())
t.Logf("First run output:\n%s", testOutput.Stdout())
t.Logf("First run errors:\n%s", testOutput.Stderr())
}
{
log.Printf("[TRACE] %s: beginning second init with state store", t.Name())
ssCfg := `terraform {
required_providers {
test = {
source = "hashicorp/test"
}
}
state_store "test_store" {
provider "test" {}
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(ssCfg), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
c := &InitCommand{
Meta: Meta{
Services: testDisco(ts),
testingOverrides: tOverrides,
ProviderSource: providerSource,
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
args := []string{
"-enable-pluggable-state-storage-experiment=true",
}
code := c.Run(args)
testOutput := done(t)
if code == 0 {
t.Fatalf("expected migration from cloud to fail: \n%s", testOutput.Stdout())
}
log.Printf("[TRACE] %s: second init with state store complete", t.Name())
expectedMsg := "Migrating state from HCP Terraform or Terraform Enterprise to another backend is not \nyet implemented."
if !strings.Contains(testOutput.Stderr(), expectedMsg) {
t.Fatalf("expected error message %q not found: \n%s", expectedMsg, testOutput.Stderr())
}
}
}
// Test that config-parsing errors that prevent initialising the pluggable state store are identified and returned
// before Terraform attempts to initialise the store.
//
// These errors include omitting the necessary entry in required_providers, or causing an issue with how require_providers
// is parsed. This test uses the first scenario for simplicity.
func TestInit_configErrorsImpactingStateStore(t *testing.T) {
td := t.TempDir()
t.Chdir(td)
cfg1 := `terraform {
required_providers {
foobar = {
source = "hashicorp/foobar"
}
}
state_store "test_store" {
provider "test" {} # missing from required_providers
value = "foobar"
}
}
`
if err := os.WriteFile(filepath.Join(td, "main.tf"), []byte(cfg1), 0644); err != nil {
t.Fatalf("err: %s", err)
}
ui := cli.NewMockUi()
view, done := testView(t)
initCmd := &InitCommand{
Meta: Meta{
Ui: ui,
View: view,
AllowExperimentalFeatures: true,
},
}
log.Printf("[TRACE] TestInit_configErrorsImpactingStateStore: init start")
args := []string{"-enable-pluggable-state-storage-experiment"}
code := initCmd.Run(args)
testOutput := done(t)
if code != 1 {
t.Fatalf("expected apply to fail with code 1, got code %d: \n%s", code, testOutput.All())
}
log.Printf("[TRACE] TestInit_configErrorsImpactingStateStore: init complete")
t.Logf("init output:\n%s", testOutput.Stdout())
t.Logf("init errors:\n%s", testOutput.Stderr())
expectedErrs := []string{
// Pre-amble text that's shown when a config-parsing error occurs during init.
"Error: Terraform encountered problems during initialisation, including problems with the configuration, described below.",
// This parsing error previously wouldn't be reported before initialising the backend, so
// Terraform attempted to use a state store in the missing provider.
"Error: Missing entry in required_providers",
}
for _, e := range expectedErrs {
if !strings.Contains(cleanString(testOutput.Stderr()), e) {
t.Fatalf("unexpected error, expected %q, given: %s", e, testOutput.Stderr())
}
}
}
// newMockProviderSource is a helper to succinctly construct a mock provider
// source that contains a set of packages matching the given provider versions
// that are available for installation (from temporary local files).
//
// The caller must call the returned close callback once the source is no
// longer needed, at which point it will clean up all of the temporary files
// and the packages in the source will no longer be available for installation.
//
// Provider addresses must be valid source strings, and passing only the
// provider name will be interpreted as a "default" provider under
// registry.terraform.io/hashicorp. If you need more control over the
// provider addresses, pass a full provider source string.
//
// This function also registers providers as belonging to the current platform,
// to ensure that they will be available to a provider installer operating in
// its default configuration.
//
// In case of any errors while constructing the source, this function will
// abort the current test using the given testing.T. Therefore a caller can
// assume that if this function returns then the result is valid and ready
// to use.
func newMockProviderSource(t *testing.T, availableProviderVersions map[string][]string) (source *getproviders.MockSource, close func()) {
t.Helper()
var packages []getproviders.PackageMeta
var closes []func()
close = func() {
for _, f := range closes {
f()
}
}
for source, versions := range availableProviderVersions {
addr := addrs.MustParseProviderSourceString(source)
for _, versionStr := range versions {
version, err := getproviders.ParseVersion(versionStr)
if err != nil {
close()
t.Fatalf("failed to parse %q as a version number for %q: %s", versionStr, addr.ForDisplay(), err)
}
meta, close, err := getproviders.FakeInstallablePackageMeta(addr, version, getproviders.VersionList{getproviders.MustParseVersion("5.0")}, getproviders.CurrentPlatform, "")
if err != nil {
close()
t.Fatalf("failed to prepare fake package for %s %s: %s", addr.ForDisplay(), versionStr, err)
}
closes = append(closes, close)
packages = append(packages, meta)
}
}
return getproviders.NewMockSource(packages, nil), close
}
// newMockProviderSourceViaHTTP is similar to newMockProviderSource except that the metadata (PackageMeta) for each provider
// reports that the provider is going to be accessed via HTTP
//
// Provider binaries are not available via the mock HTTP provider source. This source is sufficient only to allow Terraform
// to complete the provider installation process while believing it's installing providers over HTTP.
// This method is not sufficient to enable Terraform to use providers with those names.
//
// When using `newMockProviderSourceViaHTTP` to set a value for `(Meta).ProviderSource` in a test, also set up `testOverrides`
// in the same Meta. That way the provider source will allow the download process to complete, and when Terraform attempts to use
// those binaries it will instead use the testOverride providers.
func newMockProviderSourceViaHTTP(t *testing.T, availableProviderVersions map[string][]string, address string) (source *getproviders.MockSource) {
t.Helper()
var packages []getproviders.PackageMeta
var closes []func()
close := func() {
for _, f := range closes {
f()
}
}
for source, versions := range availableProviderVersions {
addr := addrs.MustParseProviderSourceString(source)
for _, versionStr := range versions {
version, err := getproviders.ParseVersion(versionStr)
if err != nil {
close()
t.Fatalf("failed to parse %q as a version number for %q: %s", versionStr, addr.ForDisplay(), err)
}
meta, close, err := getproviders.FakePackageMetaViaHTTP(addr, version, getproviders.VersionList{getproviders.MustParseVersion("5.0")}, getproviders.CurrentPlatform, address, "")
if err != nil {
close()
t.Fatalf("failed to prepare fake package for %s %s: %s", addr.ForDisplay(), versionStr, err)
}
closes = append(closes, close)
packages = append(packages, meta)
}
}
t.Cleanup(close)
return getproviders.NewMockSource(packages, nil)
}
// newMockProviderSourceUsingTestHttpServer is a helper that makes it easier to use newMockProviderSourceViaHTTP.
// This helper sets up a test HTTP server for use with newMockProviderSourceViaHTTP, and configures a handler that will respond when
// Terraform attempts to download provider binaries during installation. The mock source is returned ready to use and all cleanup is
// handled internally to this helper.
//
// This source is not sufficient for providers to be available to use during a test; when using this helper, also set up testOverrides in
// the same Meta to provide the actual provider implementations for use during the test.
//
// Currently this helper only allows one provider/version to be mocked. In future we could extend it to allow multiple providers/versions.
func newMockProviderSourceUsingTestHttpServer(t *testing.T, p addrs.Provider, v getproviders.Version) *getproviders.MockSource {
// Get un-started server so we can obtain the port it'll run on.
server := httptest.NewUnstartedServer(nil)
// Set up mock provider source that mocks installation via HTTP.
source := newMockProviderSourceViaHTTP(
t,
map[string][]string{
fmt.Sprintf("%s/%s", p.Namespace, p.Type): {v.String()},
},
server.Listener.Addr().String(),
)
// Supply a download location so that the installation completes ok
// while Terraform still believes it's downloading a provider via HTTP.
providerMetadata, err := source.PackageMeta(
context.Background(),
p,
v,
getproviders.CurrentPlatform,
)
if err != nil {
t.Fatalf("failed to get provider metadata: %s", err)
}
// Make Terraform believe it's downloading the provider.
// Any requests to the test server that aren't for that purpose will cause the test to fail.
server.Config = &http.Server{Handler: http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
providerLocationPath := strings.ReplaceAll(
providerMetadata.Location.String(),
"http://"+server.Listener.Addr().String(),
"",
)
// This is the URL that the init command will hit to download the provider, so we return a valid provider archive.
if r.URL.Path == providerLocationPath {
// This code returns data in the temporary file that's created by the mock provider source.
// This 'downloaded' is not used when Terraform uses the provider after the mock installation completes;
// Terraform will look for will use testOverrides in the Meta set up for this test.
//
// Although it's not used later we need to use this file (versus empty or made-up bytes) to enable installation
// logic to receive data with the correct checksum.
f, err := os.Open(providerMetadata.Filename)
if err != nil {
t.Fatalf("failed to open mock source file: %s", err)
}
defer f.Close()
archiveBytes, err := io.ReadAll(f)
if err != nil {
t.Fatalf("failed to read mock source file: %s", err)
}
w.WriteHeader(http.StatusOK)
w.Write(archiveBytes)
return
} else {
t.Fatalf("unexpected URL path: %s", r.URL.Path)
}
})}
server.Start()
t.Cleanup(server.Close)
return source
}
// installFakeProviderPackages installs a fake package for the given provider
// names (interpreted as a "default" provider address) and versions into the
// local plugin cache for the given "meta".
//
// Any test using this must also use t.TempDir and t.Chdir from the testing library
// or some similar mechanism to make sure that it isn't writing directly into a test
// fixture or source directory within the codebase.
//
// If a requested package cannot be installed for some reason, this function
// will abort the test using the given testing.T. Therefore if this function
// returns the caller can assume that the requested providers have been
// installed.
func installFakeProviderPackages(t *testing.T, meta *Meta, providerVersions map[string][]string) {
t.Helper()
cacheDir := meta.providerLocalCacheDir()
installFakeProviderPackagesElsewhere(t, cacheDir, providerVersions)
}
// installFakeProviderPackagesElsewhere is a variant of installFakeProviderPackages
// that will install packages into the given provider cache directory, rather
// than forcing the use of the local cache of the current "Meta".
func installFakeProviderPackagesElsewhere(t *testing.T, cacheDir *providercache.Dir, providerVersions map[string][]string) {
t.Helper()
// It can be hard to spot the mistake of forgetting to use t.TempDir and
// t.Chdir from the testing library before modifying the working directory,
// so we'll use a simple heuristic here to try to detect that mistake
// and make a noisy error about it instead.
wd, err := os.Getwd()
if err == nil {
wd = filepath.Clean(wd)
// If the directory we're in is named "command" or if we're under a
// directory named "testdata" then we'll assume a mistake and generate
// an error. This will cause the test to fail but won't block it from
// running.
if filepath.Base(wd) == "command" || filepath.Base(wd) == "testdata" || strings.Contains(filepath.ToSlash(wd), "/testdata/") {
t.Errorf("installFakeProviderPackage may be used only by tests that switch to a temporary working directory, e.g. using t.TempDir and t.Chdir from the testing library")
}
}
for name, versions := range providerVersions {
addr := addrs.NewDefaultProvider(name)
for _, versionStr := range versions {
version, err := getproviders.ParseVersion(versionStr)
if err != nil {
t.Fatalf("failed to parse %q as a version number for %q: %s", versionStr, name, err)
}
meta, close, err := getproviders.FakeInstallablePackageMeta(addr, version, getproviders.VersionList{getproviders.MustParseVersion("5.0")}, getproviders.CurrentPlatform, "")
// We're going to install all these fake packages before we return,
// so we don't need to preserve them afterwards.
defer close()
if err != nil {
t.Fatalf("failed to prepare fake package for %s %s: %s", name, versionStr, err)
}
_, err = cacheDir.InstallPackage(context.Background(), meta, nil)
if err != nil {
t.Fatalf("failed to install fake package for %s %s: %s", name, versionStr, err)
}
}
}
}
// expectedPackageInstallPath is a companion to installFakeProviderPackages
// that returns the path where the provider with the given name and version
// would be installed and, relatedly, where the installer will expect to
// find an already-installed version.
//
// Just as with installFakeProviderPackages, this function is a shortcut helper
// for "default-namespaced" providers as we commonly use in tests. If you need
// more control over the provider addresses, use functions of the underlying
// getproviders and providercache packages instead.
//
// The result always uses forward slashes, even on Windows, for consistency
// with how the getproviders and providercache packages build paths.
func expectedPackageInstallPath(name, version string, exe bool) string {
platform := getproviders.CurrentPlatform
command: new cache directory .terraform/providers for providers Terraform v0.10 introduced .terraform/plugins as a cache directory for automatically-installed plugins, Terraform v0.13 later reorganized the directory structure inside but retained its purpose as a cache. The local cache used to also serve as a record of specifically which packages were selected in a particular working directory, with the intent that a second run of "terraform init" would always select the same packages again. That meant that in some sense it behaved a bit like a local filesystem mirror directory, even though that wasn't its intended purpose. Due to some unfortunate miscommunications, somewhere a long the line we published some documentation that _recommended_ using the cache directory as if it were a filesystem mirror directory when working with Terraform Cloud. That was really only working as an accident of implementation details, and Terraform v0.14 is now going to break that because the source of record for the currently-selected provider versions is now the public-facing dependency lock file rather than the contents of an existing local cache directory on disk. After some consideration of how to move forward here, this commit implements a compromise that tries to avoid silently doing anything surprising while still giving useful guidance to folks who were previously using the unsupported strategy. Specifically: - The local cache directory will now be .terraform/providers rather than .terraform/plugins, because .terraform/plugins is effectively "poisoned" by the incorrect usage that we can't reliably distinguish from prior version correct usage. - The .terraform/plugins directory is now the "legacy cache directory". It is intentionally _not_ now a filesystem mirror directory, because that would risk incorrectly interpreting providers automatically installed by Terraform v0.13 as if they were a local mirror, and thus upgrades and checksum fetches from the origin registry would be blocked. - Because of the previous two points, someone who _was_ trying to use the legacy cache directory as a filesystem mirror would see installation fail for any providers they manually added to the legacy directory. To avoid leaving that user stumped as to what went wrong, there's a heuristic for the case where a non-official provider fails installation and yet we can see it in the legacy cache directory. If that heuristic matches then we'll produce a warning message hinting to move the provider under the terraform.d/plugins directory, which is a _correct_ location for "bundled" provider plugins that belong only to a single configuration (as opposed to being installed globally on a system). This does unfortunately mean that anyone who was following the incorrectly-documented pattern will now encounter an error (and the aforementioned warning hint) after upgrading to Terraform v0.14. This seems like the safest compromise because Terraform can't automatically infer the intent of files it finds in .terraform/plugins in order to decide automatically how best to handle them. The internals of the .terraform directory are always considered implementation detail for a particular Terraform version and so switching to a new directory for the _actual_ cache directory fits within our usual set of guarantees, though it's definitely non-ideal in isolation but okay when taken in the broader context of this problem, where the alternative would be silent misbehavior when upgrading.
2020-10-13 18:03:56 -04:00
baseDir := ".terraform/providers"
if exe {
p := fmt.Sprintf("registry.terraform.io/hashicorp/%s/%s/%s/terraform-provider-%s_%s", name, version, platform, name, version)
if platform.OS == "windows" {
p += ".exe"
}
return filepath.ToSlash(filepath.Join(baseDir, p))
}
return filepath.ToSlash(filepath.Join(
baseDir, fmt.Sprintf("registry.terraform.io/hashicorp/%s/%s/%s", name, version, platform),
))
}
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
// TODO: introduce pssName as argument here to aid testing migrations
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
func mockPluggableStateStorageProvider() *testing_provider.MockProvider {
// Create a mock provider to use for PSS
// Get mock provider factory to be used during init
//
// This imagines a provider called `test` that contains
// a pluggable state store implementation called `store`.
pssName := "test_store"
mock := testing_provider.MockProvider{
GetProviderSchemaResponse: &providers.GetProviderSchemaResponse{
Provider: providers.Schema{
Body: &configschema.Block{
Attributes: map[string]*configschema.Attribute{
"region": {Type: cty.String, Optional: true},
},
},
},
PSS: Add tests showing `state` subcommands being used with PSS (#37891) * test: Add E2E tests for `state list` and `state show` commands * test: Update `mockPluggableStateStorageProvider` to log a warning during tests where the values in `MockStates` aren't compatible with the `ReadStateBytesFn` default function. Make existing test set an appropriate value in `MockStates`. * test: Update `mockPluggableStateStorageProvider` helper to include a resource schema * test: Add command-level test for `state list` showing integration with pluggable state storage code. * test: Add command-level test for `state show` showing integration with pluggable state storage code. * test: Add command-level test for `state pull` showing integration with pluggable state storage code. * test: Add command-level test for `state identities` showing integration with pluggable state storage code. * test: Add command-level test for `state rm` showing integration with pluggable state storage code. * test: Add command-level test for `state mv` showing integration with pluggable state storage code. * test: Add command-level test for `state push` showing integration with pluggable state storage code. * test: Add command-level test for `state replace-provider` showing integration with pluggable state storage code. * test: Change shared test fixture to not be named after a specific command under test. This test fixure is reused across tests that need the config to define a state store but otherwise rely on the mock provider to set up the test scenario. * test: Update test to use shared test fixture * test: Remove redundant test fixture The internal/command/testdata/state-commands-state-store and internal/command/testdata/state-store-unchanged test fixtures are the same. * fix: Re-add logic for setting chunk size in the context of E2E tests using grpcwrap package This was removed, incorrectly, in https://github.com/hashicorp/terraform/pull/37899 * refactor: Let panic happen if there's incompatibility between mock returned from `mockPluggableStateStorageProvider` and the `MockStates` that's been set in the mock * test: Refactor to contain paths in reused variable, remove unnecessary .gitkeep * test: Remove unneeded test code
2025-12-09 12:24:38 -05:00
DataSources: map[string]providers.Schema{},
ResourceTypes: map[string]providers.Schema{
"test_instance": {
Body: &configschema.Block{
Attributes: map[string]*configschema.Attribute{
"input": {Type: cty.String, Optional: true},
"id": {Type: cty.String, Computed: true},
},
},
},
},
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
ListResourceTypes: map[string]providers.Schema{},
StateStores: map[string]providers.Schema{
pssName: {
Body: &configschema.Block{
Attributes: map[string]*configschema.Attribute{
"value": {
Type: cty.String,
Required: true,
},
},
},
},
},
},
}
mock.GetStatesFn = func(req providers.GetStatesRequest) providers.GetStatesResponse {
states := slices.Sorted(maps.Keys(mock.MockStates))
return providers.GetStatesResponse{
States: states,
}
}
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
mock.ConfigureStateStoreFn = func(req providers.ConfigureStateStoreRequest) providers.ConfigureStateStoreResponse {
return providers.ConfigureStateStoreResponse{
Capabilities: providers.StateStoreServerCapabilities{
ChunkSize: 1234, // arbitrary number that isn't 0
},
}
}
mock.WriteStateBytesFn = func(req providers.WriteStateBytesRequest) providers.WriteStateBytesResponse {
// Workspaces exist once the artefact representing it is written
if _, exist := mock.MockStates[req.StateId]; !exist {
// Ensure non-nil map
if mock.MockStates == nil {
mock.MockStates = make(map[string]interface{})
}
}
mock.MockStates[req.StateId] = req.Bytes
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
return providers.WriteStateBytesResponse{
Diagnostics: nil, // success
}
}
mock.ReadStateBytesFn = func(req providers.ReadStateBytesRequest) providers.ReadStateBytesResponse {
state := []byte{}
if v, exist := mock.MockStates[req.StateId]; exist {
PSS: Add tests showing `state` subcommands being used with PSS (#37891) * test: Add E2E tests for `state list` and `state show` commands * test: Update `mockPluggableStateStorageProvider` to log a warning during tests where the values in `MockStates` aren't compatible with the `ReadStateBytesFn` default function. Make existing test set an appropriate value in `MockStates`. * test: Update `mockPluggableStateStorageProvider` helper to include a resource schema * test: Add command-level test for `state list` showing integration with pluggable state storage code. * test: Add command-level test for `state show` showing integration with pluggable state storage code. * test: Add command-level test for `state pull` showing integration with pluggable state storage code. * test: Add command-level test for `state identities` showing integration with pluggable state storage code. * test: Add command-level test for `state rm` showing integration with pluggable state storage code. * test: Add command-level test for `state mv` showing integration with pluggable state storage code. * test: Add command-level test for `state push` showing integration with pluggable state storage code. * test: Add command-level test for `state replace-provider` showing integration with pluggable state storage code. * test: Change shared test fixture to not be named after a specific command under test. This test fixure is reused across tests that need the config to define a state store but otherwise rely on the mock provider to set up the test scenario. * test: Update test to use shared test fixture * test: Remove redundant test fixture The internal/command/testdata/state-commands-state-store and internal/command/testdata/state-store-unchanged test fixtures are the same. * fix: Re-add logic for setting chunk size in the context of E2E tests using grpcwrap package This was removed, incorrectly, in https://github.com/hashicorp/terraform/pull/37899 * refactor: Let panic happen if there's incompatibility between mock returned from `mockPluggableStateStorageProvider` and the `MockStates` that's been set in the mock * test: Refactor to contain paths in reused variable, remove unnecessary .gitkeep * test: Remove unneeded test code
2025-12-09 12:24:38 -05:00
state = v.([]byte) // If this panics, the mock has been set up with a bad MockStates value
PSS: Implement initialisation of new working directory (or use of `-reconfigure` flag) while using `state_store` (#37732) * Minor fixes in diagnostics This can only be done once modules have been parsed and the required providers data is available. There are multiple places where config is parsed, into either Config or Module structs, so this needs to be implemented in multiple places. * Rename test to make it specific to use of backend block in config * Update initBackend to accept whole initArgs collection * Only process --backend-config data, when setting up a `backend`, if that data isn't empty * Simplify how mock provider factories are made in tests * Update mock provider's default logic to track and manage existing workspaces * Add `ProviderSchema` method to `Pluggable` structs. This allows calling code to access the provider schema when using provider configuration data. * Add function for converting a providerreqs.Version to a hashicorp/go-version Version. This is needed for using locks when creating the backend state file. * Implement initial version of init new working directories using `stateStore_C_s`. Default to creating the default workspace if no workspaces exist. * Update test fixtures to match the hashicorp/test mock provider used in PSS tests * Allow tests to obtain locks that include `testingOverrides` providers. The `testingOverrides` field will only be set in tests, so this should not impact end users. * Add tests showing TF can initialize a working directory for the first time (and do the same when forced by -reconfigure flag). Remove replaced tests. * Add -create-default-workspace flag, to be used to disable creating the default workspace by default when -input=false (i.e for use in CI). Refactor creation of default workspace logic. Add tests. * Allow reattached providers to be used during init for PSS * Rename variable to `backendHash` so relation to `backend` is clearer * Allow `(m *Meta) Backend` to return warning diagnostics * Protect against nil testingOverrides in providerFactoriesFromLocks * Add test case seeing what happens if default workspace selected, doesn't exist, but other workspaces do exist. The consequences here are due to using `selectWorkspace` in `stateStore_C_s`, matching what's done in `backend_C_r_s`. * Address code consistency check failure on PR * Refactor use of mock in test that's experiencing EOF error... * Remove test that requires test to supply input for user prompt This test passes when run in isolation but fails when run alongside other tests, even when skipping all other tests using `testStdinPipe`. I don't think the value of this test is great enough to start changing how we test stdin input. * Allow -create-default-workspace to be used regardless of whether input is enabled or disabled * Add TF_SKIP_CREATE_DEFAULT_WORKSPACE environment variable * Responses to feedback, including making testStdinPipe helper log details of errors copying data to stdin. Note: We cannot call t.Fatal from a non-test goroutine. * Use Errorf instead * Allow backend state files to not include version data when a builtin or reattached provider is in use. * Add clarifying comment about re-attached providers when finding the matching entry in required_providers * Report that the default workspace was created to the view * Refactor: use error comparison via `errors.Is` to identify when no workspaces exist. * Move handling of TF_ENABLE_PLUGGABLE_STATE_STORAGE into init's ParseInit func. * Validate that PSS-related flags can only be used when experiments are enabled, enforce coupling of PSS-related flags when in use. * Slight rewording of output message about default workspace * Update test to assert new output about default workspace
2025-10-15 05:44:21 -04:00
}
return providers.ReadStateBytesResponse{
Bytes: state,
Diagnostics: nil, // success
}
}
return &mock
}