mirror of
https://github.com/postgres/postgres.git
synced 2026-03-22 18:33:19 -04:00
Commit 84d5efa7e3 missed some multibyte issues caused by short-circuit
logic in the callers. The callers assumed that if the predicate string
is longer than the label string, then it couldn't possibly be a match,
but it can be when using case-insensitive matching (LVAR_INCASE) if
casefolding changes the byte length.
Fix by refactoring to get rid of the short-circuit logic as well as
the function pointer, and consolidate the logic in a replacement
function ltree_label_match().
Discussion: https://postgr.es/m/02c6ef6cf56a5013ede61ad03c7a26affd27d449.camel@j-davis.com
Backpatch-through: 14
329 lines
7.4 KiB
C
329 lines
7.4 KiB
C
/*
|
|
* op function for ltree and lquery
|
|
* Teodor Sigaev <teodor@stack.net>
|
|
* contrib/ltree/lquery_op.c
|
|
*/
|
|
#include "postgres.h"
|
|
|
|
#include <ctype.h>
|
|
|
|
#include "catalog/pg_collation.h"
|
|
#include "ltree.h"
|
|
#include "miscadmin.h"
|
|
#include "utils/array.h"
|
|
#include "utils/formatting.h"
|
|
|
|
PG_FUNCTION_INFO_V1(ltq_regex);
|
|
PG_FUNCTION_INFO_V1(ltq_rregex);
|
|
|
|
PG_FUNCTION_INFO_V1(lt_q_regex);
|
|
PG_FUNCTION_INFO_V1(lt_q_rregex);
|
|
|
|
#define NEXTVAL(x) ( (lquery*)( (char*)(x) + INTALIGN( VARSIZE(x) ) ) )
|
|
|
|
static char *
|
|
getlexeme(char *start, char *end, int *len)
|
|
{
|
|
char *ptr;
|
|
|
|
while (start < end && t_iseq(start, '_'))
|
|
start += pg_mblen_range(start, end);
|
|
|
|
ptr = start;
|
|
if (ptr >= end)
|
|
return NULL;
|
|
|
|
while (ptr < end && !t_iseq(ptr, '_'))
|
|
ptr += pg_mblen_range(ptr, end);
|
|
|
|
*len = ptr - start;
|
|
return start;
|
|
}
|
|
|
|
bool
|
|
compare_subnode(ltree_level *t, char *qn, int len, bool prefix, bool ci)
|
|
{
|
|
char *endt = t->name + t->len;
|
|
char *endq = qn + len;
|
|
char *tn;
|
|
int lent,
|
|
lenq;
|
|
bool isok;
|
|
|
|
while ((qn = getlexeme(qn, endq, &lenq)) != NULL)
|
|
{
|
|
tn = t->name;
|
|
isok = false;
|
|
while ((tn = getlexeme(tn, endt, &lent)) != NULL)
|
|
{
|
|
if (ltree_label_match(qn, lenq, tn, lent, prefix, ci))
|
|
{
|
|
isok = true;
|
|
break;
|
|
}
|
|
tn += lent;
|
|
}
|
|
|
|
if (!isok)
|
|
return false;
|
|
qn += lenq;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
/*
|
|
* Check if the label matches the predicate string. If 'prefix' is true, then
|
|
* the predicate string is treated as a prefix. If 'ci' is true, then the
|
|
* predicate string is case-insensitive (and locale-aware).
|
|
*/
|
|
bool
|
|
ltree_label_match(const char *pred, size_t pred_len, const char *label,
|
|
size_t label_len, bool prefix, bool ci)
|
|
{
|
|
static pg_locale_t locale = NULL;
|
|
char *fpred; /* casefolded predicate */
|
|
size_t fpred_len = pred_len;
|
|
char *flabel; /* casefolded label */
|
|
size_t flabel_len = label_len;
|
|
size_t len;
|
|
bool res;
|
|
|
|
/* fast path for binary match or binary prefix match */
|
|
if ((pred_len == label_len || (prefix && pred_len < label_len)) &&
|
|
strncmp(pred, label, pred_len) == 0)
|
|
return true;
|
|
else if (!ci)
|
|
return false;
|
|
|
|
/*
|
|
* Slow path for case-insensitive comparison: case fold and then compare.
|
|
* This path is necessary even if pred_len > label_len, because the byte
|
|
* lengths may change after casefolding.
|
|
*/
|
|
if (!locale)
|
|
locale = pg_database_locale();
|
|
|
|
fpred = palloc(fpred_len + 1);
|
|
len = pg_strfold(fpred, fpred_len + 1, pred, pred_len, locale);
|
|
if (len > fpred_len)
|
|
{
|
|
/* grow buffer if needed and retry */
|
|
fpred_len = len;
|
|
fpred = repalloc(fpred, fpred_len + 1);
|
|
len = pg_strfold(fpred, fpred_len + 1, pred, pred_len, locale);
|
|
}
|
|
Assert(len <= fpred_len);
|
|
fpred_len = len;
|
|
|
|
flabel = palloc(flabel_len + 1);
|
|
len = pg_strfold(flabel, flabel_len + 1, label, label_len, locale);
|
|
if (len > flabel_len)
|
|
{
|
|
/* grow buffer if needed and retry */
|
|
flabel_len = len;
|
|
flabel = repalloc(flabel, flabel_len + 1);
|
|
len = pg_strfold(flabel, flabel_len + 1, label, label_len, locale);
|
|
}
|
|
Assert(len <= flabel_len);
|
|
flabel_len = len;
|
|
|
|
if ((fpred_len == flabel_len || (prefix && fpred_len < flabel_len)) &&
|
|
strncmp(fpred, flabel, fpred_len) == 0)
|
|
res = true;
|
|
else
|
|
res = false;
|
|
|
|
pfree(fpred);
|
|
pfree(flabel);
|
|
|
|
return res;
|
|
}
|
|
|
|
/*
|
|
* See if an lquery_level matches an ltree_level
|
|
*
|
|
* This accounts for all flags including LQL_NOT, but does not
|
|
* consider repetition counts.
|
|
*/
|
|
static bool
|
|
checkLevel(lquery_level *curq, ltree_level *curt)
|
|
{
|
|
lquery_variant *curvar = LQL_FIRST(curq);
|
|
bool success;
|
|
|
|
success = (curq->flag & LQL_NOT) ? false : true;
|
|
|
|
/* numvar == 0 means '*' which matches anything */
|
|
if (curq->numvar == 0)
|
|
return success;
|
|
|
|
for (int i = 0; i < curq->numvar; i++)
|
|
{
|
|
bool prefix = (curvar->flag & LVAR_ANYEND);
|
|
bool ci = (curvar->flag & LVAR_INCASE);
|
|
|
|
if (curvar->flag & LVAR_SUBLEXEME)
|
|
{
|
|
if (compare_subnode(curt, curvar->name, curvar->len, prefix, ci))
|
|
return success;
|
|
}
|
|
else if (ltree_label_match(curvar->name, curvar->len, curt->name,
|
|
curt->len, prefix, ci))
|
|
return success;
|
|
|
|
curvar = LVAR_NEXT(curvar);
|
|
}
|
|
return !success;
|
|
}
|
|
|
|
/*
|
|
* Try to match an lquery (of qlen items) to an ltree (of tlen items)
|
|
*/
|
|
static bool
|
|
checkCond(lquery_level *curq, int qlen,
|
|
ltree_level *curt, int tlen)
|
|
{
|
|
/* Since this function recurses, it could be driven to stack overflow */
|
|
check_stack_depth();
|
|
|
|
/* Pathological patterns could take awhile, too */
|
|
CHECK_FOR_INTERRUPTS();
|
|
|
|
/* Loop while we have query items to consider */
|
|
while (qlen > 0)
|
|
{
|
|
int low,
|
|
high;
|
|
lquery_level *nextq;
|
|
|
|
/*
|
|
* Get min and max repetition counts for this query item, dealing with
|
|
* the backwards-compatibility hack that the low/high fields aren't
|
|
* meaningful for non-'*' items unless LQL_COUNT is set.
|
|
*/
|
|
if ((curq->flag & LQL_COUNT) || curq->numvar == 0)
|
|
low = curq->low, high = curq->high;
|
|
else
|
|
low = high = 1;
|
|
|
|
/*
|
|
* We may limit "high" to the remaining text length; this avoids
|
|
* separate tests below.
|
|
*/
|
|
if (high > tlen)
|
|
high = tlen;
|
|
|
|
/* Fail if a match of required number of items is impossible */
|
|
if (high < low)
|
|
return false;
|
|
|
|
/*
|
|
* Recursively check the rest of the pattern against each possible
|
|
* start point following some of this item's match(es).
|
|
*/
|
|
nextq = LQL_NEXT(curq);
|
|
qlen--;
|
|
|
|
for (int matchcnt = 0; matchcnt < high; matchcnt++)
|
|
{
|
|
/*
|
|
* If we've consumed an acceptable number of matches of this item,
|
|
* and the rest of the pattern matches beginning here, we're good.
|
|
*/
|
|
if (matchcnt >= low && checkCond(nextq, qlen, curt, tlen))
|
|
return true;
|
|
|
|
/*
|
|
* Otherwise, try to match one more text item to this query item.
|
|
*/
|
|
if (!checkLevel(curq, curt))
|
|
return false;
|
|
|
|
curt = LEVEL_NEXT(curt);
|
|
tlen--;
|
|
}
|
|
|
|
/*
|
|
* Once we've consumed "high" matches, we can succeed only if the rest
|
|
* of the pattern matches beginning here. Loop around (if you prefer,
|
|
* think of this as tail recursion).
|
|
*/
|
|
curq = nextq;
|
|
}
|
|
|
|
/*
|
|
* Once we're out of query items, we match only if there's no remaining
|
|
* text either.
|
|
*/
|
|
return (tlen == 0);
|
|
}
|
|
|
|
Datum
|
|
ltq_regex(PG_FUNCTION_ARGS)
|
|
{
|
|
ltree *tree = PG_GETARG_LTREE_P(0);
|
|
lquery *query = PG_GETARG_LQUERY_P(1);
|
|
bool res;
|
|
|
|
res = checkCond(LQUERY_FIRST(query), query->numlevel,
|
|
LTREE_FIRST(tree), tree->numlevel);
|
|
|
|
PG_FREE_IF_COPY(tree, 0);
|
|
PG_FREE_IF_COPY(query, 1);
|
|
PG_RETURN_BOOL(res);
|
|
}
|
|
|
|
Datum
|
|
ltq_rregex(PG_FUNCTION_ARGS)
|
|
{
|
|
PG_RETURN_DATUM(DirectFunctionCall2(ltq_regex,
|
|
PG_GETARG_DATUM(1),
|
|
PG_GETARG_DATUM(0)
|
|
));
|
|
}
|
|
|
|
Datum
|
|
lt_q_regex(PG_FUNCTION_ARGS)
|
|
{
|
|
ltree *tree = PG_GETARG_LTREE_P(0);
|
|
ArrayType *_query = PG_GETARG_ARRAYTYPE_P(1);
|
|
lquery *query = (lquery *) ARR_DATA_PTR(_query);
|
|
bool res = false;
|
|
int num = ArrayGetNItems(ARR_NDIM(_query), ARR_DIMS(_query));
|
|
|
|
if (ARR_NDIM(_query) > 1)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
|
|
errmsg("array must be one-dimensional")));
|
|
if (array_contains_nulls(_query))
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
|
|
errmsg("array must not contain nulls")));
|
|
|
|
while (num > 0)
|
|
{
|
|
if (DatumGetBool(DirectFunctionCall2(ltq_regex,
|
|
PointerGetDatum(tree), PointerGetDatum(query))))
|
|
{
|
|
|
|
res = true;
|
|
break;
|
|
}
|
|
num--;
|
|
query = NEXTVAL(query);
|
|
}
|
|
|
|
PG_FREE_IF_COPY(tree, 0);
|
|
PG_FREE_IF_COPY(_query, 1);
|
|
PG_RETURN_BOOL(res);
|
|
}
|
|
|
|
Datum
|
|
lt_q_rregex(PG_FUNCTION_ARGS)
|
|
{
|
|
PG_RETURN_DATUM(DirectFunctionCall2(lt_q_regex,
|
|
PG_GETARG_DATUM(1),
|
|
PG_GETARG_DATUM(0)
|
|
));
|
|
}
|