/*
* regexp_fixed_prefix - extract fixed prefix, if any, for a regexp
*
* The result is NULL if there is no fixed prefix, else a palloc'd string.
* If it is an exact match, not just a prefix, *exact is returned as TRUE.
*/
char *
regexp_fixed_prefix(text *text_re, bool case_insensitive, Oid collation,
bool *exact)
{
char *result;
regex_t *re;
int cflags;
int re_result;
pg_wchar *str;
size_t slen;
size_t maxlen;
char errMsg[100];
*exact = false; /* default result */
/* Compile RE */
cflags = REG_ADVANCED;
if (case_insensitive)
cflags |= REG_ICASE;
re = RE_compile_and_cache(text_re, cflags, collation);
/* Examine it to see if there's a fixed prefix */
re_result = pg_regprefix(re, &str, &slen);
switch (re_result)
{
case REG_NOMATCH:
return NULL;
case REG_PREFIX:
/* continue with wchar conversion */
break;
case REG_EXACT:
*exact = true;
/* continue with wchar conversion */
break;
default:
/* re failed??? */
CHECK_FOR_INTERRUPTS();
pg_regerror(re_result, re, errMsg, sizeof(errMsg));
ereport(ERROR,
(errcode(ERRCODE_INVALID_REGULAR_EXPRESSION),
errmsg("regular expression failed: %s", errMsg)));
break;
}
/* Convert pg_wchar result back to database encoding */
maxlen = pg_database_encoding_max_length() * slen + 1;
result = (char *) palloc(maxlen);
slen = pg_wchar2mb_with_len(str, result, slen);
Assert(slen < maxlen);
free(str);
return result;
}
/*
* RE_compile_and_execute - compile and execute a RE
*
* Returns TRUE on match, FALSE on no match
*
* text_re --- the pattern, expressed as a TEXT object
* dat --- the data to match against (need not be null-terminated)
* dat_len --- the length of the data string
* cflags --- compile options for the pattern
* nmatch, pmatch --- optional return area for match details
*
* Both pattern and data are given in the database encoding. We internally
* convert to array of pg_wchar which is what Spencer's regex package wants.
*/
static bool
RE_compile_and_execute(text *text_re, char *dat, int dat_len,
int cflags, int nmatch, regmatch_t *pmatch)
{
regex_t *re;
/* Compile RE */
re = RE_compile_and_cache(text_re, cflags);
return RE_execute(re, dat, dat_len, nmatch, pmatch);
}
/*
* textregexreplace_noopt()
* Return a string matched by a regular expression, with replacement.
*
* This version doesn't have an option argument: we default to case
* sensitive match, replace the first instance only.
*/
datum_t textregexreplace_noopt(PG_FUNC_ARGS)
{
text *s = ARG_TEXT_PP(0);
text *p = ARG_TEXT_PP(1);
text *r = ARG_TEXT_PP(2);
regex_t *re;
re = RE_compile_and_cache(p, REG_ADVANCED, PG_COLLATION());
RET_TEXT_P(replace_text_regexp(s, (void *)re, r, false));
}
/*
* textregexreplace()
* Return a string matched by a regular expression, with replacement.
*/
Datum
textregexreplace(PG_FUNCTION_ARGS)
{
text *s = PG_GETARG_TEXT_P(0);
text *p = PG_GETARG_TEXT_P(1);
text *r = PG_GETARG_TEXT_P(2);
text *opt = PG_GETARG_TEXT_P(3);
char *opt_p = VARDATA(opt);
int opt_len = (VARSIZE(opt) - VARHDRSZ);
int i;
bool glob = false;
bool ignorecase = false;
regex_t *re;
/* parse options */
for (i = 0; i < opt_len; i++)
{
switch (opt_p[i])
{
case 'i':
ignorecase = true;
break;
case 'g':
glob = true;
break;
default:
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid option of regexp_replace: %c",
opt_p[i])));
break;
}
}
if (ignorecase)
re = RE_compile_and_cache(p, regex_flavor | REG_ICASE);
else
re = RE_compile_and_cache(p, regex_flavor);
PG_RETURN_TEXT_P(replace_text_regexp(s, (void *) re, r, glob));
}
/*
* textregexreplace_noopt()
* Return a string matched by a regular expression, with replacement.
*
* This version doesn't have an option argument: we default to case
* sensitive match, replace the first instance only.
*/
Datum
textregexreplace_noopt(PG_FUNCTION_ARGS)
{
text *s = PG_GETARG_TEXT_P(0);
text *p = PG_GETARG_TEXT_P(1);
text *r = PG_GETARG_TEXT_P(2);
regex_t *re;
re = RE_compile_and_cache(p, regex_flavor);
PG_RETURN_TEXT_P(replace_text_regexp(s, (void *) re, r, false));
}
/*
* textregexsubstr()
* Return a substring matched by a regular expression.
*/
Datum
textregexsubstr(PG_FUNCTION_ARGS)
{
text *s = PG_GETARG_TEXT_PP(0);
text *p = PG_GETARG_TEXT_PP(1);
regex_t *re;
regmatch_t pmatch[2];
int so,
eo;
/* Compile RE */
re = RE_compile_and_cache(p, REG_ADVANCED, PG_GET_COLLATION());
/*
* We pass two regmatch_t structs to get info about the overall match and
* the match for the first parenthesized subexpression (if any). If there
* is a parenthesized subexpression, we return what it matched; else
* return what the whole regexp matched.
*/
if (!RE_execute(re,
VARDATA_ANY(s), VARSIZE_ANY_EXHDR(s),
2, pmatch))
PG_RETURN_NULL(); /* definitely no match */
if (re->re_nsub > 0)
{
/* has parenthesized subexpressions, use the first one */
so = pmatch[1].rm_so;
eo = pmatch[1].rm_eo;
}
else
{
/* no parenthesized subexpression, use whole match */
so = pmatch[0].rm_so;
eo = pmatch[0].rm_eo;
}
/*
* It is possible to have a match to the whole pattern but no match for a
* subexpression; for example 'foo(bar)?' is considered to match 'foo' but
* there is no subexpression match. So this extra test for match failure
* is not redundant.
*/
if (so < 0 || eo < 0)
PG_RETURN_NULL();
return DirectFunctionCall3(text_substr,
PointerGetDatum(s),
Int32GetDatum(so + 1),
Int32GetDatum(eo - so));
}
/*
* textregexreplace()
* Return a string matched by a regular expression, with replacement.
*/
datum_t textregexreplace(PG_FUNC_ARGS)
{
text *s = ARG_TEXT_PP(0);
text *p = ARG_TEXT_PP(1);
text *r = ARG_TEXT_PP(2);
text *opt = ARG_TEXT_PP(3);
regex_t *re;
pg_re_flags flags;
parse_re_flags(&flags, opt);
re = RE_compile_and_cache(p, flags.cflags, PG_COLLATION());
RET_TEXT_P(replace_text_regexp(s, (void *)re, r, flags.glob));
}
/*
* textregexreplace()
* Return a string matched by a regular expression, with replacement.
*/
Datum
textregexreplace(PG_FUNCTION_ARGS)
{
text *s = PG_GETARG_TEXT_PP(0);
text *p = PG_GETARG_TEXT_PP(1);
text *r = PG_GETARG_TEXT_PP(2);
text *opt = PG_GETARG_TEXT_PP(3);
regex_t *re;
pg_re_flags flags;
parse_re_flags(&flags, opt);
re = RE_compile_and_cache(p, flags.cflags);
PG_RETURN_TEXT_P(replace_text_regexp(s, (void *) re, r, flags.glob));
}
/*
* RE_compile_and_execute - compile and execute a RE
*
* Returns TRUE on match, FALSE on no match
*
* text_re --- the pattern, expressed as an *untoasted* TEXT object
* dat --- the data to match against (need not be null-terminated)
* dat_len --- the length of the data string
* cflags --- compile options for the pattern
* nmatch, pmatch --- optional return area for match details
*
* Both pattern and data are given in the database encoding. We internally
* convert to array of pg_wchar which is what Spencer's regex package wants.
*/
static bool
RE_compile_and_execute(text *text_re, char *dat, int dat_len,
int cflags, int nmatch, regmatch_t *pmatch)
{
pg_wchar *data;
size_t data_len;
int regexec_result;
regex_t *re;
char errMsg[100];
/* Convert data string to wide characters */
data = (pg_wchar *) palloc((dat_len + 1) * sizeof(pg_wchar));
data_len = pg_mb2wchar_with_len(dat, data, dat_len);
/* Compile RE */
re = RE_compile_and_cache(text_re, cflags);
/* Perform RE match and return result */
regexec_result = pg_regexec(re,
data,
data_len,
0,
NULL, /* no details */
nmatch,
pmatch,
0);
pfree(data);
if (regexec_result != REG_OKAY && regexec_result != REG_NOMATCH)
{
/* re failed??? */
pg_regerror(regexec_result, re, errMsg, sizeof(errMsg));
ereport(ERROR,
(errcode(ERRCODE_INVALID_REGULAR_EXPRESSION),
errmsg("regular expression failed: %s", errMsg)));
}
return (regexec_result == REG_OKAY);
}
/*
* setup_regexp_matches --- do the initial matching for regexp_matches()
* or regexp_split()
*
* To avoid having to re-find the compiled pattern on each call, we do
* all the matching in one swoop. The returned regexp_matches_ctx contains
* the locations of all the substrings matching the pattern.
*
* The three bool parameters have only two patterns (one for each caller)
* but it seems clearer to distinguish the functionality this way than to
* key it all off one "is_split" flag.
*/
static regexp_matches_ctx *
setup_regexp_matches(text *orig_str, text *pattern, text *flags,
Oid collation,
bool force_glob, bool use_subpatterns,
bool ignore_degenerate)
{
regexp_matches_ctx *matchctx = palloc0(sizeof(regexp_matches_ctx));
int orig_len;
pg_wchar *wide_str;
int wide_len;
pg_re_flags re_flags;
regex_t *cpattern;
regmatch_t *pmatch;
int pmatch_len;
int array_len;
int array_idx;
int prev_match_end;
int start_search;
/* save original string --- we'll extract result substrings from it */
matchctx->orig_str = orig_str;
/* convert string to pg_wchar form for matching */
orig_len = VARSIZE_ANY_EXHDR(orig_str);
wide_str = (pg_wchar *) palloc(sizeof(pg_wchar) * (orig_len + 1));
wide_len = pg_mb2wchar_with_len(VARDATA_ANY(orig_str), wide_str, orig_len);
/* determine options */
parse_re_flags(&re_flags, flags);
if (force_glob)
{
/* user mustn't specify 'g' for regexp_split */
if (re_flags.glob)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("regexp_split does not support the global option")));
/* but we find all the matches anyway */
re_flags.glob = true;
}
/* set up the compiled pattern */
cpattern = RE_compile_and_cache(pattern, re_flags.cflags, collation);
/* do we want to remember subpatterns? */
if (use_subpatterns && cpattern->re_nsub > 0)
{
matchctx->npatterns = cpattern->re_nsub;
pmatch_len = cpattern->re_nsub + 1;
}
else
{
use_subpatterns = false;
matchctx->npatterns = 1;
pmatch_len = 1;
}
/* temporary output space for RE package */
pmatch = palloc(sizeof(regmatch_t) * pmatch_len);
/* the real output space (grown dynamically if needed) */
array_len = re_flags.glob ? 256 : 32;
matchctx->match_locs = (int *) palloc(sizeof(int) * array_len);
array_idx = 0;
/* search for the pattern, perhaps repeatedly */
prev_match_end = 0;
start_search = 0;
while (RE_wchar_execute(cpattern, wide_str, wide_len, start_search,
pmatch_len, pmatch))
{
/*
* If requested, ignore degenerate matches, which are zero-length
* matches occurring at the start or end of a string or just after a
* previous match.
*/
if (!ignore_degenerate ||
(pmatch[0].rm_so < wide_len &&
pmatch[0].rm_eo > prev_match_end))
{
/* enlarge output space if needed */
while (array_idx + matchctx->npatterns * 2 > array_len)
{
array_len *= 2;
matchctx->match_locs = (int *) repalloc(matchctx->match_locs,
sizeof(int) * array_len);
}
/* save this match's locations */
if (use_subpatterns)
{
int i;
for (i = 1; i <= matchctx->npatterns; i++)
{
matchctx->match_locs[array_idx++] = pmatch[i].rm_so;
matchctx->match_locs[array_idx++] = pmatch[i].rm_eo;
}
}
else
{
matchctx->match_locs[array_idx++] = pmatch[0].rm_so;
matchctx->match_locs[array_idx++] = pmatch[0].rm_eo;
}
matchctx->nmatches++;
}
prev_match_end = pmatch[0].rm_eo;
/* if not glob, stop after one match */
if (!re_flags.glob)
break;
/*
* Advance search position. Normally we start the next search at the
* end of the previous match; but if the match was of zero length, we
* have to advance by one character, or we'd just find the same match
* again.
*/
start_search = prev_match_end;
if (pmatch[0].rm_so == pmatch[0].rm_eo)
start_search++;
if (start_search > wide_len)
break;
}
/* Clean up temp storage */
pfree(wide_str);
pfree(pmatch);
return matchctx;
}
/*
* setup_regexp_matches --- do the initial matching for regexp_matches()
* or regexp_split()
*
* To avoid having to re-find the compiled pattern on each call, we do
* all the matching in one swoop. The returned regexp_matches_ctx contains
* the locations of all the substrings matching the pattern.
*
* The four bool parameters have only two patterns (one for matching, one for
* splitting) but it seems clearer to distinguish the functionality this way
* than to key it all off one "is_split" flag. We don't currently assume that
* fetching_unmatched is exclusive of fetching the matched text too; if it's
* set, the conversion buffer is large enough to fetch any single matched or
* unmatched string, but not any larger substring. (In practice, when splitting
* the matches are usually small anyway, and it didn't seem worth complicating
* the code further.)
*/
static regexp_matches_ctx *
setup_regexp_matches(text *orig_str, text *pattern, text *flags,
Oid collation,
bool force_glob,
bool use_subpatterns,
bool ignore_degenerate,
bool fetching_unmatched)
{
regexp_matches_ctx *matchctx = palloc0(sizeof(regexp_matches_ctx));
int eml = pg_database_encoding_max_length();
int orig_len;
pg_wchar *wide_str;
int wide_len;
pg_re_flags re_flags;
regex_t *cpattern;
regmatch_t *pmatch;
int pmatch_len;
int array_len;
int array_idx;
int prev_match_end;
int prev_valid_match_end;
int start_search;
int maxlen = 0; /* largest fetch length in characters */
/* save original string --- we'll extract result substrings from it */
matchctx->orig_str = orig_str;
/* convert string to pg_wchar form for matching */
orig_len = VARSIZE_ANY_EXHDR(orig_str);
wide_str = (pg_wchar *) palloc(sizeof(pg_wchar) * (orig_len + 1));
wide_len = pg_mb2wchar_with_len(VARDATA_ANY(orig_str), wide_str, orig_len);
/* determine options */
parse_re_flags(&re_flags, flags);
if (force_glob)
{
/* user mustn't specify 'g' for regexp_split */
if (re_flags.glob)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("regexp_split does not support the global option")));
/* but we find all the matches anyway */
re_flags.glob = true;
}
/* set up the compiled pattern */
cpattern = RE_compile_and_cache(pattern, re_flags.cflags, collation);
/* do we want to remember subpatterns? */
if (use_subpatterns && cpattern->re_nsub > 0)
{
matchctx->npatterns = cpattern->re_nsub;
pmatch_len = cpattern->re_nsub + 1;
}
else
{
use_subpatterns = false;
matchctx->npatterns = 1;
pmatch_len = 1;
}
/* temporary output space for RE package */
pmatch = palloc(sizeof(regmatch_t) * pmatch_len);
/*
* the real output space (grown dynamically if needed)
*
* use values 2^n-1, not 2^n, so that we hit the limit at 2^28-1 rather
* than at 2^27
*/
array_len = re_flags.glob ? 255 : 31;
matchctx->match_locs = (int *) palloc(sizeof(int) * array_len);
array_idx = 0;
/* search for the pattern, perhaps repeatedly */
prev_match_end = 0;
prev_valid_match_end = 0;
start_search = 0;
while (RE_wchar_execute(cpattern, wide_str, wide_len, start_search,
pmatch_len, pmatch))
{
/*
* If requested, ignore degenerate matches, which are zero-length
* matches occurring at the start or end of a string or just after a
* previous match.
*/
if (!ignore_degenerate ||
(pmatch[0].rm_so < wide_len &&
pmatch[0].rm_eo > prev_match_end))
{
/* enlarge output space if needed */
while (array_idx + matchctx->npatterns * 2 + 1 > array_len)
{
array_len += array_len + 1; /* 2^n-1 => 2^(n+1)-1 */
if (array_len > MaxAllocSize/sizeof(int))
ereport(ERROR,
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
errmsg("too many regular expression matches")));
matchctx->match_locs = (int *) repalloc(matchctx->match_locs,
sizeof(int) * array_len);
}
/* save this match's locations */
if (use_subpatterns)
{
int i;
for (i = 1; i <= matchctx->npatterns; i++)
{
int so = pmatch[i].rm_so;
int eo = pmatch[i].rm_eo;
matchctx->match_locs[array_idx++] = so;
matchctx->match_locs[array_idx++] = eo;
if (so >= 0 && eo >= 0 && (eo - so) > maxlen)
maxlen = (eo - so);
}
}
else
{
int so = pmatch[0].rm_so;
int eo = pmatch[0].rm_eo;
matchctx->match_locs[array_idx++] = so;
matchctx->match_locs[array_idx++] = eo;
if (so >= 0 && eo >= 0 && (eo - so) > maxlen)
maxlen = (eo - so);
}
matchctx->nmatches++;
/*
* check length of unmatched portion between end of previous valid
* (nondegenerate, or degenerate but not ignored) match and start
* of current one
*/
if (fetching_unmatched &&
pmatch[0].rm_so >= 0 &&
(pmatch[0].rm_so - prev_valid_match_end) > maxlen)
maxlen = (pmatch[0].rm_so - prev_valid_match_end);
prev_valid_match_end = pmatch[0].rm_eo;
}
prev_match_end = pmatch[0].rm_eo;
/* if not glob, stop after one match */
if (!re_flags.glob)
break;
/*
* Advance search position. Normally we start the next search at the
* end of the previous match; but if the match was of zero length, we
* have to advance by one character, or we'd just find the same match
* again.
*/
start_search = prev_match_end;
if (pmatch[0].rm_so == pmatch[0].rm_eo)
start_search++;
if (start_search > wide_len)
break;
}
/*
* check length of unmatched portion between end of last match and end of
* input string
*/
if (fetching_unmatched &&
(wide_len - prev_valid_match_end) > maxlen)
maxlen = (wide_len - prev_valid_match_end);
/*
* Keep a note of the end position of the string for the benefit of
* splitting code.
*/
matchctx->match_locs[array_idx] = wide_len;
if (eml > 1)
{
int64 maxsiz = eml * (int64) maxlen;
int conv_bufsiz;
/*
* Make the conversion buffer large enough for any substring of
* interest.
*
* Worst case: assume we need the maximum size (maxlen*eml), but take
* advantage of the fact that the original string length in bytes is an
* upper bound on the byte length of any fetched substring (and we know
* that len+1 is safe to allocate because the varlena header is longer
* than 1 byte).
*/
if (maxsiz > orig_len)
conv_bufsiz = orig_len + 1;
else
conv_bufsiz = maxsiz + 1; /* safe since maxsiz < 2^30 */
matchctx->conv_buf = palloc(conv_bufsiz);
matchctx->conv_bufsiz = conv_bufsiz;
matchctx->wide_str = wide_str;
}
else
{
/* No need to keep the wide string if we're in a single-byte charset. */
pfree(wide_str);
matchctx->wide_str = NULL;
matchctx->conv_buf = NULL;
matchctx->conv_bufsiz = 0;
}
/* Clean up temp storage */
pfree(pmatch);
return matchctx;
}
