#define SCF_TRIE_RESTUDY 0x4000 /* Do restudy? */
#define SCF_SEEN_ACCEPT 0x8000
-#define UTF (RExC_utf8 != 0)
-#define LOC ((RExC_flags & RXf_PMf_LOCALE) != 0)
-#define FOLD ((RExC_flags & RXf_PMf_FOLD) != 0)
+#define UTF cBOOL(RExC_utf8)
+#define LOC cBOOL(RExC_flags & RXf_PMf_LOCALE)
+#define UNI_SEMANTICS cBOOL(RExC_flags & RXf_PMf_UNICODE)
+#define FOLD cBOOL(RExC_flags & RXf_PMf_FOLD)
#define OOB_UNICODE 12345678
#define OOB_NAMEDCLASS -1
#if PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS
#define EXPERIMENTAL_INPLACESCAN
-#endif /*RE_TRACK_PATTERN_OFFSETS*/
+#endif /*PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS*/
#define DEBUG_STUDYDATA(str,data,depth) \
DEBUG_OPTIMISE_MORE_r(if(data){ \
Dumps the final compressed table form of the trie to Perl_debug_log.
Used for debugging make_trie().
*/
-
+
STATIC void
S_dump_trie(pTHX_ const struct _reg_trie_data *trie, HV *widecharmap,
AV *revcharmap, U32 depth)
if (trie->jump)
trie->jump[0] = (U16)(nextbranch - convert);
- /* XXXX */
- if ( !trie->states[trie->startstate].wordnum && trie->bitmap &&
- ( (char *)jumper - (char *)convert) >= (int)sizeof(struct regnode_charclass) )
+ /* If the start state is not accepting (meaning there is no empty string/NOTHING)
+ * and there is a bitmap
+ * and the first "jump target" node we found leaves enough room
+ * then convert the TRIE node into a TRIEC node, with the bitmap
+ * embedded inline in the opcode - this is hypothetically faster.
+ */
+ if ( !trie->states[trie->startstate].wordnum
+ && trie->bitmap
+ && ( (char *)jumper - (char *)convert) >= (int)sizeof(struct regnode_charclass) )
{
OP( convert ) = TRIEC;
Copy(trie->bitmap, ((struct regnode_charclass *)convert)->bitmap, ANYOF_BITMAP_SIZE, char);
/* These are the cases when once a subexpression
fails at a particular position, it cannot succeed
even after backtracking at the enclosing scope.
-
+
XXXX what if minimal match and we are at the
initial run of {n,m}? */
if ((mincount != maxcount - 1) && (maxcount != REG_INFTY))
#if 0
while ( nxt1 && (OP(nxt1) != WHILEM)) {
regnode *nnxt = regnext(nxt1);
-
if (nnxt == nxt) {
if (reg_off_by_arg[OP(nxt1)])
ARG_SET(nxt1, nxt2 - nxt1);
if (UTF)
old = utf8_hop((U8*)s, old) - (U8*)s;
-
l -= old;
/* Get the added string: */
last_str = newSVpvn_utf8(s + old, l, UTF);
if (flags & SCF_DO_STCLASS_AND) {
for (value = 0; value < 256; value++)
if (!is_VERTWS_cp(value))
- ANYOF_BITMAP_CLEAR(data->start_class, value);
- }
- else {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ else {
for (value = 0; value < 256; value++)
if (is_VERTWS_cp(value))
- ANYOF_BITMAP_SET(data->start_class, value);
- }
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
if (flags & SCF_DO_STCLASS_OR)
cl_and(data->start_class, and_withp);
flags &= ~SCF_DO_STCLASS;
data->pos_delta += 1;
data->longest = &(data->longest_float);
}
-
}
else if (OP(scan) == FOLDCHAR) {
int d = ARG(scan)==0xDF ? 1 : 2;
if (flags & SCF_DO_STCLASS_AND) {
if (!(data->start_class->flags & ANYOF_LOCALE)) {
ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NALNUM);
- for (value = 0; value < 256; value++)
- if (!isALNUM(value))
- ANYOF_BITMAP_CLEAR(data->start_class, value);
+ if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (!isWORDCHAR_L1(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ } else {
+ for (value = 0; value < 256; value++) {
+ if (!isALNUM(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ }
}
}
else {
if (data->start_class->flags & ANYOF_LOCALE)
ANYOF_CLASS_SET(data->start_class,ANYOF_ALNUM);
- else {
- for (value = 0; value < 256; value++)
- if (isALNUM(value))
- ANYOF_BITMAP_SET(data->start_class, value);
- }
+ else if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (isWORDCHAR_L1(value)) {
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ }
+ } else {
+ for (value = 0; value < 256; value++) {
+ if (isALNUM(value)) {
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ }
+ }
}
break;
case ALNUML:
if (flags & SCF_DO_STCLASS_AND) {
if (!(data->start_class->flags & ANYOF_LOCALE)) {
ANYOF_CLASS_CLEAR(data->start_class,ANYOF_ALNUM);
- for (value = 0; value < 256; value++)
- if (isALNUM(value))
- ANYOF_BITMAP_CLEAR(data->start_class, value);
+ if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (isWORDCHAR_L1(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ } else {
+ for (value = 0; value < 256; value++) {
+ if (isALNUM(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ }
}
}
else {
else {
for (value = 0; value < 256; value++)
if (!isALNUM(value))
- ANYOF_BITMAP_SET(data->start_class, value);
+ ANYOF_BITMAP_SET(data->start_class, value);
}
}
break;
if (flags & SCF_DO_STCLASS_AND) {
if (!(data->start_class->flags & ANYOF_LOCALE)) {
ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NSPACE);
- for (value = 0; value < 256; value++)
- if (!isSPACE(value))
- ANYOF_BITMAP_CLEAR(data->start_class, value);
+ if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (!isSPACE_L1(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ } else {
+ for (value = 0; value < 256; value++) {
+ if (!isSPACE(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ }
}
}
else {
- if (data->start_class->flags & ANYOF_LOCALE)
+ if (data->start_class->flags & ANYOF_LOCALE) {
ANYOF_CLASS_SET(data->start_class,ANYOF_SPACE);
- else {
- for (value = 0; value < 256; value++)
- if (isSPACE(value))
- ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ else if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (isSPACE_L1(value)) {
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ }
+ } else {
+ for (value = 0; value < 256; value++) {
+ if (isSPACE(value)) {
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ }
}
}
break;
if (flags & SCF_DO_STCLASS_AND) {
if (!(data->start_class->flags & ANYOF_LOCALE)) {
ANYOF_CLASS_CLEAR(data->start_class,ANYOF_SPACE);
- for (value = 0; value < 256; value++)
- if (isSPACE(value))
- ANYOF_BITMAP_CLEAR(data->start_class, value);
+ if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (isSPACE_L1(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ } else {
+ for (value = 0; value < 256; value++) {
+ if (isSPACE(value)) {
+ ANYOF_BITMAP_CLEAR(data->start_class, value);
+ }
+ }
+ }
}
}
else {
if (data->start_class->flags & ANYOF_LOCALE)
ANYOF_CLASS_SET(data->start_class,ANYOF_NSPACE);
- else {
- for (value = 0; value < 256; value++)
- if (!isSPACE(value))
- ANYOF_BITMAP_SET(data->start_class, value);
- }
+ else if (FLAGS(scan) & USE_UNI) {
+ for (value = 0; value < 256; value++) {
+ if (!isSPACE_L1(value)) {
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ }
+ }
+ else {
+ for (value = 0; value < 256; value++) {
+ if (!isSPACE(value)) {
+ ANYOF_BITMAP_SET(data->start_class, value);
+ }
+ }
+ }
}
break;
case NSPACEL:
else {
for (value = 0; value < 256; value++)
if (isDIGIT(value))
- ANYOF_BITMAP_SET(data->start_class, value);
+ ANYOF_BITMAP_SET(data->start_class, value);
}
}
break;
else {
for (value = 0; value < 256; value++)
if (!isDIGIT(value))
- ANYOF_BITMAP_SET(data->start_class, value);
+ ANYOF_BITMAP_SET(data->start_class, value);
}
}
break;
struct regexp *r;
register regexp_internal *ri;
STRLEN plen;
- char *exp = SvPV(pattern, plen);
- char* xend = exp + plen;
+ char *exp;
+ char* xend;
regnode *scan;
I32 flags;
I32 minlen = 0;
+
+ /* these are all flags - maybe they should be turned
+ * into a single int with different bit masks */
+ I32 sawlookahead = 0;
I32 sawplus = 0;
I32 sawopen = 0;
+
U8 jump_ret = 0;
dJMPENV;
scan_data_t data;
RExC_state_t RExC_state;
RExC_state_t * const pRExC_state = &RExC_state;
#ifdef TRIE_STUDY_OPT
- int restudied= 0;
+ int restudied;
RExC_state_t copyRExC_state;
#endif
GET_RE_DEBUG_FLAGS_DECL;
RExC_utf8 = RExC_orig_utf8 = SvUTF8(pattern);
- DEBUG_COMPILE_r({
- SV *dsv= sv_newmortal();
- RE_PV_QUOTED_DECL(s, RExC_utf8,
- dsv, exp, plen, 60);
- PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
- PL_colors[4],PL_colors[5],s);
- });
/* Longjmp back to here if have to switch in midstream to utf8 */
if (! RExC_orig_utf8) {
JMPENV_PUSH(jump_ret);
}
- if (jump_ret != 0) {
+ if (jump_ret == 0) { /* First time through */
+ exp = SvPV(pattern, plen);
+ xend = exp + plen;
+
+ DEBUG_COMPILE_r({
+ SV *dsv= sv_newmortal();
+ RE_PV_QUOTED_DECL(s, RExC_utf8,
+ dsv, exp, plen, 60);
+ PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
+ PL_colors[4],PL_colors[5],s);
+ });
+ }
+ else { /* longjumped back */
STRLEN len = plen;
- /* Here, we longjmped back. If the cause was other than changing to
- * utf8, pop our own setjmp, and longjmp to the correct handler */
+ /* If the cause for the longjmp was other than changing to utf8, pop
+ * our own setjmp, and longjmp to the correct handler */
if (jump_ret != UTF8_LONGJMP) {
JMPENV_POP;
JMPENV_JUMP(jump_ret);
SAVEFREEPV(exp);
}
+#ifdef TRIE_STUDY_OPT
+ restudied = 0;
+#endif
+
RExC_precomp = exp;
RExC_flags = pm_flags;
RExC_sawback = 0;
r->extflags = pm_flags;
{
bool has_p = ((r->extflags & RXf_PMf_KEEPCOPY) == RXf_PMf_KEEPCOPY);
- bool has_minus = ((r->extflags & RXf_PMf_STD_PMMOD) != RXf_PMf_STD_PMMOD);
+ bool has_charset = cBOOL(r->extflags & (RXf_PMf_LOCALE|RXf_PMf_UNICODE));
+
+ /* The caret is output if there are any defaults: if not all the STD
+ * flags are set, or if no character set specifier is needed */
+ bool has_default =
+ (((r->extflags & RXf_PMf_STD_PMMOD) != RXf_PMf_STD_PMMOD)
+ || ! has_charset);
bool has_runon = ((RExC_seen & REG_SEEN_RUN_ON_COMMENT)==REG_SEEN_RUN_ON_COMMENT);
U16 reganch = (U16)((r->extflags & RXf_PMf_STD_PMMOD)
>> RXf_PMf_STD_PMMOD_SHIFT);
const char *fptr = STD_PAT_MODS; /*"msix"*/
char *p;
/* Allocate for the worst case, which is all the std flags are turned
- * on, but this means no caret. We never output a minus, as all those
- * are defaults, so are covered by the caret */
+ * on. If more precision is desired, we could do a population count of
+ * the flags set. This could be done with a small lookup table, or by
+ * shifting, masking and adding, or even, when available, assembly
+ * language for a machine-language population count.
+ * We never output a minus, as all those are defaults, so are
+ * covered by the caret */
const STRLEN wraplen = plen + has_p + has_runon
+ + has_default /* If needs a caret */
+ + has_charset /* If needs a character set specifier */
+ (sizeof(STD_PAT_MODS) - 1)
+ (sizeof("(?:)") - 1);
- p = sv_grow(MUTABLE_SV(rx), wraplen + 1);
+ p = sv_grow(MUTABLE_SV(rx), wraplen + 1); /* +1 for the ending NUL */
SvPOK_on(rx);
SvFLAGS(rx) |= SvUTF8(pattern);
*p++='('; *p++='?';
- if (has_minus) { /* If a default, cover it using the caret */
+
+ /* If a default, cover it using the caret */
+ if (has_default) {
*p++= DEFAULT_PAT_MOD;
}
+ if (has_charset) {
+ if (r->extflags & RXf_PMf_LOCALE) {
+ *p++ = LOCALE_PAT_MOD;
+ } else {
+ *p++ = UNICODE_PAT_MOD;
+ }
+ }
if (has_p)
*p++ = KEEPCOPY_PAT_MOD; /*'p'*/
{
}
reStudy:
- r->minlen = minlen = sawplus = sawopen = 0;
+ r->minlen = minlen = sawlookahead = sawplus = sawopen = 0;
Zero(r->substrs, 1, struct reg_substr_data);
#ifdef TRIE_STUDY_OPT
I32 last_close = 0; /* pointed to by data */
regnode *first= scan;
regnode *first_next= regnext(first);
-
/*
* Skip introductions and multiplicators >= 1
* so that we can extract the 'meat' of the pattern that must
/* An OR of *one* alternative - should not happen now. */
(OP(first) == BRANCH && OP(first_next) != BRANCH) ||
/* for now we can't handle lookbehind IFMATCH*/
- (OP(first) == IFMATCH && !first->flags) ||
+ (OP(first) == IFMATCH && !first->flags && (sawlookahead = 1)) ||
(OP(first) == PLUS) ||
(OP(first) == MINMOD) ||
/* An {n,m} with n>0 */
first = NEXTOPER(first);
goto again;
}
- if (sawplus && (!sawopen || !RExC_sawback)
+ if (sawplus && !sawlookahead && (!sawopen || !RExC_sawback)
&& !(RExC_seen & REG_SEEN_EVAL)) /* May examine pos and $& */
/* x+ must match at the 1st pos of run of x's */
r->intflags |= PREGf_SKIP;
that follow */
has_use_defaults = TRUE;
STD_PMMOD_FLAGS_CLEAR(&RExC_flags);
+ RExC_flags &= ~(RXf_PMf_LOCALE|RXf_PMf_UNICODE);
goto parse_flags;
default:
--RExC_parse;
{
U32 posflags = 0, negflags = 0;
U32 *flagsp = &posflags;
+ bool has_charset_modifier = 0;
while (*RExC_parse) {
/* && strchr("iogcmsx", *RExC_parse) */
and must be globally applied -- japhy */
switch (*RExC_parse) {
CASE_STD_PMMOD_FLAGS_PARSE_SET(flagsp);
+ case LOCALE_PAT_MOD:
+ if (has_charset_modifier || flagsp == &negflags) {
+ goto fail_modifiers;
+ }
+ posflags |= RXf_PMf_LOCALE;
+ negflags |= RXf_PMf_UNICODE;
+ has_charset_modifier = 1;
+ break;
+ case UNICODE_PAT_MOD:
+ if (has_charset_modifier || flagsp == &negflags) {
+ goto fail_modifiers;
+ }
+ posflags |= RXf_PMf_UNICODE;
+ negflags |= RXf_PMf_LOCALE;
+ has_charset_modifier = 1;
+ break;
+ case DUAL_PAT_MOD:
+ if (has_use_defaults
+ || has_charset_modifier
+ || flagsp == &negflags)
+ {
+ goto fail_modifiers;
+ }
+ negflags |= (RXf_PMf_LOCALE|RXf_PMf_UNICODE);
+ has_charset_modifier = 1;
+ break;
case ONCE_PAT_MOD: /* 'o' */
case GLOBAL_PAT_MOD: /* 'g' */
if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
* if there is a minus, it means will be trying to
* re-specify a default which is an error */
if (has_use_defaults || flagsp == &negflags) {
+ fail_modifiers:
RExC_parse++;
vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
/*NOTREACHED*/
*flagp |= HASWIDTH;
goto finish_meta_pat;
case 'w':
- ret = reg_node(pRExC_state, (U8)(LOC ? ALNUML : ALNUM));
+ if (LOC) {
+ ret = reg_node(pRExC_state, (U8)(ALNUML));
+ } else {
+ ret = reg_node(pRExC_state, (U8)(ALNUM));
+ FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0;
+ }
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'W':
- ret = reg_node(pRExC_state, (U8)(LOC ? NALNUML : NALNUM));
+ if (LOC) {
+ ret = reg_node(pRExC_state, (U8)(NALNUML));
+ } else {
+ ret = reg_node(pRExC_state, (U8)(NALNUM));
+ FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0;
+ }
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'b':
RExC_seen_zerolen++;
RExC_seen |= REG_SEEN_LOOKBEHIND;
- ret = reg_node(pRExC_state, (U8)(LOC ? BOUNDL : BOUND));
+ if (LOC) {
+ ret = reg_node(pRExC_state, (U8)(BOUNDL));
+ } else {
+ ret = reg_node(pRExC_state, (U8)(BOUND));
+ FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0;
+ }
*flagp |= SIMPLE;
goto finish_meta_pat;
case 'B':
RExC_seen_zerolen++;
RExC_seen |= REG_SEEN_LOOKBEHIND;
- ret = reg_node(pRExC_state, (U8)(LOC ? NBOUNDL : NBOUND));
+ if (LOC) {
+ ret = reg_node(pRExC_state, (U8)(NBOUNDL));
+ } else {
+ ret = reg_node(pRExC_state, (U8)(NBOUND));
+ FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0;
+ }
*flagp |= SIMPLE;
goto finish_meta_pat;
case 's':
- ret = reg_node(pRExC_state, (U8)(LOC ? SPACEL : SPACE));
+ if (LOC) {
+ ret = reg_node(pRExC_state, (U8)(SPACEL));
+ } else {
+ ret = reg_node(pRExC_state, (U8)(SPACE));
+ FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0;
+ }
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'S':
- ret = reg_node(pRExC_state, (U8)(LOC ? NSPACEL : NSPACE));
+ if (LOC) {
+ ret = reg_node(pRExC_state, (U8)(NSPACEL));
+ } else {
+ ret = reg_node(pRExC_state, (U8)(NSPACE));
+ FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0;
+ }
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'd':
case '0': case '1': case '2': case '3':case '4':
case '5': case '6': case '7': case '8':case '9':
if (*p == '0' ||
- (isOCTAL(p[1]) && atoi(p) >= RExC_npar))
+ (isDIGIT(p[1]) && atoi(p) >= RExC_npar))
{
I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
STRLEN numlen = 3;
}
}
-
-#define _C_C_T_(NAME,TEST,WORD) \
+/* No locale test */
+#define _C_C_T_NOLOC_(NAME,TEST,WORD) \
ANYOF_##NAME: \
- if (LOC) \
- ANYOF_CLASS_SET(ret, ANYOF_##NAME); \
- else { \
for (value = 0; value < 256; value++) \
if (TEST) \
ANYOF_BITMAP_SET(ret, value); \
- } \
yesno = '+'; \
what = WORD; \
break; \
case ANYOF_N##NAME: \
- if (LOC) \
- ANYOF_CLASS_SET(ret, ANYOF_N##NAME); \
- else { \
for (value = 0; value < 256; value++) \
if (!TEST) \
ANYOF_BITMAP_SET(ret, value); \
- } \
yesno = '!'; \
what = WORD; \
break
-#define _C_C_T_NOLOC_(NAME,TEST,WORD) \
+/* Like the above, but there are differences if we are in uni-8-bit or not, so
+ * there are two tests passed in, to use depending on that. There aren't any
+ * cases where the label is different from the name, so no need for that
+ * parameter */
+#define _C_C_T_(NAME,TEST_8,TEST_7,WORD) \
ANYOF_##NAME: \
- for (value = 0; value < 256; value++) \
- if (TEST) \
- ANYOF_BITMAP_SET(ret, value); \
+ if (LOC) ANYOF_CLASS_SET(ret, ANYOF_##NAME); \
+ else if (UNI_SEMANTICS) { \
+ for (value = 0; value < 256; value++) { \
+ if (TEST_8) ANYOF_BITMAP_SET(ret, value); \
+ } \
+ } \
+ else { \
+ for (value = 0; value < 256; value++) { \
+ if (TEST_7) ANYOF_BITMAP_SET(ret, value); \
+ } \
+ } \
yesno = '+'; \
what = WORD; \
break; \
case ANYOF_N##NAME: \
- for (value = 0; value < 256; value++) \
- if (!TEST) \
- ANYOF_BITMAP_SET(ret, value); \
+ if (LOC) ANYOF_CLASS_SET(ret, ANYOF_N##NAME); \
+ else if (UNI_SEMANTICS) { \
+ for (value = 0; value < 256; value++) { \
+ if (! TEST_8) ANYOF_BITMAP_SET(ret, value); \
+ } \
+ } \
+ else { \
+ for (value = 0; value < 256; value++) { \
+ if (! TEST_7) ANYOF_BITMAP_SET(ret, value); \
+ } \
+ } \
yesno = '!'; \
what = WORD; \
break
#ifdef EBCDIC
UV literal_endpoint = 0;
#endif
- UV stored = 0; /* number of chars stored in the class */
+ UV stored = 0; /* 0, 1, or more than 1 chars stored in the class */
regnode * const orig_emit = RExC_emit; /* Save the original RExC_emit in
case we need to change the emitted regop to an EXACT. */
if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
- if (!SIZE_ONLY && !need_class)
- ANYOF_CLASS_ZERO(ret);
-
- need_class = 1;
+ /* What matches in a locale is not known until runtime, so need to
+ * (one time per class) allocate extra space to pass to regexec.
+ * The space will contain a bit for each named class that is to be
+ * matched against. This isn't needed for \p{} and pseudo-classes,
+ * as they are not affected by locale, and hence are dealt with
+ * separately */
+ if (LOC && namedclass < ANYOF_MAX && ! need_class) {
+ need_class = 1;
+ if (SIZE_ONLY) {
+ RExC_size += ANYOF_CLASS_ADD_SKIP;
+ }
+ else {
+ RExC_emit += ANYOF_CLASS_ADD_SKIP;
+ ANYOF_CLASS_ZERO(ret);
+ }
+ ANYOF_FLAGS(ret) |= ANYOF_CLASS|ANYOF_LARGE;
+ }
/* a bad range like a-\d, a-[:digit:] ? */
if (range) {
* --jhi */
switch ((I32)namedclass) {
- case _C_C_T_(ALNUMC, isALNUMC(value), POSIX_CC_UNI_NAME("Alnum"));
- case _C_C_T_(ALPHA, isALPHA(value), POSIX_CC_UNI_NAME("Alpha"));
- case _C_C_T_(BLANK, isBLANK(value), POSIX_CC_UNI_NAME("Blank"));
- case _C_C_T_(CNTRL, isCNTRL(value), POSIX_CC_UNI_NAME("Cntrl"));
- case _C_C_T_(GRAPH, isGRAPH(value), POSIX_CC_UNI_NAME("Graph"));
- case _C_C_T_(LOWER, isLOWER(value), POSIX_CC_UNI_NAME("Lower"));
- case _C_C_T_(PRINT, isPRINT(value), POSIX_CC_UNI_NAME("Print"));
- case _C_C_T_(PSXSPC, isPSXSPC(value), POSIX_CC_UNI_NAME("Space"));
- case _C_C_T_(PUNCT, isPUNCT(value), POSIX_CC_UNI_NAME("Punct"));
- case _C_C_T_(UPPER, isUPPER(value), POSIX_CC_UNI_NAME("Upper"));
+ case _C_C_T_(ALNUMC, isALNUMC_L1(value), isALNUMC(value), "XPosixAlnum");
+ case _C_C_T_(ALPHA, isALPHA_L1(value), isALPHA(value), "XPosixAlpha");
+ case _C_C_T_(BLANK, isBLANK_L1(value), isBLANK(value), "XPosixBlank");
+ case _C_C_T_(CNTRL, isCNTRL_L1(value), isCNTRL(value), "XPosixCntrl");
+ case _C_C_T_(GRAPH, isGRAPH_L1(value), isGRAPH(value), "XPosixGraph");
+ case _C_C_T_(LOWER, isLOWER_L1(value), isLOWER(value), "XPosixLower");
+ case _C_C_T_(PRINT, isPRINT_L1(value), isPRINT(value), "XPosixPrint");
+ case _C_C_T_(PSXSPC, isPSXSPC_L1(value), isPSXSPC(value), "XPosixSpace");
+ case _C_C_T_(PUNCT, isPUNCT_L1(value), isPUNCT(value), "XPosixPunct");
+ case _C_C_T_(UPPER, isUPPER_L1(value), isUPPER(value), "XPosixUpper");
#ifdef BROKEN_UNICODE_CHARCLASS_MAPPINGS
- case _C_C_T_(ALNUM, isALNUM(value), "Word");
- case _C_C_T_(SPACE, isSPACE(value), "SpacePerl");
+ /* \s, \w match all unicode if utf8. */
+ case _C_C_T_(SPACE, isSPACE_L1(value), isSPACE(value), "SpacePerl");
+ case _C_C_T_(ALNUM, isWORDCHAR_L1(value), isALNUM(value), "Word");
#else
- case _C_C_T_(SPACE, isSPACE(value), "PerlSpace");
- case _C_C_T_(ALNUM, isALNUM(value), "PerlWord");
+ /* \s, \w match ascii and locale only */
+ case _C_C_T_(SPACE, isSPACE_L1(value), isSPACE(value), "PerlSpace");
+ case _C_C_T_(ALNUM, isWORDCHAR_L1(value), isALNUM(value), "PerlWord");
#endif
- case _C_C_T_(XDIGIT, isXDIGIT(value), "XDigit");
+ case _C_C_T_(XDIGIT, isXDIGIT_L1(value), isXDIGIT(value), "XPosixXDigit");
case _C_C_T_NOLOC_(VERTWS, is_VERTWS_latin1(&value), "VertSpace");
case _C_C_T_NOLOC_(HORIZWS, is_HORIZWS_latin1(&value), "HorizSpace");
case ANYOF_ASCII:
/* Strings such as "+utf8::isWord\n" */
Perl_sv_catpvf(aTHX_ listsv, "%cutf8::Is%s\n", yesno, what);
}
- if (LOC)
- ANYOF_FLAGS(ret) |= ANYOF_CLASS;
+ stored+=2; /* can't optimize this class */
continue;
}
} /* end of namedclass \blah */
range = 0; /* this range (if it was one) is done now */
}
- if (need_class) {
- ANYOF_FLAGS(ret) |= ANYOF_LARGE;
- if (SIZE_ONLY)
- RExC_size += ANYOF_CLASS_ADD_SKIP;
- else
- RExC_emit += ANYOF_CLASS_ADD_SKIP;
- }
if (SIZE_ONLY)
/* nextchar()
- Advance that parse position, and optionall absorbs
+ Advance that parse position, and optionally absorbs
"whitespace" from the inputstream.
Without /x "whitespace" means (?#...) style comments only,
https://perl5.git.perl.org / perl5.git / blobdiff