#define WORST 0 /* Worst case. */
#define HASWIDTH 0x01 /* Known to match non-null strings. */
-/* Simple enough to be STAR/PLUS operand, in an EXACT node must be a single
+/* Simple enough to be STAR/PLUS operand; in an EXACT node must be a single
* character, and if utf8, must be invariant. Note that this is not the same
* thing as REGNODE_SIMPLE */
#define SIMPLE 0x02
#define FOLD cBOOL(RExC_flags & RXf_PMf_FOLD)
-#define OOB_UNICODE 12345678
#define OOB_NAMEDCLASS -1
+/* There is no code point that is out-of-bounds, so this is problematic. But
+ * its only current use is to initialize a variable that is always set before
+ * looked at. */
+#define OOB_UNICODE 0xDEADBEEF
+
#define CHR_SVLEN(sv) (UTF ? sv_len_utf8(sv) : SvCUR(sv))
#define CHR_DIST(a,b) (UTF ? utf8_distance(a,b) : a - b)
* problematic sequences. This delta is used by the caller to adjust the
* min length of the match, and the delta between min and max, so that the
* optimizer doesn't reject these possibilities based on size constraints.
- * 2) These sequences are not currently correctly handled by the trie code
- * either, so it changes the joined node type to ops that are not handled
- * by trie's, those new ops being EXACTFU_SS and EXACTFU_TRICKYFOLD.
+ * 2) These sequences require special handling by the trie code, so it
+ * changes the joined node type to ops for the trie's benefit, those new
+ * ops being EXACTFU_SS and EXACTFU_TRICKYFOLD.
* 3) This is sufficient for the two Greek sequences (described below), but
* the one involving the Sharp s (\xDF) needs more. The node type
* EXACTFU_SS is used for an EXACTFU node that contains at least one "ss"
data->flags |= (OP(scan) == MEOL
? SF_BEFORE_MEOL
: SF_BEFORE_SEOL);
+ SCAN_COMMIT(pRExC_state, data, minlenp);
+
}
else if ( PL_regkind[OP(scan)] == BRANCHJ
/* Lookbehind, or need to calculate parens/evals/stclass: */
* the original pattern needs upgrading to utf8.
*/
-bool
+static bool
S_compile_runtime_code(pTHX_ RExC_state_t * const pRExC_state,
char *pat, STRLEN plen)
{
return;
}
-
void
Perl__invlist_union_maybe_complement_2nd(pTHX_ SV* const a, SV* const b, bool complement_b, SV** output)
{
return TRUE;
}
+PERL_STATIC_INLINE UV
+S_invlist_highest(pTHX_ SV* const invlist)
+{
+ /* Returns the highest code point that matches an inversion list. This API
+ * has an ambiguity, as it returns 0 under either the highest is actually
+ * 0, or if the list is empty. If this distinction matters to you, check
+ * for emptiness before calling this function */
+
+ UV len = invlist_len(invlist);
+ UV *array;
+
+ PERL_ARGS_ASSERT_INVLIST_HIGHEST;
+
+ if (len == 0) {
+ return 0;
+ }
+
+ array = invlist_array(invlist);
+
+ /* The last element in the array in the inversion list always starts a
+ * range that goes to infinity. That range may be for code points that are
+ * matched in the inversion list, or it may be for ones that aren't
+ * matched. In the latter case, the highest code point in the set is one
+ * less than the beginning of this range; otherwise it is the final element
+ * of this range: infinity */
+ return (ELEMENT_RANGE_MATCHES_INVLIST(len - 1))
+ ? UV_MAX
+ : array[len - 1] - 1;
+}
+
#ifndef PERL_IN_XSUB_RE
SV *
Perl__invlist_contents(pTHX_ SV* const invlist)
return uv;
}
+PERL_STATIC_INLINE void
+S_alloc_maybe_populate_EXACT(pTHX_ RExC_state_t *pRExC_state, regnode *node, STRLEN len, UV code_point)
+{
+ /* This knows the details about sizing an EXACTish node, and potentially
+ * populating it with a single character. If <len> is non-zero, it assumes
+ * that the node has already been populated, and just does the sizing,
+ * ignoring <code_point>. Otherwise it looks at <code_point> and
+ * calculates what <len> should be. In pass 1, it sizes the node
+ * appropriately. In pass 2, it additionally will populate the node's
+ * STRING with <code_point>, if <len> is 0.
+ *
+ * It knows that under FOLD, UTF characters and the Latin Sharp S must be
+ * folded (the latter only when the rules indicate it can match 'ss') */
+
+ bool len_passed_in = cBOOL(len != 0);
+ U8 character[UTF8_MAXBYTES_CASE+1];
+
+ PERL_ARGS_ASSERT_ALLOC_MAYBE_POPULATE_EXACT;
+
+ if (! len_passed_in) {
+ if (UTF) {
+ if (FOLD) {
+ to_uni_fold(NATIVE_TO_UNI(code_point), character, &len);
+ }
+ else {
+ uvchr_to_utf8( character, code_point);
+ len = UTF8SKIP(character);
+ }
+ }
+ else if (! FOLD
+ || code_point != LATIN_SMALL_LETTER_SHARP_S
+ || MORE_ASCII_RESTRICTED
+ || ! AT_LEAST_UNI_SEMANTICS)
+ {
+ *character = (U8) code_point;
+ len = 1;
+ }
+ else {
+ *character = 's';
+ *(character + 1) = 's';
+ len = 2;
+ }
+ }
+
+ if (SIZE_ONLY) {
+ RExC_size += STR_SZ(len);
+ }
+ else {
+ RExC_emit += STR_SZ(len);
+ STR_LEN(node) = len;
+ if (! len_passed_in) {
+ Copy((char *) character, STRING(node), len, char);
+ }
+ }
+}
/*
- regatom - the lowest level
*flagp |= HASWIDTH;
goto finish_meta_pat;
case 'w':
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = ALNUML;
- break;
- case REGEX_UNICODE_CHARSET:
- op = ALNUMU;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = ALNUMA;
- break;
- case REGEX_DEPENDS_CHARSET:
- op = ALNUM;
- break;
- default:
- goto bad_charset;
+ op = ALNUM + get_regex_charset(RExC_flags);
+ if (op > ALNUMA) { /* /aa is same as /a */
+ op = ALNUMA;
}
ret = reg_node(pRExC_state, op);
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'W':
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = NALNUML;
- break;
- case REGEX_UNICODE_CHARSET:
- op = NALNUMU;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = NALNUMA;
- break;
- case REGEX_DEPENDS_CHARSET:
- op = NALNUM;
- break;
- default:
- goto bad_charset;
+ op = NALNUM + get_regex_charset(RExC_flags);
+ if (op > NALNUMA) { /* /aa is same as /a */
+ op = NALNUMA;
}
ret = reg_node(pRExC_state, op);
*flagp |= HASWIDTH|SIMPLE;
case 'b':
RExC_seen_zerolen++;
RExC_seen |= REG_SEEN_LOOKBEHIND;
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = BOUNDL;
- break;
- case REGEX_UNICODE_CHARSET:
- op = BOUNDU;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = BOUNDA;
- break;
- case REGEX_DEPENDS_CHARSET:
- op = BOUND;
- break;
- default:
- goto bad_charset;
+ op = BOUND + get_regex_charset(RExC_flags);
+ if (op > BOUNDA) { /* /aa is same as /a */
+ op = BOUNDA;
}
ret = reg_node(pRExC_state, op);
FLAGS(ret) = get_regex_charset(RExC_flags);
case 'B':
RExC_seen_zerolen++;
RExC_seen |= REG_SEEN_LOOKBEHIND;
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = NBOUNDL;
- break;
- case REGEX_UNICODE_CHARSET:
- op = NBOUNDU;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = NBOUNDA;
- break;
- case REGEX_DEPENDS_CHARSET:
- op = NBOUND;
- break;
- default:
- goto bad_charset;
+ op = NBOUND + get_regex_charset(RExC_flags);
+ if (op > NBOUNDA) { /* /aa is same as /a */
+ op = NBOUNDA;
}
ret = reg_node(pRExC_state, op);
FLAGS(ret) = get_regex_charset(RExC_flags);
*flagp |= SIMPLE;
goto finish_meta_pat;
case 's':
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = SPACEL;
- break;
- case REGEX_UNICODE_CHARSET:
- op = SPACEU;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = SPACEA;
- break;
- case REGEX_DEPENDS_CHARSET:
- op = SPACE;
- break;
- default:
- goto bad_charset;
+ op = SPACE + get_regex_charset(RExC_flags);
+ if (op > SPACEA) { /* /aa is same as /a */
+ op = SPACEA;
}
ret = reg_node(pRExC_state, op);
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'S':
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = NSPACEL;
- break;
- case REGEX_UNICODE_CHARSET:
- op = NSPACEU;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = NSPACEA;
- break;
- case REGEX_DEPENDS_CHARSET:
- op = NSPACE;
- break;
- default:
- goto bad_charset;
- }
- ret = reg_node(pRExC_state, op);
- *flagp |= HASWIDTH|SIMPLE;
- goto finish_meta_pat;
- case 'd':
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = DIGITL;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = DIGITA;
- break;
- case REGEX_DEPENDS_CHARSET: /* No difference between these */
- case REGEX_UNICODE_CHARSET:
- op = DIGIT;
- break;
- default:
- goto bad_charset;
+ op = NSPACE + get_regex_charset(RExC_flags);
+ if (op > NSPACEA) { /* /aa is same as /a */
+ op = NSPACEA;
}
ret = reg_node(pRExC_state, op);
*flagp |= HASWIDTH|SIMPLE;
goto finish_meta_pat;
case 'D':
- switch (get_regex_charset(RExC_flags)) {
- case REGEX_LOCALE_CHARSET:
- op = NDIGITL;
- break;
- case REGEX_ASCII_RESTRICTED_CHARSET:
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- op = NDIGITA;
- break;
- case REGEX_DEPENDS_CHARSET: /* No difference between these */
- case REGEX_UNICODE_CHARSET:
- op = NDIGIT;
- break;
- default:
- goto bad_charset;
+ op = NDIGIT;
+ goto join_D_and_d;
+ case 'd':
+ op = DIGIT;
+ join_D_and_d:
+ {
+ U8 offset = get_regex_charset(RExC_flags);
+ if (offset == REGEX_UNICODE_CHARSET) {
+ offset = REGEX_DEPENDS_CHARSET;
+ }
+ else if (offset == REGEX_ASCII_MORE_RESTRICTED_CHARSET) {
+ offset = REGEX_ASCII_RESTRICTED_CHARSET;
+ }
+ op += offset;
}
ret = reg_node(pRExC_state, op);
*flagp |= HASWIDTH|SIMPLE;
bool is_exactfu_sharp_s;
ender = 0;
- node_type = ((! FOLD) ? EXACT
- : (LOC)
- ? EXACTFL
- : (MORE_ASCII_RESTRICTED)
- ? EXACTFA
- : (AT_LEAST_UNI_SEMANTICS)
- ? EXACTFU
- : EXACTF);
+ if (! FOLD) {
+ node_type = EXACT;
+ }
+ else {
+ node_type = get_regex_charset(RExC_flags);
+ if (node_type >= REGEX_ASCII_RESTRICTED_CHARSET) {
+ node_type--; /* /a is same as /u, and map /aa's offset to
+ what /a's would have been, so there is no
+ hole */
+ }
+ node_type += EXACTF;
+ }
ret = reg_node(pRExC_state, node_type);
s = STRING(ret);
if (len == 1 && UNI_IS_INVARIANT(ender))
*flagp |= SIMPLE;
- if (SIZE_ONLY)
- RExC_size += STR_SZ(len);
- else {
- STR_LEN(ret) = len;
- RExC_emit += STR_SZ(len);
- }
+ alloc_maybe_populate_EXACT(pRExC_state, ret, len, 0);
}
break;
}
return(ret);
-
-/* Jumped to when an unrecognized character set is encountered */
-bad_charset:
- Perl_croak(aTHX_ "panic: Unknown regex character set encoding: %u", get_regex_charset(RExC_flags));
- return(NULL);
}
STATIC char *
switch (skip) {
case 4:
if (memEQ(posixcc, "word", 4)) /* this is not POSIX, this is the Perl \w */
- namedclass = complement ? ANYOF_NALNUM : ANYOF_ALNUM;
+ namedclass = ANYOF_ALNUM;
break;
case 5:
/* Names all of length 5. */
switch (posixcc[4]) {
case 'a':
if (memEQ(posixcc, "alph", 4)) /* alpha */
- namedclass = complement ? ANYOF_NALPHA : ANYOF_ALPHA;
+ namedclass = ANYOF_ALPHA;
break;
case 'e':
if (memEQ(posixcc, "spac", 4)) /* space */
- namedclass = complement ? ANYOF_NPSXSPC : ANYOF_PSXSPC;
+ namedclass = ANYOF_PSXSPC;
break;
case 'h':
if (memEQ(posixcc, "grap", 4)) /* graph */
- namedclass = complement ? ANYOF_NGRAPH : ANYOF_GRAPH;
+ namedclass = ANYOF_GRAPH;
break;
case 'i':
if (memEQ(posixcc, "asci", 4)) /* ascii */
- namedclass = complement ? ANYOF_NASCII : ANYOF_ASCII;
+ namedclass = ANYOF_ASCII;
break;
case 'k':
if (memEQ(posixcc, "blan", 4)) /* blank */
- namedclass = complement ? ANYOF_NBLANK : ANYOF_BLANK;
+ namedclass = ANYOF_BLANK;
break;
case 'l':
if (memEQ(posixcc, "cntr", 4)) /* cntrl */
- namedclass = complement ? ANYOF_NCNTRL : ANYOF_CNTRL;
+ namedclass = ANYOF_CNTRL;
break;
case 'm':
if (memEQ(posixcc, "alnu", 4)) /* alnum */
- namedclass = complement ? ANYOF_NALNUMC : ANYOF_ALNUMC;
+ namedclass = ANYOF_ALNUMC;
break;
case 'r':
if (memEQ(posixcc, "lowe", 4)) /* lower */
- namedclass = complement ? ANYOF_NLOWER : ANYOF_LOWER;
+ namedclass = ANYOF_LOWER;
else if (memEQ(posixcc, "uppe", 4)) /* upper */
- namedclass = complement ? ANYOF_NUPPER : ANYOF_UPPER;
+ namedclass = ANYOF_UPPER;
break;
case 't':
if (memEQ(posixcc, "digi", 4)) /* digit */
- namedclass = complement ? ANYOF_NDIGIT : ANYOF_DIGIT;
+ namedclass = ANYOF_DIGIT;
else if (memEQ(posixcc, "prin", 4)) /* print */
- namedclass = complement ? ANYOF_NPRINT : ANYOF_PRINT;
+ namedclass = ANYOF_PRINT;
else if (memEQ(posixcc, "punc", 4)) /* punct */
- namedclass = complement ? ANYOF_NPUNCT : ANYOF_PUNCT;
+ namedclass = ANYOF_PUNCT;
break;
}
break;
case 6:
if (memEQ(posixcc, "xdigit", 6))
- namedclass = complement ? ANYOF_NXDIGIT : ANYOF_XDIGIT;
+ namedclass = ANYOF_XDIGIT;
break;
}
if (namedclass == OOB_NAMEDCLASS)
Simple_vFAIL3("POSIX class [:%.*s:] unknown",
t - s - 1, s + 1);
+
+ /* The #defines are structured so each complement is +1 to
+ * the normal one */
+ if (complement) {
+ namedclass++;
+ }
assert (posixcc[skip] == ':');
assert (posixcc[skip+1] == ']');
} else if (!SIZE_ONLY) {
* determined at run-time
* run_time_list is a SV* that contains text names of properties that are to
* be computed at run time. This concatenates <Xpropertyname>
- * to it, apppropriately
+ * to it, appropriately
* This is essentially DO_POSIX, but we know only the Latin1 values at compile
* time */
#define DO_POSIX_LATIN1_ONLY_KNOWN(node, class, destlist, sourcelist, \
}
/* Like DO_POSIX_LATIN1_ONLY_KNOWN, but for the complement. A combination of
- * this and DO_N_POSIX */
+ * this and DO_N_POSIX. Sets <matches_above_unicode> only if it can; unchanged
+ * otherwise */
#define DO_N_POSIX_LATIN1_ONLY_KNOWN(node, class, destlist, sourcelist, \
- l1_sourcelist, Xpropertyname, run_time_list) \
+ l1_sourcelist, Xpropertyname, run_time_list, matches_above_unicode) \
if (AT_LEAST_ASCII_RESTRICTED) { \
_invlist_union_complement_2nd(destlist, sourcelist, &destlist); \
} \
else { \
Perl_sv_catpvf(aTHX_ run_time_list, "!utf8::%s\n", Xpropertyname); \
+ matches_above_unicode = TRUE; \
if (LOC) { \
- ANYOF_CLASS_SET(node, namedclass); \
+ ANYOF_CLASS_SET(node, namedclass); \
} \
else { \
SV* scratch_list = NULL; \
return;
}
+/* The names of properties whose definitions are not known at compile time are
+ * stored in this SV, after a constant heading. So if the length has been
+ * changed since initialization, then there is a run-time definition. */
+#define HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION (SvCUR(listsv) != initial_listsv_len)
+
/*
parse a class specification and produce either an ANYOF node that
matches the pattern or perhaps will be optimized into an EXACTish node
UV value = 0; /* XXX:dmq: needs to be referenceable (unfortunately) */
register regnode *ret;
STRLEN numlen;
- IV namedclass;
+ IV namedclass = OOB_NAMEDCLASS;
char *rangebegin = NULL;
bool need_class = 0;
bool allow_full_fold = TRUE; /* Assume wants multi-char folding */
STRLEN initial_listsv_len = 0; /* Kind of a kludge to see if it is more
than just initialized. */
SV* properties = NULL; /* Code points that match \p{} \P{} */
+ SV* posixes = NULL; /* Code points that match classes like, [:word:],
+ extended beyond the Latin1 range */
UV element_count = 0; /* Number of distinct elements in the class.
Optimizations may be possible if this is tiny */
UV n;
/* Set if a component of this character class is user-defined; just passed
* on to the engine */
- UV has_user_defined_property = 0;
+ bool has_user_defined_property = FALSE;
/* inversion list of code points this node matches only when the target
* string is in UTF-8. (Because is under /d) */
UV literal_endpoint = 0;
#endif
UV stored = 0; /* how many chars stored in the bitmap */
+ bool invert = FALSE; /* Is this class to be complemented */
+
+ /* Is there any thing like \W or [:^digit:] that matches above the legal
+ * Unicode range? */
+ bool runtime_posix_matches_above_Unicode = FALSE;
regnode * const orig_emit = RExC_emit; /* Save the original RExC_emit in
case we need to change the emitted regop to an EXACT. */
if (UCHARAT(RExC_parse) == '^') { /* Complement of range. */
RExC_naughty++;
RExC_parse++;
- if (!SIZE_ONLY)
- ANYOF_FLAGS(ret) |= ANYOF_INVERT;
+ invert = TRUE;
/* We have decided to not allow multi-char folds in inverted character
* classes, due to the confusion that can happen, especially with
SV** invlistsvp;
SV* invlist;
char* name;
+
if (UCHARAT(RExC_parse) == '^') {
RExC_parse++;
n--;
Perl_sv_catpvf(aTHX_ listsv, "%cutf8::%s\n",
(value == 'p' ? '+' : '!'),
name);
- has_user_defined_property = 1;
+ has_user_defined_property = TRUE;
/* We don't know yet, so have to assume that the
* property could match something in the Latin1 range,
- * hence something that isn't utf8 */
+ * hence something that isn't utf8. Note that this
+ * would cause things in <depends_list> to match
+ * inappropriately, except that any \p{}, including
+ * this one forces Unicode semantics, which means there
+ * is <no depends_list> */
ANYOF_FLAGS(ret) |= ANYOF_NONBITMAP_NON_UTF8;
}
else {
/* Here, did get the swash and its inversion list. If
* the swash is from a user-defined property, then this
* whole character class should be regarded as such */
- SV** user_defined_svp =
- hv_fetchs(MUTABLE_HV(SvRV(swash)),
- "USER_DEFINED", FALSE);
- if (user_defined_svp) {
- has_user_defined_property
- |= SvUV(*user_defined_svp);
- }
+ has_user_defined_property =
+ _is_swash_user_defined(swash);
/* Invert if asking for the complement */
if (value == 'P') {
- _invlist_union_complement_2nd(properties, invlist, &properties);
+ _invlist_union_complement_2nd(properties,
+ invlist,
+ &properties);
/* The swash can't be used as-is, because we've
* inverted things; delay removing it to here after
literal_endpoint++;
#endif
- if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
-
- /* 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) {
+ /* What matches in a locale is not known until runtime. This
+ * includes what the Posix classes (like \w, [:space:]) match.
+ * Room must be reserved (one time per class) to store such
+ * classes, either if Perl is compiled so that locale nodes always
+ * should have this space, or if there is such class info to be
+ * stored. 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
+ && ! need_class
+ && (ANYOF_LOCALE == ANYOF_CLASS
+ || (namedclass > OOB_NAMEDCLASS && namedclass < ANYOF_MAX)))
+ {
need_class = 1;
if (SIZE_ONLY) {
RExC_size += ANYOF_CLASS_SKIP - ANYOF_SKIP;
ANYOF_FLAGS(ret) |= ANYOF_CLASS;
}
+ if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
+
/* a bad range like a-\d, a-[:digit:]. The '-' is taken as a
* literal, as is the character that began the false range, i.e.
* the 'a' in the examples */
}
range = 0; /* this was not a true range */
+ element_count += 2; /* So counts for three values */
}
- if (!SIZE_ONLY) {
-
- /* Possible truncation here but in some 64-bit environments
- * the compiler gets heartburn about switch on 64-bit values.
- * A similar issue a little earlier when switching on value.
- * --jhi */
+ if (! SIZE_ONLY) {
switch ((I32)namedclass) {
case ANYOF_ALNUMC: /* C's alnum, in contrast to \w */
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
PL_PosixAlnum, PL_L1PosixAlnum, "XPosixAlnum", listsv);
break;
case ANYOF_NALNUMC:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixAlnum, PL_L1PosixAlnum, "XPosixAlnum", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixAlnum, PL_L1PosixAlnum, "XPosixAlnum", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_ALPHA:
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
PL_PosixAlpha, PL_L1PosixAlpha, "XPosixAlpha", listsv);
break;
case ANYOF_NALPHA:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixAlpha, PL_L1PosixAlpha, "XPosixAlpha", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixAlpha, PL_L1PosixAlpha, "XPosixAlpha", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_ASCII:
if (LOC) {
ANYOF_CLASS_SET(ret, namedclass);
}
else {
- _invlist_union(properties, PL_ASCII, &properties);
+ _invlist_union(posixes, PL_ASCII, &posixes);
}
break;
case ANYOF_NASCII:
ANYOF_CLASS_SET(ret, namedclass);
}
else {
- _invlist_union_complement_2nd(properties,
- PL_ASCII, &properties);
+ _invlist_union_complement_2nd(posixes,
+ PL_ASCII, &posixes);
if (DEPENDS_SEMANTICS) {
ANYOF_FLAGS(ret) |= ANYOF_NON_UTF8_LATIN1_ALL;
}
}
break;
case ANYOF_BLANK:
- DO_POSIX(ret, namedclass, properties,
+ DO_POSIX(ret, namedclass, posixes,
PL_PosixBlank, PL_XPosixBlank);
break;
case ANYOF_NBLANK:
- DO_N_POSIX(ret, namedclass, properties,
+ DO_N_POSIX(ret, namedclass, posixes,
PL_PosixBlank, PL_XPosixBlank);
break;
case ANYOF_CNTRL:
- DO_POSIX(ret, namedclass, properties,
+ DO_POSIX(ret, namedclass, posixes,
PL_PosixCntrl, PL_XPosixCntrl);
break;
case ANYOF_NCNTRL:
- DO_N_POSIX(ret, namedclass, properties,
+ DO_N_POSIX(ret, namedclass, posixes,
PL_PosixCntrl, PL_XPosixCntrl);
break;
case ANYOF_DIGIT:
/* There are no digits in the Latin1 range outside of
* ASCII, so call the macro that doesn't have to resolve
* them */
- DO_POSIX_LATIN1_ONLY_KNOWN_L1_RESOLVED(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN_L1_RESOLVED(ret, namedclass, posixes,
PL_PosixDigit, "XPosixDigit", listsv);
break;
case ANYOF_NDIGIT:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixDigit, PL_PosixDigit, "XPosixDigit", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixDigit, PL_PosixDigit, "XPosixDigit", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_GRAPH:
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
PL_PosixGraph, PL_L1PosixGraph, "XPosixGraph", listsv);
break;
case ANYOF_NGRAPH:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixGraph, PL_L1PosixGraph, "XPosixGraph", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixGraph, PL_L1PosixGraph, "XPosixGraph", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_HORIZWS:
/* For these, we use the cp_list, as /d doesn't make a
Xname = "XPosixLower";
}
if (namedclass == ANYOF_LOWER) {
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
ascii_source, l1_source, Xname, listsv);
}
else {
DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass,
- properties, ascii_source, l1_source, Xname, listsv);
+ posixes, ascii_source, l1_source, Xname, listsv,
+ runtime_posix_matches_above_Unicode);
}
break;
}
case ANYOF_PRINT:
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
PL_PosixPrint, PL_L1PosixPrint, "XPosixPrint", listsv);
break;
case ANYOF_NPRINT:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixPrint, PL_L1PosixPrint, "XPosixPrint", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixPrint, PL_L1PosixPrint, "XPosixPrint", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_PUNCT:
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
PL_PosixPunct, PL_L1PosixPunct, "XPosixPunct", listsv);
break;
case ANYOF_NPUNCT:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixPunct, PL_L1PosixPunct, "XPosixPunct", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixPunct, PL_L1PosixPunct, "XPosixPunct", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_PSXSPC:
- DO_POSIX(ret, namedclass, properties,
+ DO_POSIX(ret, namedclass, posixes,
PL_PosixSpace, PL_XPosixSpace);
break;
case ANYOF_NPSXSPC:
- DO_N_POSIX(ret, namedclass, properties,
+ DO_N_POSIX(ret, namedclass, posixes,
PL_PosixSpace, PL_XPosixSpace);
break;
case ANYOF_SPACE:
- DO_POSIX(ret, namedclass, properties,
+ DO_POSIX(ret, namedclass, posixes,
PL_PerlSpace, PL_XPerlSpace);
break;
case ANYOF_NSPACE:
- DO_N_POSIX(ret, namedclass, properties,
+ DO_N_POSIX(ret, namedclass, posixes,
PL_PerlSpace, PL_XPerlSpace);
break;
case ANYOF_UPPER: /* Same as LOWER, above */
Xname = "XPosixUpper";
}
if (namedclass == ANYOF_UPPER) {
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
ascii_source, l1_source, Xname, listsv);
}
else {
DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass,
- properties, ascii_source, l1_source, Xname, listsv);
+ posixes, ascii_source, l1_source, Xname, listsv,
+ runtime_posix_matches_above_Unicode);
}
break;
}
case ANYOF_ALNUM: /* Really is 'Word' */
- DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
+ DO_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
PL_PosixWord, PL_L1PosixWord, "XPosixWord", listsv);
break;
case ANYOF_NALNUM:
- DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, properties,
- PL_PosixWord, PL_L1PosixWord, "XPosixWord", listsv);
+ DO_N_POSIX_LATIN1_ONLY_KNOWN(ret, namedclass, posixes,
+ PL_PosixWord, PL_L1PosixWord, "XPosixWord", listsv,
+ runtime_posix_matches_above_Unicode);
break;
case ANYOF_VERTWS:
/* For these, we use the cp_list, as /d doesn't make a
PL_VertSpace, &cp_list);
break;
case ANYOF_XDIGIT:
- DO_POSIX(ret, namedclass, properties,
+ DO_POSIX(ret, namedclass, posixes,
PL_PosixXDigit, PL_XPosixXDigit);
break;
case ANYOF_NXDIGIT:
- DO_N_POSIX(ret, namedclass, properties,
+ DO_N_POSIX(ret, namedclass, posixes,
PL_PosixXDigit, PL_XPosixXDigit);
break;
case ANYOF_MAX:
break;
}
- continue;
+ continue; /* Go get next character */
}
} /* end of namedclass \blah */
}
}
else {
- prevvalue = value; /* save the beginning of the range */
+ prevvalue = value; /* save the beginning of the potential range */
if (RExC_parse+1 < RExC_end
&& *RExC_parse == '-'
&& RExC_parse[1] != ']')
}
}
+ /* Here, <prevvalue> is the beginning of the range, if any; or <value>
+ * if not */
+
/* non-Latin1 code point implies unicode semantics. Must be set in
* pass1 so is there for the whole of pass 2 */
if (value > 255) {
RExC_uni_semantics = 1;
}
- /* now is the next time */
+ /* Ready to process either the single value, or the completed range */
if (!SIZE_ONLY) {
#ifndef EBCDIC
cp_list = _add_range_to_invlist(cp_list, prevvalue, value);
}
range = 0; /* this range (if it was one) is done now */
+ } /* End of loop through all the text within the brackets */
+
+ /* If the character class contains only a single element, it may be
+ * optimizable into another node type which is smaller and runs faster.
+ * Check if this is the case for this class */
+ if (element_count == 1) {
+ U8 op = END;
+
+ if (namedclass > OOB_NAMEDCLASS) { /* this is a named class, like \w or
+ [:digit:] or \p{foo} */
+
+ /* Certain named classes have equivalents that can appear outside a
+ * character class, e.g. \w, \H. We use these instead of a
+ * character class. */
+ switch ((I32)namedclass) {
+ U8 offset;
+
+ /* The first group is for node types that depend on the charset
+ * modifier to the regex. We first calculate the base node
+ * type, and if it should be inverted */
+
+ case ANYOF_NALNUM:
+ invert = ! invert;
+ /* FALLTHROUGH */
+ case ANYOF_ALNUM:
+ op = ALNUM;
+ goto join_charset_classes;
+
+ case ANYOF_NSPACE:
+ invert = ! invert;
+ /* FALLTHROUGH */
+ case ANYOF_SPACE:
+ op = SPACE;
+ goto join_charset_classes;
+
+ case ANYOF_NDIGIT:
+ invert = ! invert;
+ /* FALLTHROUGH */
+ case ANYOF_DIGIT:
+ op = DIGIT;
+
+ join_charset_classes:
+
+ /* Now that we have the base node type, we take advantage
+ * of the enum ordering of the charset modifiers to get the
+ * exact node type, For example the base SPACE also has
+ * SPACEL, SPACEU, and SPACEA */
+
+ offset = get_regex_charset(RExC_flags);
+
+ /* /aa is the same as /a for these */
+ if (offset == REGEX_ASCII_MORE_RESTRICTED_CHARSET) {
+ offset = REGEX_ASCII_RESTRICTED_CHARSET;
+ }
+ else if (op == DIGIT && offset == REGEX_UNICODE_CHARSET) {
+ offset = REGEX_DEPENDS_CHARSET; /* There is no DIGITU */
+ }
+
+ op += offset;
+
+ /* The number of varieties of each of these is the same,
+ * hence, so is the delta between the normal and
+ * complemented nodes */
+ if (invert) {
+ op += NALNUM - ALNUM;
+ }
+ break;
+
+ /* The second group doesn't depend of the charset modifiers.
+ * We just have normal and complemented */
+ case ANYOF_NHORIZWS:
+ invert = ! invert;
+ /* FALLTHROUGH */
+ case ANYOF_HORIZWS:
+ op = (invert) ? NHORIZWS : HORIZWS;
+ break;
+
+ case ANYOF_NVERTWS:
+ invert = ! invert;
+ /* FALLTHROUGH */
+ case ANYOF_VERTWS:
+ op = (invert) ? NVERTWS : VERTWS;
+ break;
+
+
+ }
+ }
+ else if (! LOC) {
+ if (invert && prevvalue == '\n' && value == '\n') {
+ op = REG_ANY; /* Optimize [^\n] */
+ }
+ else if (prevvalue == '0' && value == '9') {
+ op = (invert) ? NDIGITA : DIGITA;
+ }
+ }
+
+ /* Here, we have changed <op> away from its initial value iff we found
+ * an optimization */
+ if (op != END) {
+
+ /* Throw away this ANYOF regnode, and emit the calculated one,
+ * which should correspond to the beginning, not current, state of
+ * the parse */
+ const char * cur_parse = RExC_parse;
+ RExC_parse = (char *)orig_parse;
+ RExC_emit = (regnode *)orig_emit;
+
+ ret = reg_node(pRExC_state, op);
+
+ RExC_parse = (char *) cur_parse;
+
+ SvREFCNT_dec(listsv);
+ return ret;
+ }
}
if (SIZE_ONLY)
SV* fold_intersection = NULL;
- const UV highest_index = invlist_len(cp_list) - 1;
-
/* In the Latin1 range, the characters that can be folded-to or -from
* are precisely the alphabetic characters. If the highest code point
* is within Latin1, we can use the compiled-in list, and not have to
- * go out to disk. If the last element in the array is in the
- * inversion list set, it starts a range that goes to infinity, so the
- * maximum of the inversion list is definitely above Latin1.
- * Otherwise, it starts a range that isn't in the set, so the max is
- * one less than it */
- if (! ELEMENT_RANGE_MATCHES_INVLIST(highest_index)
- && invlist_array(cp_list)[highest_index] <= 256)
- {
+ * go out to disk. */
+ if (invlist_highest(cp_list) < 256) {
_invlist_intersection(PL_L1PosixAlpha, cp_list, &fold_intersection);
}
else {
- /* This is a list of all the characters that participate in folds
- * (except marks, etc in multi-char folds */
+ /* Here, there are non-Latin1 code points, so we will have to go
+ * fetch the list of all the characters that participate in folds
+ */
if (! PL_utf8_foldable) {
- SV* swash = swash_init("utf8", "Cased", &PL_sv_undef, 1, 0);
+ SV* swash = swash_init("utf8", "_Perl_Any_Folds",
+ &PL_sv_undef, 1, 0);
PL_utf8_foldable = _swash_to_invlist(swash);
SvREFCNT_dec(swash);
}
}
else {
depends_list =
- add_cp_to_invlist(depends_list, PL_fold_latin1[j]);
+ add_cp_to_invlist(depends_list, PL_fold_latin1[j]);
}
}
switch (j) {
case 'k':
case 'K':
- /* KELVIN SIGN */
cp_list =
- add_cp_to_invlist(cp_list, 0x212A);
+ add_cp_to_invlist(cp_list, KELVIN_SIGN);
break;
case 's':
case 'S':
- /* LATIN SMALL LETTER LONG S */
- cp_list =
- add_cp_to_invlist(cp_list, 0x017F);
+ cp_list = add_cp_to_invlist(cp_list,
+ LATIN_SMALL_LETTER_LONG_S);
break;
case MICRO_SIGN:
cp_list = add_cp_to_invlist(cp_list,
break;
case LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE:
case LATIN_SMALL_LETTER_A_WITH_RING_ABOVE:
- /* ANGSTROM SIGN */
cp_list =
- add_cp_to_invlist(cp_list, 0x212B);
+ add_cp_to_invlist(cp_list, ANGSTROM_SIGN);
break;
case LATIN_SMALL_LETTER_Y_WITH_DIAERESIS:
cp_list = add_cp_to_invlist(cp_list,
/* /aa doesn't allow folds between ASCII and non-;
* /l doesn't allow them between above and below
* 256 */
- if ((MORE_ASCII_RESTRICTED && (isASCII(c) != isASCII(j)))
+ if ((MORE_ASCII_RESTRICTED
+ && (isASCII(c) != isASCII(j)))
|| (LOC && ((c < 256) != (j < 256))))
{
continue;
cp_list = add_cp_to_invlist(cp_list, c);
}
else {
- depends_list = add_cp_to_invlist(depends_list, c);
+ depends_list = add_cp_to_invlist(depends_list, c);
}
}
}
SvREFCNT_dec(fold_intersection);
}
- /* And combine the result (if any) with any inversion list from properties.
- * The lists are kept separate up to now because we don't want to fold the
- * properties */
- if (properties) {
+ /* And combine the result (if any) with any inversion list from posix
+ * classes. The lists are kept separate up to now because we don't want to
+ * fold the classes (folding of those is automatically handled by the swash
+ * fetching code) */
+ if (posixes) {
if (AT_LEAST_UNI_SEMANTICS) {
if (cp_list) {
- _invlist_union(cp_list, properties, &cp_list);
- SvREFCNT_dec(properties);
+ _invlist_union(cp_list, posixes, &cp_list);
+ SvREFCNT_dec(posixes);
}
else {
- cp_list = properties;
+ cp_list = posixes;
}
}
else {
- /* Under /d, we put the things that match only when the target
- * string is utf8, into a separate list */
+ /* Under /d, we put into a separate list the Latin1 things that
+ * match only when the target string is utf8 */
SV* nonascii_but_latin1_properties = NULL;
- _invlist_intersection(properties, PL_Latin1,
+ _invlist_intersection(posixes, PL_Latin1,
&nonascii_but_latin1_properties);
_invlist_subtract(nonascii_but_latin1_properties, PL_ASCII,
&nonascii_but_latin1_properties);
- _invlist_subtract(properties, nonascii_but_latin1_properties,
- &properties);
+ _invlist_subtract(posixes, nonascii_but_latin1_properties,
+ &posixes);
if (cp_list) {
- _invlist_union(cp_list, properties, &cp_list);
- SvREFCNT_dec(properties);
+ _invlist_union(cp_list, posixes, &cp_list);
+ SvREFCNT_dec(posixes);
}
else {
- cp_list = properties;
+ cp_list = posixes;
}
if (depends_list) {
}
}
+ /* And combine the result (if any) with any inversion list from properties.
+ * The lists are kept separate up to now so that we can distinguish the two
+ * in regards to matching above-Unicode. A run-time warning is generated
+ * if a Unicode property is matched against a non-Unicode code point. But,
+ * we allow user-defined properties to match anything, without any warning,
+ * and we also suppress the warning if there is a portion of the character
+ * class that isn't a Unicode property, and which matches above Unicode, \W
+ * or [\x{110000}] for example.
+ * (Note that in this case, unlike the Posix one above, there is no
+ * <depends_list>, because having a Unicode property forces Unicode
+ * semantics */
+ if (properties) {
+ bool warn_super = ! has_user_defined_property;
+ if (cp_list) {
+
+ /* If it matters to the final outcome, see if a non-property
+ * component of the class matches above Unicode. If so, the
+ * warning gets suppressed. This is true even if just a single
+ * such code point is specified, as though not strictly correct if
+ * another such code point is matched against, the fact that they
+ * are using above-Unicode code points indicates they should know
+ * the issues involved */
+ if (warn_super) {
+ bool non_prop_matches_above_Unicode =
+ runtime_posix_matches_above_Unicode
+ | (invlist_highest(cp_list) > PERL_UNICODE_MAX);
+ if (invert) {
+ non_prop_matches_above_Unicode =
+ ! non_prop_matches_above_Unicode;
+ }
+ warn_super = ! non_prop_matches_above_Unicode;
+ }
+
+ _invlist_union(properties, cp_list, &cp_list);
+ SvREFCNT_dec(properties);
+ }
+ else {
+ cp_list = properties;
+ }
+
+ if (warn_super) {
+ ANYOF_FLAGS(ret) |= ANYOF_WARN_SUPER;
+ }
+ }
+
+ /* Here, we have calculated what code points should be in the character
+ * class.
+ *
+ * Now we can see about various optimizations. Fold calculation (which we
+ * did above) needs to take place before inversion. Otherwise /[^k]/i
+ * would invert to include K, which under /i would match k, which it
+ * shouldn't. */
+
+ /* Optimize inverted simple patterns (e.g. [^a-z]). Note that we haven't
+ * set the FOLD flag yet, so this does optimize those. It doesn't
+ * optimize locale. Doing so perhaps could be done as long as there is
+ * nothing like \w in it; some thought also would have to be given to the
+ * interaction with above 0x100 chars */
+ if (invert
+ && ! LOC
+ && ! depends_list
+ && ! unicode_alternate
+ && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION)
+ {
+ _invlist_invert(cp_list);
+
+ /* Any swash can't be used as-is, because we've inverted things */
+ if (swash) {
+ SvREFCNT_dec(swash);
+ swash = NULL;
+ }
+
+ /* Clear the invert flag since have just done it here */
+ invert = FALSE;
+ }
+
/* Here, <cp_list> contains all the code points we can determine at
* compile time that match under all conditions. Go through it, and
* for things that belong in the bitmap, put them there, and delete from
}
}
+ if (invert) {
+ ANYOF_FLAGS(ret) |= ANYOF_INVERT;
+ }
+
/* Combine the two lists into one. */
if (depends_list) {
if (cp_list) {
}
}
- /* Here, we have calculated what code points should be in the character
- * class. <cp_list> does not overlap the bitmap except possibly in the
- * case of DEPENDS rules.
- *
- * Now we can see about various optimizations. Fold calculation (which we
- * did above) needs to take place before inversion. Otherwise /[^k]/i
- * would invert to include K, which under /i would match k, which it
- * shouldn't. */
-
- /* Optimize inverted simple patterns (e.g. [^a-z]). Note that we haven't
- * set the FOLD flag yet, so this does optimize those. It doesn't
- * optimize locale. Doing so perhaps could be done as long as there is
- * nothing like \w in it; some thought also would have to be given to the
- * interaction with above 0x100 chars */
- if ((ANYOF_FLAGS(ret) & ANYOF_INVERT)
- && ! LOC
- && ! unicode_alternate
- /* In case of /d, there are some things that should match only when in
- * not in the bitmap, i.e., they require UTF8 to match. These are
- * listed in cp_list, but if ANYOF_NONBITMAP_NON_UTF8 is set in this
- * case, they don't require UTF8, so can invert here */
- && (! cp_list
- || ! DEPENDS_SEMANTICS
- || (ANYOF_FLAGS(ret) & ANYOF_NONBITMAP_NON_UTF8))
- && SvCUR(listsv) == initial_listsv_len)
- {
- int i;
- if (! cp_list) {
- for (i = 0; i < 256; ++i) {
- if (ANYOF_BITMAP_TEST(ret, i)) {
- ANYOF_BITMAP_CLEAR(ret, i);
- }
- else {
- ANYOF_BITMAP_SET(ret, i);
- prevvalue = value;
- value = i;
- }
- }
- /* The inversion means that everything above 255 is matched */
- ANYOF_FLAGS(ret) |= ANYOF_UNICODE_ALL;
- }
- else {
- /* Here, also has things outside the bitmap that may overlap with
- * the bitmap. We have to sync them up, so that they get inverted
- * in both places. Earlier, we removed all overlaps except in the
- * case of /d rules, so no syncing is needed except for this case
- */
- SV *remove_list = NULL;
-
- if (DEPENDS_SEMANTICS) {
- UV start, end;
-
- /* Set the bits that correspond to the ones that aren't in the
- * bitmap. Otherwise, when we invert, we'll miss these.
- * Earlier, we removed from the cp_list all code points
- * < 128, so there is no extra work here */
- invlist_iterinit(cp_list);
- while (invlist_iternext(cp_list, &start, &end)) {
- if (start > 255) { /* The bit map goes to 255 */
- break;
- }
- if (end > 255) {
- end = 255;
- }
- for (i = start; i <= (int) end; ++i) {
- ANYOF_BITMAP_SET(ret, i);
- prevvalue = value;
- value = i;
- }
- }
- }
-
- /* Now invert both the bitmap and the cp_list. Anything in the
- * bitmap has to also be removed from the non-bitmap, but again,
- * there should not be overlap unless is /d rules. */
- _invlist_invert(cp_list);
-
- /* Any swash can't be used as-is, because we've inverted things */
- if (swash) {
- SvREFCNT_dec(swash);
- swash = NULL;
- }
-
- for (i = 0; i < 256; ++i) {
- if (ANYOF_BITMAP_TEST(ret, i)) {
- ANYOF_BITMAP_CLEAR(ret, i);
- if (DEPENDS_SEMANTICS) {
- if (! remove_list) {
- remove_list = _new_invlist(2);
- }
- remove_list = add_cp_to_invlist(remove_list, i);
- }
- }
- else {
- ANYOF_BITMAP_SET(ret, i);
- prevvalue = value;
- value = i;
- }
- }
-
- /* And do the removal */
- if (DEPENDS_SEMANTICS) {
- if (remove_list) {
- _invlist_subtract(cp_list, remove_list, &cp_list);
- SvREFCNT_dec(remove_list);
- }
- }
- else {
- /* There is no overlap for non-/d, so just delete anything
- * below 256 */
- _invlist_intersection(cp_list, PL_AboveLatin1, &cp_list);
- }
- }
-
- stored = 256 - stored;
-
- /* Clear the invert flag since have just done it here */
- ANYOF_FLAGS(ret) &= ~ANYOF_INVERT;
- }
-
/* Folding in the bitmap is taken care of above, but not for locale (for
* which we have to wait to see what folding is in effect at runtime), and
* for some things not in the bitmap (only the upper latin folds in this
* FI'. */
if (! cp_list
&& ! unicode_alternate
- && SvCUR(listsv) == initial_listsv_len
+ && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
&& ! (ANYOF_FLAGS(ret) & (ANYOF_INVERT|ANYOF_UNICODE_ALL))
&& (((stored == 1 && ((! (ANYOF_FLAGS(ret) & ANYOF_LOCALE))
|| (! ANYOF_CLASS_TEST_ANY_SET(ret)))))
swash = NULL;
}
if (! cp_list
- && SvCUR(listsv) == initial_listsv_len
+ && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
&& ! unicode_alternate)
{
ARG_SET(ret, ANYOF_NONBITMAP_EMPTY);
* av[2] stores the multicharacter foldings, used later in
* regexec.c:S_reginclass().
* av[3] stores the cp_list inversion list for use in addition or
- * instead of av[0]; not used if av[1] isn't NULL
+ * instead of av[0]; used only if av[1] is NULL
* av[4] is set if any component of the class is from a user-defined
- * property; not used if av[1] isn't NULL */
+ * property; used only if av[1] is NULL */
AV * const av = newAV();
SV *rv;
- av_store(av, 0, (SvCUR(listsv) == initial_listsv_len)
- ? &PL_sv_undef
- : listsv);
+ av_store(av, 0, (HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION)
+ ? listsv
+ : &PL_sv_undef);
if (swash) {
av_store(av, 1, swash);
SvREFCNT_dec(cp_list);
}
return ret;
}
+#undef HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
/* reg_skipcomment()