* "One Ring to rule them all, One Ring to find them..."
*/
+/* This file contains functions for executing a regular expression. See
+ * also regcomp.c which funnily enough, contains functions for compiling
+ * a regular expression.
+ *
+ * This file is also copied at build time to ext/re/re_exec.c, where
+ * it's built with -DPERL_EXT_RE_BUILD -DPERL_EXT_RE_DEBUG -DPERL_EXT.
+ * This causes the main functions to be compiled under new names and with
+ * debugging support added, which makes "use re 'debug'" work.
+
+ */
+
/* NOTE: this is derived from Henry Spencer's regexp code, and should not
* confused with the original package (see point 3 below). Thanks, Henry!
*/
**** Alterations to Henry's code are...
****
**** Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
- **** 2000, 2001, 2002, 2003, 2004, by Larry Wall and others
+ **** 2000, 2001, 2002, 2003, 2004, 2005, by Larry Wall and others
****
**** You may distribute under the terms of either the GNU General Public
**** License or the Artistic License, as specified in the README file.
I32 ml_anch;
register char *other_last = Nullch; /* other substr checked before this */
char *check_at = Nullch; /* check substr found at this pos */
+ I32 multiline = prog->reganch & PMf_MULTILINE;
#ifdef DEBUGGING
char *i_strpos = strpos;
SV *dsv = PERL_DEBUG_PAD_ZERO(0);
if (prog->reganch & ROPT_ANCH) { /* Match at beg-of-str or after \n */
ml_anch = !( (prog->reganch & ROPT_ANCH_SINGLE)
|| ( (prog->reganch & ROPT_ANCH_BOL)
- && !PL_multiline ) ); /* Check after \n? */
+ && !multiline ) ); /* Check after \n? */
if (!ml_anch) {
if ( !(prog->reganch & (ROPT_ANCH_GPOS /* Checked by the caller */
else if (prog->reganch & ROPT_CANY_SEEN)
s = fbm_instr((U8*)(s + start_shift),
(U8*)(strend - end_shift),
- check, PL_multiline ? FBMrf_MULTILINE : 0);
+ check, multiline ? FBMrf_MULTILINE : 0);
else
s = fbm_instr(HOP3(s, start_shift, strend),
HOP3(strend, -end_shift, strbeg),
- check, PL_multiline ? FBMrf_MULTILINE : 0);
+ check, multiline ? FBMrf_MULTILINE : 0);
/* Update the count-of-usability, remove useless subpatterns,
unshift s. */
HOP3(HOP3(last1, prog->anchored_offset, strend)
+ SvCUR(must), -(SvTAIL(must)!=0), strbeg),
must,
- PL_multiline ? FBMrf_MULTILINE : 0
+ multiline ? FBMrf_MULTILINE : 0
);
DEBUG_r(PerlIO_printf(Perl_debug_log,
"%s anchored substr `%s%.*s%s'%s",
s = fbm_instr((unsigned char*)s,
(unsigned char*)last + SvCUR(must)
- (SvTAIL(must)!=0),
- must, PL_multiline ? FBMrf_MULTILINE : 0);
+ must, multiline ? FBMrf_MULTILINE : 0);
DEBUG_r(PerlIO_printf(Perl_debug_log, "%s floating substr `%s%.*s%s'%s",
(s ? "Found" : "Contradicts"),
PL_colors[0],
char *m;
STRLEN ln;
STRLEN lnc;
+ register STRLEN uskip;
unsigned int c1;
unsigned int c2;
char *e;
switch (OP(c)) {
case ANYOF:
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if ((ANYOF_FLAGS(c) & ANYOF_UNICODE) ||
!UTF8_IS_INVARIANT((U8)s[0]) ?
reginclass(c, (U8*)s, 0, do_utf8) :
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
if (UTF) {
STRLEN ulen1, ulen2;
U8 *sm = (U8 *) m;
- U8 tmpbuf1[UTF8_MAXLEN_UCLC+1];
- U8 tmpbuf2[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
+ U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
to_utf8_lower((U8*)m, tmpbuf1, &ulen1);
to_utf8_upper((U8*)m, tmpbuf2, &ulen2);
- c1 = utf8n_to_uvchr(tmpbuf1, UTF8_MAXLEN_UCLC,
+ c1 = utf8n_to_uvchr(tmpbuf1, UTF8_MAXBYTES_CASE,
0, ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
- c2 = utf8n_to_uvchr(tmpbuf2, UTF8_MAXLEN_UCLC,
+ c2 = utf8n_to_uvchr(tmpbuf2, UTF8_MAXBYTES_CASE,
0, ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
lnc = 0;
while (sm < ((U8 *) m + ln)) {
if (do_utf8) {
UV c, f;
- U8 tmpbuf [UTF8_MAXLEN+1];
- U8 foldbuf[UTF8_MAXLEN_FOLD+1];
+ U8 tmpbuf [UTF8_MAXBYTES+1];
+ U8 foldbuf[UTF8_MAXBYTES_CASE+1];
STRLEN len, foldlen;
if (c1 == c2) {
/* Upper and lower of 1st char are equal -
* probably not a "letter". */
while (s <= e) {
- c = utf8n_to_uvchr((U8*)s, UTF8_MAXLEN, &len,
+ c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
if ( c == c1
}
else {
while (s <= e) {
- c = utf8n_to_uvchr((U8*)s, UTF8_MAXLEN, &len,
+ c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
tmp = ((OP(c) == BOUND ?
isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0);
LOAD_UTF8_CHARCLASS(alnum,"a");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (tmp == !(OP(c) == BOUND ?
swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) :
isALNUM_LC_utf8((U8*)s)))
if ((norun || regtry(prog, s)))
goto got_it;
}
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
tmp = ((OP(c) == NBOUND ?
isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0);
LOAD_UTF8_CHARCLASS(alnum,"a");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (tmp == !(OP(c) == NBOUND ?
swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) :
isALNUM_LC_utf8((U8*)s)))
tmp = !tmp;
else if ((norun || regtry(prog, s)))
goto got_it;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case ALNUM:
if (do_utf8) {
LOAD_UTF8_CHARCLASS(alnum,"a");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case ALNUML:
PL_reg_flags |= RF_tainted;
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (isALNUM_LC_utf8((U8*)s)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case NALNUM:
if (do_utf8) {
LOAD_UTF8_CHARCLASS(alnum,"a");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (!swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case NALNUML:
PL_reg_flags |= RF_tainted;
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (!isALNUM_LC_utf8((U8*)s)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case SPACE:
if (do_utf8) {
LOAD_UTF8_CHARCLASS(space," ");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case SPACEL:
PL_reg_flags |= RF_tainted;
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (*s == ' ' || isSPACE_LC_utf8((U8*)s)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case NSPACE:
if (do_utf8) {
LOAD_UTF8_CHARCLASS(space," ");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (!(*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8))) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case NSPACEL:
PL_reg_flags |= RF_tainted;
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (!(*s == ' ' || isSPACE_LC_utf8((U8*)s))) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case DIGIT:
if (do_utf8) {
LOAD_UTF8_CHARCLASS(digit,"0");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (swash_fetch(PL_utf8_digit,(U8*)s, do_utf8)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case DIGITL:
PL_reg_flags |= RF_tainted;
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (isDIGIT_LC_utf8((U8*)s)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case NDIGIT:
if (do_utf8) {
LOAD_UTF8_CHARCLASS(digit,"0");
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (!swash_fetch(PL_utf8_digit,(U8*)s, do_utf8)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
case NDIGITL:
PL_reg_flags |= RF_tainted;
if (do_utf8) {
- while (s < strend) {
+ while (s + (uskip = UTF8SKIP(s)) <= strend) {
if (!isDIGIT_LC_utf8((U8*)s)) {
if (tmp && (norun || regtry(prog, s)))
goto got_it;
}
else
tmp = 1;
- s += UTF8SKIP(s);
+ s += uskip;
}
}
else {
char *scream_olds;
SV* oreplsv = GvSV(PL_replgv);
bool do_utf8 = DO_UTF8(sv);
+ I32 multiline = prog->reganch & PMf_MULTILINE;
#ifdef DEBUGGING
SV *dsv0 = PERL_DEBUG_PAD_ZERO(0);
SV *dsv1 = PERL_DEBUG_PAD_ZERO(1);
if (prog->reganch & (ROPT_ANCH & ~ROPT_ANCH_GPOS)) {
if (s == startpos && regtry(prog, startpos))
goto got_it;
- else if (PL_multiline || (prog->reganch & ROPT_IMPLICIT)
+ else if (multiline || (prog->reganch & ROPT_IMPLICIT)
|| (prog->reganch & ROPT_ANCH_MBOL)) /* XXXX SBOL? */
{
char *end;
end_shift, &scream_pos, 0))
: (s = fbm_instr((unsigned char*)HOP3(s, back_min, strend),
(unsigned char*)strend, must,
- PL_multiline ? FBMrf_MULTILINE : 0))) ) {
+ multiline ? FBMrf_MULTILINE : 0))) ) {
/* we may be pointing at the wrong string */
if ((flags & REXEC_SCREAM) && RX_MATCH_COPIED(prog))
s = strbeg + (s - SvPVX(sv));
if (SvTAIL(float_real)) {
if (memEQ(strend - len + 1, little, len - 1))
last = strend - len + 1;
- else if (!PL_multiline)
+ else if (!multiline)
last = memEQ(strend - len, little, len)
? strend - len : Nullch;
else
switch (OP(scan)) {
case BOL:
- if (locinput == PL_bostr || (PL_multiline &&
- (nextchr || locinput < PL_regeol) && locinput[-1] == '\n') )
+ if (locinput == PL_bostr)
{
/* regtill = regbol; */
break;
break;
sayNO;
case EOL:
- if (PL_multiline)
- goto meol;
- else
goto seol;
case MEOL:
- meol:
if ((nextchr || locinput < PL_regeol) && nextchr != '\n')
sayNO;
break;
if (l >= PL_regeol)
sayNO;
if (NATIVE_TO_UNI(*(U8*)s) !=
- utf8n_to_uvuni((U8*)l, UTF8_MAXLEN, &ulen,
+ utf8n_to_uvuni((U8*)l, UTF8_MAXBYTES, &ulen,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY))
sayNO;
if (l >= PL_regeol)
sayNO;
if (NATIVE_TO_UNI(*((U8*)l)) !=
- utf8n_to_uvuni((U8*)s, UTF8_MAXLEN, &ulen,
+ utf8n_to_uvuni((U8*)s, UTF8_MAXBYTES, &ulen,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY))
sayNO;
*/
if (OP(scan) == REFF) {
STRLEN ulen1, ulen2;
- U8 tmpbuf1[UTF8_MAXLEN_UCLC+1];
- U8 tmpbuf2[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
+ U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
while (s < e) {
if (l >= PL_regeol)
sayNO;
CHECKPOINT lastcp;
/* We suppose that the next guy does not need
- backtracking: in particular, it is of constant length,
+ backtracking: in particular, it is of constant non-zero length,
and has no parenths to influence future backrefs. */
ln = ARG1(scan); /* min to match */
n = ARG2(scan); /* max to match */
minmod = 0;
if (ln && regrepeat_hard(scan, ln, &l) < ln)
sayNO;
- /* if we matched something zero-length we don't need to
- backtrack - capturing parens are already defined, so
- the caveat in the maximal case doesn't apply
-
- XXXX if ln == 0, we can redo this check first time
- through the following loop
- */
- if (ln && l == 0)
- n = ln; /* don't backtrack */
locinput = PL_reginput;
if (HAS_TEXT(next) || JUMPABLE(next)) {
regnode *text_node = next;
c1 = c2 = -1000;
assume_ok_MM:
REGCP_SET(lastcp);
- /* This may be improved if l == 0. */
- while (n >= ln || (n == REG_INFTY && ln > 0 && l)) { /* ln overflow ? */
+ while (n >= ln || (n == REG_INFTY && ln > 0)) { /* ln overflow ? */
/* If it could work, try it. */
if (c1 == -1000 ||
UCHARAT(PL_reginput) == c1 ||
}
else {
n = regrepeat_hard(scan, n, &l);
- /* if we matched something zero-length we don't need to
- backtrack, unless the minimum count is zero and we
- are capturing the result - in that case the capture
- being defined or not may affect later execution
- */
- if (n != 0 && l == 0 && !(paren && ln == 0))
- ln = n; /* don't backtrack */
locinput = PL_reginput;
DEBUG_r(
PerlIO_printf(Perl_debug_log,
else { /* UTF */
if (OP(text_node) == EXACTF || OP(text_node) == REFF) {
STRLEN ulen1, ulen2;
- U8 tmpbuf1[UTF8_MAXLEN_UCLC+1];
- U8 tmpbuf2[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
+ U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
to_utf8_lower((U8*)s, tmpbuf1, &ulen1);
to_utf8_upper((U8*)s, tmpbuf2, &ulen2);
- c1 = utf8n_to_uvuni(tmpbuf1, UTF8_MAXLEN, 0,
+ c1 = utf8n_to_uvuni(tmpbuf1, UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
- c2 = utf8n_to_uvuni(tmpbuf2, UTF8_MAXLEN, 0,
+ c2 = utf8n_to_uvuni(tmpbuf2, UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
}
else {
- c2 = c1 = utf8n_to_uvchr(s, UTF8_MAXLEN, 0,
+ c2 = c1 = utf8n_to_uvchr(s, UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
}
* utf8_distance(old, locinput) */
while (locinput <= e &&
utf8n_to_uvchr((U8*)locinput,
- UTF8_MAXLEN, &len,
+ UTF8_MAXBYTES, &len,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY) != (UV)c1) {
locinput += len;
* utf8_distance(old, locinput) */
while (locinput <= e) {
UV c = utf8n_to_uvchr((U8*)locinput,
- UTF8_MAXLEN, &len,
+ UTF8_MAXBYTES, &len,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
if (c == (UV)c1 || c == (UV)c2)
if (c1 != -1000) {
if (do_utf8)
c = utf8n_to_uvchr((U8*)PL_reginput,
- UTF8_MAXLEN, 0,
+ UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
else
n = regrepeat(scan, n);
locinput = PL_reginput;
if (ln < n && PL_regkind[(U8)OP(next)] == EOL &&
- ((!PL_multiline && OP(next) != MEOL) ||
+ (OP(next) != MEOL ||
OP(next) == SEOL || OP(next) == EOS))
{
ln = n; /* why back off? */
if (c1 != -1000) {
if (do_utf8)
c = utf8n_to_uvchr((U8*)PL_reginput,
- UTF8_MAXLEN, 0,
+ UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
else
if (c1 != -1000) {
if (do_utf8)
c = utf8n_to_uvchr((U8*)PL_reginput,
- UTF8_MAXLEN, 0,
+ UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
else
/*
- regrepeat_hard - repeatedly match something, report total lenth and length
*
- * The repeater is supposed to have constant length.
+ * The repeater is supposed to have constant non-zero length.
*/
STATIC I32
STRLEN plen;
if (do_utf8 && !UTF8_IS_INVARIANT(c))
- c = utf8n_to_uvchr(p, UTF8_MAXLEN, &len,
+ c = utf8n_to_uvchr(p, UTF8_MAXBYTES, &len,
ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
plen = lenp ? *lenp : UNISKIP(NATIVE_TO_UNI(c));
}
}
if (!match) {
- U8 tmpbuf[UTF8_MAXLEN_FOLD+1];
+ U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
STRLEN tmplen;
to_utf8_fold(p, tmpbuf, &tmplen);
{
SV* sv;
if (prog->float_substr && !prog->float_utf8) {
- prog->float_utf8 = sv = NEWSV(117, 0);
- SvSetSV(sv, prog->float_substr);
+ prog->float_utf8 = sv = newSVsv(prog->float_substr);
sv_utf8_upgrade(sv);
if (SvTAIL(prog->float_substr))
SvTAIL_on(sv);
prog->check_utf8 = sv;
}
if (prog->anchored_substr && !prog->anchored_utf8) {
- prog->anchored_utf8 = sv = NEWSV(118, 0);
- SvSetSV(sv, prog->anchored_substr);
+ prog->anchored_utf8 = sv = newSVsv(prog->anchored_substr);
sv_utf8_upgrade(sv);
if (SvTAIL(prog->anchored_substr))
SvTAIL_on(sv);
{
SV* sv;
if (prog->float_utf8 && !prog->float_substr) {
- prog->float_substr = sv = NEWSV(117, 0);
- SvSetSV(sv, prog->float_utf8);
+ prog->float_substr = sv = newSVsv(prog->float_utf8);
if (sv_utf8_downgrade(sv, TRUE)) {
if (SvTAIL(prog->float_utf8))
SvTAIL_on(sv);
prog->check_substr = sv;
}
if (prog->anchored_utf8 && !prog->anchored_substr) {
- prog->anchored_substr = sv = NEWSV(118, 0);
- SvSetSV(sv, prog->anchored_utf8);
+ prog->anchored_substr = sv = newSVsv(prog->anchored_utf8);
if (sv_utf8_downgrade(sv, TRUE)) {
if (SvTAIL(prog->anchored_utf8))
SvTAIL_on(sv);