EBCDIC: the non-printable characters are different.

[p5sagit/p5-mst-13.2.git] / regexec.c

diff --git a/regexec.c b/regexec.c
index df4a31b..8bd2284 100644 (file)
--- a/regexec.c
+++ b/regexec.c
@@ -67,7 +67,7 @@
  *
  ****    Alterations to Henry's code are...
  ****
- ****    Copyright (c) 1991-2001, Larry Wall
+ ****    Copyright (c) 1991-2002, Larry Wall
  ****
  ****    You may distribute under the terms of either the GNU General Public
  ****    License or the Artistic License, as specified in the README file.
@@ -140,13 +140,18 @@
     PL_regkind[(U8)OP(rn)] == EXACT || PL_regkind[(U8)OP(rn)] == REF \
 )
 
+/*
+  Search for mandatory following text node; for lookahead, the text must
+  follow but for lookbehind (rn->flags != 0) we skip to the next step.
+*/
 #define FIND_NEXT_IMPT(rn) STMT_START { \
     while (JUMPABLE(rn)) \
-       if (OP(rn) == SUSPEND || OP(rn) == IFMATCH || \
-           PL_regkind[(U8)OP(rn)] == CURLY) \
+       if (OP(rn) == SUSPEND || PL_regkind[(U8)OP(rn)] == CURLY) \
            rn = NEXTOPER(NEXTOPER(rn)); \
        else if (OP(rn) == PLUS) \
            rn = NEXTOPER(rn); \
+       else if (OP(rn) == IFMATCH) \
+           rn = (rn->flags == 0) ? NEXTOPER(NEXTOPER(rn)) : rn + ARG(rn); \
        else rn += NEXT_OFF(rn); \
 } STMT_END 
 
@@ -919,8 +924,11 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, char *strend, char *sta
                STRLEN skip = do_utf8 ? UTF8SKIP(s) : 1;
 
                if (reginclass(c, (U8*)s, do_utf8) ||
-                   (ANYOF_UNICODE_FOLD_SHARP_S(c, s, strend) &&
-                    (skip = 2))) {
+                   (ANYOF_FOLD_SHARP_S(c, s, strend) &&
+                    /* The assignment of 2 is intentional:
+                     * for the sharp s, the skip is 2. */
+                    (skip = SHARP_S_SKIP)
+                    )) {
                    if (tmp && (norun || regtry(prog, s)))
                        goto got_it;
                    else
@@ -1016,12 +1024,12 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, char *strend, char *sta
                        c = utf8_to_uvchr((U8*)s, &len);
 
                        /* Handle some of the three Greek sigmas cases.
-                         * Note that not all the possible combinations
-                         * are handled here: some of them are handled
-                         * handled by the standard folding rules, and
-                         * some of them (the character class or ANYOF
-                         * cases) are handled during compiletime in
-                         * regexec.c:S_regclass(). */
+                        * Note that not all the possible combinations
+                        * are handled here: some of them are handled
+                        * by the standard folding rules, and some of
+                        * them (the character class or ANYOF cases)
+                        * are handled during compiletime in
+                        * regexec.c:S_regclass(). */
                        if (c == (UV)UNICODE_GREEK_CAPITAL_LETTER_SIGMA ||
                            c == (UV)UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA)
                            c = (UV)UNICODE_GREEK_SMALL_LETTER_SIGMA;
@@ -1038,7 +1046,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, char *strend, char *sta
                             if ( f != c
                                  && (f == c1 || f == c2)
                                  && (ln == foldlen ||
-                                     !ibcmp_utf8((char *)foldbuf,
+                                     !ibcmp_utf8((char *) foldbuf,
                                                  (char **)0, foldlen, do_utf8,
                                                  m,
                                                  (char **)0, ln,      UTF))
@@ -2327,7 +2335,7 @@ S_regmatch(pTHX_ regnode *prog)
                        if (l >= PL_regeol)
                             sayNO;
                        if (NATIVE_TO_UNI(*(U8*)s) !=
-                           utf8_to_uvchr((U8*)l, &ulen))
+                           utf8_to_uvuni((U8*)l, &ulen))
                             sayNO;
                        l += ulen;
                        s ++;
@@ -2339,7 +2347,7 @@ S_regmatch(pTHX_ regnode *prog)
                        if (l >= PL_regeol)
                            sayNO;
                        if (NATIVE_TO_UNI(*((U8*)l)) !=
-                           utf8_to_uvchr((U8*)s, &ulen))
+                           utf8_to_uvuni((U8*)s, &ulen))
                            sayNO;
                        s += ulen;
                        l ++;
@@ -2373,8 +2381,20 @@ S_regmatch(pTHX_ regnode *prog)
                char *e = PL_regeol;
 
                if (ibcmp_utf8(s, 0,  ln, do_utf8,
-                              l, &e, 0,  UTF))
-                    sayNO;
+                              l, &e, 0,  UTF)) {
+                    /* One more case for the sharp s:
+                     * pack("U0U*", 0xDF) =~ /ss/i,
+                     * the 0xC3 0x9F are the UTF-8
+                     * byte sequence for the U+00DF. */
+                    if (!(do_utf8 &&
+                          toLOWER(s[0]) == 's' &&
+                          ln >= 2 &&
+                          toLOWER(s[1]) == 's' &&
+                          (U8)l[0] == 0xC3 &&
+                          e - l >= 2 &&
+                          (U8)l[1] == 0x9F))
+                         sayNO;
+               }
                locinput = e;
                nextchr = UCHARAT(locinput);
                break;
@@ -2422,8 +2442,8 @@ S_regmatch(pTHX_ regnode *prog)
            /* If we might have the case of the German sharp s
             * in a casefolding Unicode character class. */
 
-           if (ANYOF_UNICODE_FOLD_SHARP_S(scan, locinput, PL_regeol)) {
-                locinput += 2;
+           if (ANYOF_FOLD_SHARP_S(scan, locinput, PL_regeol)) {
+                locinput += SHARP_S_SKIP;
                 nextchr = UCHARAT(locinput);
            }
            else