Suppress address randomization on Darwin. Do not merge into trunk.

[gnu-emacs] / src / regex.c

diff --git a/src/regex.c b/src/regex.c
index ff43fc85008d0183f70ca7f3f9ecfc43f08443e7..551f54dabc3266a15a4da82eac67d1cd2f4e159b 100644 (file)
--- a/src/regex.c
+++ b/src/regex.c
@@ -3,7 +3,7 @@
    internationalization features.)
 
    Copyright (C) 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001,
-                 2002, 2003, 2004, 2005, 2006, 2007, 2008
+                 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
                  Free Software Foundation, Inc.
 
    This program is free software; you can redistribute it and/or modify
@@ -117,6 +117,7 @@
    that make sense only in Emacs. */
 #ifdef emacs
 
+# include <setjmp.h>
 # include "lisp.h"
 # include "buffer.h"
 
@@ -146,12 +147,12 @@
 
 # define RE_MULTIBYTE_P(bufp) ((bufp)->multibyte)
 # define RE_TARGET_MULTIBYTE_P(bufp) ((bufp)->target_multibyte)
-# define RE_STRING_CHAR(p, s, multibyte) \
-  (multibyte ? (STRING_CHAR (p, s)) : (*(p)))
-# define RE_STRING_CHAR_AND_LENGTH(p, s, len, multibyte) \
-  (multibyte ? (STRING_CHAR_AND_LENGTH (p, s, len)) : ((len) = 1, *(p)))
+# define RE_STRING_CHAR(p, multibyte) \
+  (multibyte ? (STRING_CHAR (p)) : (*(p)))
+# define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) \
+  (multibyte ? (STRING_CHAR_AND_LENGTH (p, len)) : ((len) = 1, *(p)))
 
-# define RE_CHAR_TO_MULTIBYTE(c) unibyte_to_multibyte_table[(c)]
+# define RE_CHAR_TO_MULTIBYTE(c) UNIBYTE_TO_CHAR (c)
 
 # define RE_CHAR_TO_UNIBYTE(c) CHAR_TO_BYTE_SAFE (c)
 
@@ -165,7 +166,7 @@
        re_char *dtemp = (p) == (str2) ? (end1) : (p);                       \
        re_char *dlimit = ((p) > (str2) && (p) <= (end2)) ? (str2) : (str1); \
        while (dtemp-- > dlimit && !CHAR_HEAD_P (*dtemp));                   \
-       c = STRING_CHAR (dtemp, (p) - dtemp);                                \
+       c = STRING_CHAR (dtemp);                                             \
       }                                                                             \
     else                                                                    \
       {                                                                             \
@@ -179,7 +180,7 @@
 # define GET_CHAR_AFTER(c, p, len)             \
   do {                                         \
     if (target_multibyte)                      \
-      (c) = STRING_CHAR_AND_LENGTH (p, 0, len);        \
+      (c) = STRING_CHAR_AND_LENGTH (p, len);   \
     else                                       \
       {                                                \
        (c) = *p;                               \
@@ -301,11 +302,11 @@ enum syntaxcode { Swhitespace = 0, Sword = 1, Ssymbol = 2 };
 # define SAME_CHARSET_P(c1, c2) (1)
 # define MULTIBYTE_FORM_LENGTH(p, s) (1)
 # define PREV_CHAR_BOUNDARY(p, limit) ((p)--)
-# define STRING_CHAR(p, s) (*(p))
-# define RE_STRING_CHAR(p, s, multibyte) STRING_CHAR ((p), (s))
+# define STRING_CHAR(p) (*(p))
+# define RE_STRING_CHAR(p, multibyte) STRING_CHAR (p)
 # define CHAR_STRING(c, s) (*(s) = (c), 1)
-# define STRING_CHAR_AND_LENGTH(p, s, actual_len) ((actual_len) = 1, *(p))
-# define RE_STRING_CHAR_AND_LENGTH(p, s, len, multibyte) STRING_CHAR_AND_LENGTH ((p), (s), (len))
+# define STRING_CHAR_AND_LENGTH(p, actual_len) ((actual_len) = 1, *(p))
+# define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) STRING_CHAR_AND_LENGTH (p, len)
 # define RE_CHAR_TO_MULTIBYTE(c) (c)
 # define RE_CHAR_TO_UNIBYTE(c) (c)
 # define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
@@ -1756,7 +1757,7 @@ static int analyse_first _RE_ARGS ((re_char *p, re_char *pend,
   do {                                                                 \
     int len;                                                           \
     if (p == pend) return REG_EEND;                                    \
-    c = RE_STRING_CHAR_AND_LENGTH (p, pend - p, len, multibyte);       \
+    c = RE_STRING_CHAR_AND_LENGTH (p, len, multibyte);                 \
     p += len;                                                          \
   } while (0)
 
@@ -2085,7 +2086,7 @@ struct range_table_work_area
   } while (0)
 
 
-/* Both FROM and TO are mulitbyte characters.  */
+/* Both FROM and TO are multibyte characters.  */
 
 #define SETUP_MULTIBYTE_RANGE(work_area, FROM, TO)                        \
   do {                                                                    \
@@ -2400,7 +2401,7 @@ set_image_of_range_1 (work_area, start, end, translate)
 
 #endif /* emacs */
 
-/* Record the the image of the range start..end when passed through
+/* Record the image of the range start..end when passed through
    TRANSLATE.  This is not necessarily TRANSLATE(start)..TRANSLATE(end)
    and is not even necessarily contiguous.
    Normally we approximate it with the smallest contiguous range that contains
@@ -3064,6 +3065,13 @@ regex_compile (pattern, size, syntax, bufp)
                           don't need to handle them for multibyte.
                           They are distinguished by a negative wctype.  */
 
+                       /* Setup the gl_state object to its buffer-defined
+                          value.  This hardcodes the buffer-global
+                          syntax-table for ASCII chars, while the other chars
+                          will obey syntax-table properties.  It's not ideal,
+                          but it's the way it's been done until now.  */
+                       SETUP_SYNTAX_TABLE (BEGV, 0);
+
                        for (ch = 0; ch < 256; ++ch)
                          {
                            c = RE_CHAR_TO_MULTIBYTE (ch);
@@ -3797,7 +3805,7 @@ regex_compile (pattern, size, syntax, bufp)
 
                    if (c1 != c2 && (c1 = RE_CHAR_TO_UNIBYTE (c2)) >= 0)
                      c = c1;
-                 }                   
+                 }
                *b++ = c;
                len = 1;
              }
@@ -4082,8 +4090,7 @@ analyse_first (p, pend, fastmap, multibyte)
                     the corresponding multibyte character.  */
                  int c = RE_CHAR_TO_MULTIBYTE (p[1]);
 
-                 if (! CHAR_BYTE8_P (c))
-                   fastmap[CHAR_LEADING_CODE (c)] = 1;
+                 fastmap[CHAR_LEADING_CODE (c)] = 1;
                }
            }
          break;
@@ -4118,7 +4125,7 @@ analyse_first (p, pend, fastmap, multibyte)
          if (/* Any leading code can possibly start a character
                 which doesn't match the specified set of characters.  */
              not
-             || 
+             ||
              /* If we can match a character class, we can match any
                 multibyte characters.  */
              (CHARSET_RANGE_TABLE_EXISTS_P (&p[-2])
@@ -4138,7 +4145,7 @@ analyse_first (p, pend, fastmap, multibyte)
                   && match_any_multibyte_characters == false)
            {
              /* Set fastmap[I] to 1 where I is a leading code of each
-                multibyte characer in the range table. */
+                multibyte character in the range table. */
              int c, count;
              unsigned char lc1, lc2;
 
@@ -4540,8 +4547,7 @@ re_search_2 (bufp, str1, size1, str2, size2, startpos, range, regs, stop)
                      {
                        int buf_charlen;
 
-                       buf_ch = STRING_CHAR_AND_LENGTH (d, range - lim,
-                                                        buf_charlen);
+                       buf_ch = STRING_CHAR_AND_LENGTH (d, buf_charlen);
                        buf_ch = RE_TRANSLATE (translate, buf_ch);
                        if (fastmap[CHAR_LEADING_CODE (buf_ch)])
                          break;
@@ -4573,8 +4579,7 @@ re_search_2 (bufp, str1, size1, str2, size2, startpos, range, regs, stop)
                      {
                        int buf_charlen;
 
-                       buf_ch = STRING_CHAR_AND_LENGTH (d, range - lim,
-                                                        buf_charlen);
+                       buf_ch = STRING_CHAR_AND_LENGTH (d, buf_charlen);
                        if (fastmap[CHAR_LEADING_CODE (buf_ch)])
                          break;
                        range -= buf_charlen;
@@ -4591,12 +4596,9 @@ re_search_2 (bufp, str1, size1, str2, size2, startpos, range, regs, stop)
            }
          else                          /* Searching backwards.  */
            {
-             int room = (startpos >= size1
-                         ? size2 + size1 - startpos
-                         : size1 - startpos);
              if (multibyte)
                {
-                 buf_ch = STRING_CHAR (d, room);
+                 buf_ch = STRING_CHAR (d);
                  buf_ch = TRANSLATE (buf_ch);
                  if (! fastmap[CHAR_LEADING_CODE (buf_ch)])
                    goto advance;
@@ -4887,11 +4889,11 @@ mutually_exclusive_p (bufp, p1, p2)
       {
        register re_wchar_t c
          = (re_opcode_t) *p2 == endline ? '\n'
-         : RE_STRING_CHAR (p2 + 2, pend - p2 - 2, multibyte);
+         : RE_STRING_CHAR (p2 + 2, multibyte);
 
        if ((re_opcode_t) *p1 == exactn)
          {
-           if (c != RE_STRING_CHAR (p1 + 2, pend - p1 - 2, multibyte))
+           if (c != RE_STRING_CHAR (p1 + 2, multibyte))
              {
                DEBUG_PRINT3 ("  '%c' != '%c' => fast loop.\n", c, p1[2]);
                return 1;
@@ -5542,13 +5544,13 @@ re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
 
                PREFETCH ();
                if (multibyte)
-                 pat_ch = STRING_CHAR_AND_LENGTH (p, pend - p, pat_charlen);
+                 pat_ch = STRING_CHAR_AND_LENGTH (p, pat_charlen);
                else
                  {
                    pat_ch = RE_CHAR_TO_MULTIBYTE (*p);
                    pat_charlen = 1;
                  }
-               buf_ch = STRING_CHAR_AND_LENGTH (d, dend - d, buf_charlen);
+               buf_ch = STRING_CHAR_AND_LENGTH (d, buf_charlen);
 
                if (TRANSLATE (buf_ch) != pat_ch)
                  {
@@ -5570,7 +5572,7 @@ re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
                PREFETCH ();
                if (multibyte)
                  {
-                   pat_ch = STRING_CHAR_AND_LENGTH (p, pend - p, pat_charlen);
+                   pat_ch = STRING_CHAR_AND_LENGTH (p, pat_charlen);
                    pat_ch = RE_CHAR_TO_UNIBYTE (pat_ch);
                  }
                else
@@ -5610,7 +5612,7 @@ re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
            DEBUG_PRINT1 ("EXECUTING anychar.\n");
 
            PREFETCH ();
-           buf_ch = RE_STRING_CHAR_AND_LENGTH (d, dend - d, buf_charlen,
+           buf_ch = RE_STRING_CHAR_AND_LENGTH (d, buf_charlen,
                                                target_multibyte);
            buf_ch = TRANSLATE (buf_ch);
 
@@ -5658,7 +5660,7 @@ re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
              }
 
            PREFETCH ();
-           c = RE_STRING_CHAR_AND_LENGTH (d, dend - d, len, target_multibyte);
+           c = RE_STRING_CHAR_AND_LENGTH (d, len, target_multibyte);
            if (target_multibyte)
              {
                int c1;
@@ -6256,7 +6258,7 @@ re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
              UPDATE_SYNTAX_TABLE (charpos);
 #endif
              PREFETCH ();
-             c2 = RE_STRING_CHAR (d, dend - d, target_multibyte);
+             c2 = RE_STRING_CHAR (d, target_multibyte);
              s2 = SYNTAX (c2);
 
              /* Case 2: S2 is neither Sword nor Ssymbol. */
@@ -6309,7 +6311,7 @@ re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
              if (!AT_STRINGS_END (d))
                {
                  PREFETCH_NOLIMIT ();
-                 c2 = RE_STRING_CHAR (d, dend - d, target_multibyte);
+                 c2 = RE_STRING_CHAR (d, target_multibyte);
 #ifdef emacs
                  UPDATE_SYNTAX_TABLE_FORWARD (charpos + 1);
 #endif