X-Git-Url: https://git.exim.org/exim.git/blobdiff_plain/4fbcfc2ed4b301bc25f45931b0639eace3411cff..aa41d2de89da4bf43d52fd12a191742ff9b668a0:/src/src/pcre/pcre_maketables.c

diff --git a/src/src/pcre/pcre_maketables.c b/src/src/pcre/pcre_maketables.c
index d686383ac..3f6ec8415 100644
--- a/src/src/pcre/pcre_maketables.c
+++ b/src/src/pcre/pcre_maketables.c
@@ -1,4 +1,4 @@
-/* $Cambridge: exim/src/src/pcre/pcre_maketables.c,v 1.2 2005/08/08 10:22:14 ph10 Exp $ */
+/* $Cambridge: exim/src/src/pcre/pcre_maketables.c,v 1.3 2006/11/07 16:50:36 ph10 Exp $ */
 
 /*************************************************
 *      Perl-Compatible Regular Expressions       *
@@ -8,7 +8,7 @@
 and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
-           Copyright (c) 1997-2005 University of Cambridge
+           Copyright (c) 1997-2006 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -88,29 +88,22 @@ for (i = 0; i < 256; i++) *p++ = tolower(i);
 
 for (i = 0; i < 256; i++) *p++ = islower(i)? toupper(i) : tolower(i);
 
-/* Then the character class tables. Don't try to be clever and save effort
-on exclusive ones - in some locales things may be different. Note that the
-table for "space" includes everything "isspace" gives, including VT in the
-default locale. This makes it work for the POSIX class [:space:]. */
+/* Then the character class tables. Don't try to be clever and save effort on
+exclusive ones - in some locales things may be different. Note that the table
+for "space" includes everything "isspace" gives, including VT in the default
+locale. This makes it work for the POSIX class [:space:]. Note also that it is
+possible for a character to be alnum or alpha without being lower or upper,
+such as "male and female ordinals" (\xAA and \xBA) in the fr_FR locale (at
+least under Debian Linux's locales as of 12/2005). So we must test for alnum
+specially. */
 
 memset(p, 0, cbit_length);
 for (i = 0; i < 256; i++)
   {
-  if (isdigit(i))
-    {
-    p[cbit_digit  + i/8] |= 1 << (i&7);
-    p[cbit_word   + i/8] |= 1 << (i&7);
-    }
-  if (isupper(i))
-    {
-    p[cbit_upper  + i/8] |= 1 << (i&7);
-    p[cbit_word   + i/8] |= 1 << (i&7);
-    }
-  if (islower(i))
-    {
-    p[cbit_lower  + i/8] |= 1 << (i&7);
-    p[cbit_word   + i/8] |= 1 << (i&7);
-    }
+  if (isdigit(i)) p[cbit_digit  + i/8] |= 1 << (i&7);
+  if (isupper(i)) p[cbit_upper  + i/8] |= 1 << (i&7);
+  if (islower(i)) p[cbit_lower  + i/8] |= 1 << (i&7);
+  if (isalnum(i)) p[cbit_word   + i/8] |= 1 << (i&7);
   if (i == '_')   p[cbit_word   + i/8] |= 1 << (i&7);
   if (isspace(i)) p[cbit_space  + i/8] |= 1 << (i&7);
   if (isxdigit(i))p[cbit_xdigit + i/8] |= 1 << (i&7);
@@ -139,7 +132,9 @@ for (i = 0; i < 256; i++)
   meta-character, which in this sense is any character that terminates a run
   of data characters. */
 
-  if (strchr("*+?{^.$|()[", i) != 0) x += ctype_meta; *p++ = x; }
+  if (strchr("*+?{^.$|()[", i) != 0) x += ctype_meta;
+  *p++ = x;
+  }
 
 return yield;
 }