Fix caseless character class bug for characters within a range that have more

than one other case.


git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1438 2f5784b3-3f2a-0410-8824-cb99058d5e15

6 files changed, 68 insertions, 7 deletions

diff --git a/ChangeLog b/ChangeLog
index e058884..506b2c7 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -53,6 +53,12 @@ Version 8.35-RC1 xx-xxxx-201x
 
 11. Empty match is not possible, when the minimum length is greater than zero,
     and there is no \K in the pattern. Remove these unnecessary checks form JIT.
+    
+12. In a caseless character class with UCP support, when a character with more
+    than one alternative case was not the first character of a range, not all
+    the alternative cases were added to the class. For example, s and \x{17f}
+    are both alternative cases for S: the class [RST] was handled correctly,
+    but [R-T] was not.
 
 
 Version 8.34 15-December-2013
diff --git a/pcre_compile.c b/pcre_compile.c
index a3d2870..dd065a3 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -6,7 +6,7 @@
 and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
-           Copyright (c) 1997-2013 University of Cambridge
+           Copyright (c) 1997-2014 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -4077,12 +4077,16 @@ for (c = *cptr; c <= d; c++)
 
 if (c > d) return -1;  /* Reached end of range */
 
+/* Found a character that has a single other case. Search for the end of the 
+range, which is either the end of the input range, or a character that has zero 
+or more than one other cases. */
+
 *ocptr = othercase;
 next = othercase + 1;
 
 for (++c; c <= d; c++)
   {
-  if (UCD_OTHERCASE(c) != next) break;
+  if ((co = UCD_CASESET(c)) != 0 || UCD_OTHERCASE(c) != next) break;
   next++;
   }
 
@@ -4138,7 +4142,7 @@ if ((options & PCRE_CASELESS) != 0)
 
     options &= ~PCRE_CASELESS;   /* Remove for recursive calls */
     c = start;
-
+    
     while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0)
       {
       /* Handle a single character that has more than one other case. */
@@ -4201,9 +4205,9 @@ for (c = start; c <= classbits_end; c++)
 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
 if (start <= 0xff) start = 0xff + 1;
 
-if (end >= start) {
+if (end >= start) 
+  {
   pcre_uchar *uchardata = *uchardptr;
-
 #ifdef SUPPORT_UTF
   if ((options & PCRE_UTF8) != 0)  /* All UTFs use the same flag bit */
     {
diff --git a/testdata/testinput6 b/testdata/testinput6
index 1e450be..7a6a53f 100644
--- a/testdata/testinput6
+++ b/testdata/testinput6
@@ -1484,4 +1484,13 @@
     \x{a1}\x{a7}  
     \x{37e} 
 
+/[RST]+/8iW
+    Ss\x{17f}
+    
+/[R-T]+/8iW 
+    Ss\x{17f}
+
+/[q-u]+/8iW 
+    Ss\x{17f}
+
 /-- End of testinput6 --/
diff --git a/testdata/testinput7 b/testdata/testinput7
index 9d14543..6bd0586 100644
--- a/testdata/testinput7
+++ b/testdata/testinput7
@@ -829,4 +829,10 @@ of case for anything other than the ASCII letters. --/
 
 /\d+\s{0,5}=\s*\S?=\w{0,4}\W*/8WBZ
 
+/[RST]+/8iWBZ
+    
+/[R-T]+/8iWBZ 
+
+/[Q-U]+/8iWBZ 
+
 /-- End of testinput7 --/
diff --git a/testdata/testoutput6 b/testdata/testoutput6
index 6c42fce..f355e60 100644
--- a/testdata/testoutput6
+++ b/testdata/testoutput6
@@ -2445,4 +2445,16 @@ No match
     \x{37e} 
 No match
 
+/[RST]+/8iW
+    Ss\x{17f}
+ 0: Ss\x{17f}
+    
+/[R-T]+/8iW 
+    Ss\x{17f}
+ 0: Ss\x{17f}
+
+/[q-u]+/8iW 
+    Ss\x{17f}
+ 0: Ss\x{17f}
+
 /-- End of testinput6 --/
diff --git a/testdata/testoutput7 b/testdata/testoutput7
index 8970d7b..c64e049 100644
--- a/testdata/testoutput7
+++ b/testdata/testoutput7
@@ -124,7 +124,7 @@ No match
 /[z-\x{100}]/8iDZ 
 ------------------------------------------------------------------
         Bra
-        [Zz-\xff\x{39c}\x{3bc}\x{1e9e}\x{178}\x{100}-\x{101}]
+        [Zz-\xff\x{39c}\x{3bc}\x{212b}\x{1e9e}\x{212b}\x{178}\x{100}-\x{101}]
         Ket
         End
 ------------------------------------------------------------------
@@ -162,7 +162,7 @@ No match
 /[z-\x{100}]/8DZi
 ------------------------------------------------------------------
         Bra
-        [Zz-\xff\x{39c}\x{3bc}\x{1e9e}\x{178}\x{100}-\x{101}]
+        [Zz-\xff\x{39c}\x{3bc}\x{212b}\x{1e9e}\x{212b}\x{178}\x{100}-\x{101}]
         Ket
         End
 ------------------------------------------------------------------
@@ -2263,4 +2263,28 @@ No match
         End
 ------------------------------------------------------------------
 
+/[RST]+/8iWBZ
+------------------------------------------------------------------
+        Bra
+        [R-Tr-t\x{17f}]++
+        Ket
+        End
+------------------------------------------------------------------
+    
+/[R-T]+/8iWBZ 
+------------------------------------------------------------------
+        Bra
+        [R-Tr-t\x{17f}]++
+        Ket
+        End
+------------------------------------------------------------------
+
+/[Q-U]+/8iWBZ 
+------------------------------------------------------------------
+        Bra
+        [Q-Uq-u\x{17f}]++
+        Ket
+        End
+------------------------------------------------------------------
+
 /-- End of testinput7 --/