Fix bug in Oniguruma \g support.

git-svn-id: svn://vcs.exim.org/pcre/code/trunk@334 2f5784b3-3f2a-0410-8824-cb99058d5e15

3 files changed, 21 insertions, 4 deletions

diff --git a/pcre_compile.c b/pcre_compile.c
index aed8190..f7be568 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -1882,7 +1882,7 @@ while ((ptr = (uschar *)find_recurse(ptr, utf8)) != NULL)
 
   /* See if this recursion is on the forward reference list. If so, adjust the
   reference. */
-
+  
   for (hc = save_hwm; hc < cd->hwm; hc += LINK_SIZE)
     {
     offset = GET(hc, 0);
@@ -2458,7 +2458,7 @@ for (;; ptr++)
   /* Get next byte in the pattern */
 
   c = *ptr;
-
+  
   /* If we are in the pre-compile phase, accumulate the length used for the
   previous cycle of this loop. */
 
@@ -4138,7 +4138,7 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */
     bravalue = OP_CBRA;
     save_hwm = cd->hwm;
     reset_bracount = FALSE;
-
+    
     /* First deal with various "verbs" that can be introduced by '*'. */
 
     if (*(++ptr) == '*' && (cd->ctypes[ptr[1]] & ctype_letter) != 0)
@@ -5127,11 +5127,12 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */
       -ESC_g is returned only for these cases. So we don't need to check for <
       or ' if the value is -ESC_g. For the Perl syntax \g{n} the value is
       -ESC_REF+n, and for the Perl syntax \g{name} the result is -ESC_k (as
-      that is a synonym). */
+      that is a synonym for a named back reference). */
       
       if (-c == ESC_g)
         {
         const uschar *p;
+        save_hwm = cd->hwm;   /* Normally this is set when '(' is read */ 
         terminator = (*(++ptr) == '<')? '>' : '\'';
         
         /* These two statements stop the compiler for warning about possibly
diff --git a/testdata/testinput2 b/testdata/testinput2
index 4ef241d..a4a0326 100644
--- a/testdata/testinput2
+++ b/testdata/testinput2
@@ -2643,4 +2643,10 @@ a random value. /Ix
     XaaX
     XAAX 
 
+/(?=(?<regex>(?#simplesyntax)\$(?<name>[a-zA-Z_\x{7f}-\x{ff}][a-zA-Z0-9_\x{7f}-\x{ff}]*)(?:\[(?<index>[a-zA-Z0-9_\x{7f}-\x{ff}]+|\$\g<name>)\]|->\g<name>(\(.*?\))?)?|(?#simple syntax withbraces)\$\{(?:\g<name>(?<indices>\[(?:\g<index>|'(?:\\.|[^'\\])*'|"(?:\g<regex>|\\.|[^"\\])*")\])?|\g<complex>|\$\{\g<complex>\})\}|(?#complexsyntax)\{(?<complex>\$(?<segment>\g<name>(\g<indices>*|\(.*?\))?)(?:->\g<segment>)*|\$\g<complex>|\$\{\g<complex>\})\}))\{/
+
+/(?<n>a|b|c)\g<n>*/
+   abc
+   accccbbb 
+
 / End of testinput2 /
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index ae25cfa..c2e41f7 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -9505,4 +9505,14 @@ No match
  0: AA
  1: A
 
+/(?=(?<regex>(?#simplesyntax)\$(?<name>[a-zA-Z_\x{7f}-\x{ff}][a-zA-Z0-9_\x{7f}-\x{ff}]*)(?:\[(?<index>[a-zA-Z0-9_\x{7f}-\x{ff}]+|\$\g<name>)\]|->\g<name>(\(.*?\))?)?|(?#simple syntax withbraces)\$\{(?:\g<name>(?<indices>\[(?:\g<index>|'(?:\\.|[^'\\])*'|"(?:\g<regex>|\\.|[^"\\])*")\])?|\g<complex>|\$\{\g<complex>\})\}|(?#complexsyntax)\{(?<complex>\$(?<segment>\g<name>(\g<indices>*|\(.*?\))?)(?:->\g<segment>)*|\$\g<complex>|\$\{\g<complex>\})\}))\{/
+
+/(?<n>a|b|c)\g<n>*/
+   abc
+ 0: abc
+ 1: a
+   accccbbb 
+ 0: accccbbb
+ 1: a
+
 / End of testinput2 /