From a5ca95c10cb6da41568b09b7c646441773147535 Mon Sep 17 00:00:00 2001
From: Tim Waugh <twaugh@fedoraproject.org>
Date: Thu, 16 Dec 2004 17:50:47 +0000
Subject: [PATCH] - Bypass kwset matching when ignoring case and processing
 multibyte input     (bug #143079).

---
 grep-2.5.1-egf-speedup.patch | 33 +++++++++++++++++++--------------
 grep.spec                    |  4 ++++
 2 files changed, 23 insertions(+), 14 deletions(-)

diff --git a/grep-2.5.1-egf-speedup.patch b/grep-2.5.1-egf-speedup.patch
index 87e4dee..0e9a5eb 100644
--- a/grep-2.5.1-egf-speedup.patch
+++ b/grep-2.5.1-egf-speedup.patch
@@ -1,5 +1,5 @@
---- grep-2.5.1/src/search.c	2004-12-14 15:55:21.257729918 +0000
-+++ grep-2.5.1/src/search.c	2004-12-16 16:53:01.110921088 +0000
+--- grep-2.5.1/src/search.c	2004-12-16 16:53:01.110921088 +0000
++++ grep-2.5.1/src/search.c	2004-12-16 17:46:57.039678304 +0000
 @@ -39,6 +39,9 @@
  #ifdef HAVE_LIBPCRE
  # include <pcre.h>
@@ -125,7 +125,7 @@
  #endif /* MBS_SUPPORT */
  
    buflim = buf + size;
-@@ -373,18 +337,56 @@
+@@ -373,21 +337,60 @@
  	  if (kwset)
  	    {
  	      /* Find a possible match using the KWset matcher. */
@@ -134,8 +134,8 @@
 +	      size_t bytes_left = 0;
 +#endif /* MBS_SUPPORT */
 +	      size_t offset;
-+	      /* kwsexec doesn't work with match_icase and multibyte input. */
 +#ifdef MBS_SUPPORT
++	      /* kwsexec doesn't work with match_icase and multibyte input. */
 +	      if (match_icase && MB_CUR_MAX > 1)
 +		/* Avoid kwset */
 +		offset = 0;
@@ -184,8 +184,13 @@
 +#endif /* MBS_SUPPORT */
  	      while (beg > buf && beg[-1] != eol)
  		--beg;
- 	      if (kwsm.index < kwset_exact_matches)
-@@ -395,13 +397,47 @@
+-	      if (kwsm.index < kwset_exact_matches)
++	      if (!(match_icase && MB_CUR_MAX > 1) &&
++		  (kwsm.index < kwset_exact_matches))
+ 		goto success_in_beg_and_end;
+ 	      if (dfaexec (&dfa, beg, end - beg, &backref) == (size_t) -1)
+ 		continue;
+@@ -395,13 +398,47 @@
  	  else
  	    {
  	      /* No good fixed strings; start with DFA. */
@@ -233,7 +238,7 @@
  	      while (beg > buf && beg[-1] != eol)
  		--beg;
  	    }
-@@ -469,15 +505,6 @@
+@@ -469,15 +506,6 @@
      } /* for (beg = end ..) */
  
   failure:
@@ -249,7 +254,7 @@
    return (size_t) -1;
  
   success_in_beg_and_end:
-@@ -486,15 +513,6 @@
+@@ -486,15 +514,6 @@
    /* FALLTHROUGH */
  
   success_in_start_and_len:
@@ -265,7 +270,7 @@
    *match_size = len;
    return start;
  }
-@@ -504,6 +522,7 @@
+@@ -504,6 +523,7 @@
  {
    char const *beg, *lim, *err;
  
@@ -273,7 +278,7 @@
    kwsinit ();
    beg = pattern;
    do
-@@ -531,17 +550,8 @@
+@@ -531,17 +551,8 @@
    struct kwsmatch kwsmatch;
    size_t ret_val;
  #ifdef MBS_SUPPORT
@@ -293,7 +298,7 @@
  #endif /* MBS_SUPPORT */
  
    for (beg = buf; beg <= buf + size; ++beg)
-@@ -550,8 +560,33 @@
+@@ -550,8 +561,33 @@
        if (offset == (size_t) -1)
  	goto failure;
  #ifdef MBS_SUPPORT
@@ -329,7 +334,7 @@
  #endif /* MBS_SUPPORT */
        beg += offset;
        len = kwsmatch.size[0];
-@@ -587,6 +622,36 @@
+@@ -587,6 +623,36 @@
  	          if (offset == -1) {
  	            break; /* Try a different anchor. */
  	          }
@@ -366,7 +371,7 @@
  	          beg += offset;
  	          len = kwsmatch.size[0];
  	        }
-@@ -597,19 +662,31 @@
+@@ -597,19 +663,31 @@
      }
  
   failure:
@@ -406,7 +411,7 @@
    end++;
    while (buf < beg && beg[-1] != eol)
      --beg;
-@@ -618,15 +695,6 @@
+@@ -618,15 +696,6 @@
  
   success_in_beg_and_len:
    *match_size = len;
diff --git a/grep.spec b/grep.spec
index 42f9ca9..d022343 100644
--- a/grep.spec
+++ b/grep.spec
@@ -85,6 +85,10 @@ fi
 %{_mandir}/*/*
 
 %changelog
+* Thu Dec 16 2004 Tim Waugh <twaugh@redhat.com>
+- Bypass kwset matching when ignoring case and processing multibyte input
+  (bug #143079).
+
 * Tue Dec 14 2004 Tim Waugh <twaugh@redhat.com> 2.5.1-42
 - Further UTF-8 processing avoided since a '\n' byte is always an
   end-of-line character in that encoding.