Import regex changes from libc via gnulib.

3 files changed, 39 insertions, 17 deletions

diff --git a/lib/ChangeLog b/lib/ChangeLog
index ac3c19dfe..7f0ed7435 100644
--- a/lib/ChangeLog
+++ b/lib/ChangeLog
@@ -1,5 +1,24 @@
 2006-08-10  Paul Eggert  <eggert@cs.ucla.edu>
 
+	Import the following changes from libc:
+
+	2006-06-02  Jakub Jelinek  <jakub@redhat.com>
+
+	* posix/regex_internal.c (re_string_skip_chars): If no character has
+	been converted at all, set *last_wc to WEOF.  If mbrtowc failed, set wc
+	to the byte which couldn't be converted.
+	(re_string_reconstruct): Don't clear valid_raw_len before calling
+	re_string_skip_chars.  If wc is WEOF after re_string_skip_chars, set
+	tip_context using re_string_context_at.
+
+	2006-05-02  Ulrich Drepper  <drepper@redhat.com>
+
+	* posix/regex.h: g++ still cannot handled [restrict].
+
+	2006-04-21  Ulrich Drepper  <drepper@redhat.com>
+
+	* posix/regex.h: Remove special handling for VMS.
+
 	Accommodate new getaddrinfo implementation in gnulib.
 	* inet_ntop.c, inet_ntop.h, snprintf.c, snprintf.h, socket_.h:
 	New files, from gnulib.
diff --git a/lib/regex.h b/lib/regex.h
index 51857e559..e8748e195 100644
--- a/lib/regex.h
+++ b/lib/regex.h
@@ -37,12 +37,6 @@ extern "C" {
 # define __USE_GNU_REGEX 1
 #endif
 
-#ifdef __VMS
-/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
-   should be there.  */
-# include <stddef.h>
-#endif
-
 #ifdef _REGEX_LARGE_OFFSETS
 
 /* Use types and values that are wide enough to represent signed and
@@ -641,9 +635,10 @@ extern int re_exec (const char *);
 #  endif
 # endif
 #endif
-/* gcc 3.1 and up support the [restrict] syntax, but g++ doesn't.  */
+/* gcc 3.1 and up support the [restrict] syntax.  */
 #ifndef __restrict_arr
-# if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) && !defined __cplusplus
+# if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \
+     && !defined __GNUG__
 #  define __restrict_arr __restrict
 # else
 #  define __restrict_arr
diff --git a/lib/regex_internal.c b/lib/regex_internal.c
index 55fe2982a..a377d2b28 100644
--- a/lib/regex_internal.c
+++ b/lib/regex_internal.c
@@ -488,7 +488,7 @@ re_string_skip_chars (re_string_t *pstr, Idx new_raw_idx, wint_t *last_wc)
   mbstate_t prev_st;
   Idx rawbuf_idx;
   size_t mbclen;
-  wchar_t wc = 0;
+  wchar_t wc = WEOF;
 
   /* Skip the characters which are not necessary to check.  */
   for (rawbuf_idx = pstr->raw_mbs_idx + pstr->valid_raw_len;
@@ -501,7 +501,11 @@ re_string_skip_chars (re_string_t *pstr, Idx new_raw_idx, wint_t *last_wc)
 			remain_len, &pstr->cur_state);
       if (BE (mbclen == (size_t) -2 || mbclen == (size_t) -1 || mbclen == 0, 0))
 	{
-	  /* We treat these cases as a singlebyte character.  */
+	  /* We treat these cases as a single byte character.  */
+	  if (mbclen == 0 || remain_len == 0)
+	    wc = L'\0';
+	  else
+	    wc = *(unsigned char *) (pstr->raw_mbs + rawbuf_idx);
 	  mbclen = 1;
 	  pstr->cur_state = prev_st;
 	}
@@ -627,7 +631,6 @@ re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags)
 	    }
 #endif
 	  pstr->valid_len = 0;
-	  pstr->valid_raw_len = 0;
 #ifdef RE_ENABLE_I18N
 	  if (pstr->mb_cur_max > 1)
 	    {
@@ -690,6 +693,16 @@ re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags)
 
 	      if (wc == WEOF)
 		pstr->valid_len = re_string_skip_chars (pstr, idx, &wc) - idx;
+	      if (wc == WEOF)
+		pstr->tip_context
+		  = re_string_context_at (pstr, pstr->valid_raw_len - 1, eflags);
+	      else
+		pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0)
+				      && IS_WIDE_WORD_CHAR (wc))
+				     ? CONTEXT_WORD
+				     : ((IS_WIDE_NEWLINE (wc)
+					 && pstr->newline_anchor)
+					? CONTEXT_NEWLINE : 0));
 	      if (BE (pstr->valid_len, 0))
 		{
 		  for (wcs_idx = 0; wcs_idx < pstr->valid_len; ++wcs_idx)
@@ -698,17 +711,12 @@ re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags)
 		    memset (pstr->mbs, -1, pstr->valid_len);
 		}
 	      pstr->valid_raw_len = pstr->valid_len;
-	      pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0)
-				    && IS_WIDE_WORD_CHAR (wc))
-				   ? CONTEXT_WORD
-				   : ((IS_WIDE_NEWLINE (wc)
-				       && pstr->newline_anchor)
-				      ? CONTEXT_NEWLINE : 0));
 	    }
 	  else
 #endif /* RE_ENABLE_I18N */
 	    {
 	      int c = pstr->raw_mbs[pstr->raw_mbs_idx + offset - 1];
+	      pstr->valid_raw_len = 0;
 	      if (pstr->trans)
 		c = pstr->trans[c];
 	      pstr->tip_context = (bitset_contain (pstr->word_char, c)