2005-01-26 Jakub Jelinek <jakub@redhat.com>

	[BZ #693]
	* posix/regex_internal.h (DUMMY_CONSTRAINT): Rename to...
	(WORD_DELIM_CONSTRAINT): ...this.
	(NOT_WORD_DELIM_CONSTRAINT): Define.
	(re_context_type): Add INSIDE_NOTWORD and NOT_WORD_DELIM,
	change WORD_DELIM to use WORD_DELIM_CONSTRAINT.
	* posix/regcomp.c (peek_token): For \B create NOT_WORD_DELIM
	anchor instead of INSIDE_WORD.
	(parse_expression): Handle NOT_WORD_DELIM constraint.
	* posix/bug-regex19.c (tests): Adjust tests that relied on \B
	being inside word instead of not word delim.
	* posix/tst-rxspencer.c (mb_frob_pattern): Don't frob escaped
	characters.
	* posix/rxspencer/tests: Add some new tests.

diff --git a/posix/bug-regex19.c b/posix/bug-regex19.c
index 4000b19b4d539a08c4b8c9493bd9cefa130d8aa2..3a6391cb01aef7f88efdef599a0c46630de40274 100644 (file)
--- a/posix/bug-regex19.c
+++ b/posix/bug-regex19.c
@@ -170,22 +170,22 @@ static struct test_s
   {ERE, "[^k]\\B[^k]", "kBk", 0, -1},
   {ERE, "[^C]\\B[^C]", "CCCABA", 0, 3},
   {ERE, "[^C]\\B[^C]", "CBC", 0, -1},
-  {ERE, ".(\\b|\\B).", "=~AB", 0, 1},
+  {ERE, ".(\\b|\\B).", "=~AB", 0, 0},
   {ERE, ".(\\b|\\B).", "A=C", 0, 0},
   {ERE, ".(\\b|\\B).", "ABC", 0, 0},
-  {ERE, ".(\\b|\\B).", "=~\\!", 0, -1},
-  {ERE, "[^k](\\b|\\B)[^k]", "=~AB", 0, 1},
+  {ERE, ".(\\b|\\B).", "=~\\!", 0, 0},
+  {ERE, "[^k](\\b|\\B)[^k]", "=~AB", 0, 0},
   {ERE, "[^k](\\b|\\B)[^k]", "A=C", 0, 0},
   {ERE, "[^k](\\b|\\B)[^k]", "ABC", 0, 0},
-  {ERE, "[^k](\\b|\\B)[^k]", "=~kBD", 0, 3},
-  {ERE, "[^k](\\b|\\B)[^k]", "=~\\!", 0, -1},
-  {ERE, "[^k](\\b|\\B)[^k]", "=~kB", 0, -1},
-  {ERE, "[^C](\\b|\\B)[^C]", "=~AB", 0, 1},
+  {ERE, "[^k](\\b|\\B)[^k]", "=~kBD", 0, 0},
+  {ERE, "[^k](\\b|\\B)[^k]", "=~\\!", 0, 0},
+  {ERE, "[^k](\\b|\\B)[^k]", "=~kB", 0, 0},
+  {ERE, "[^C](\\b|\\B)[^C]", "=~AB", 0, 0},
   {ERE, "[^C](\\b|\\B)[^C]", "A=C", 0, 0},
   {ERE, "[^C](\\b|\\B)[^C]", "ABC", 0, 0},
-  {ERE, "[^C](\\b|\\B)[^C]", "=~CBD", 0, 3},
-  {ERE, "[^C](\\b|\\B)[^C]", "=~\\!", 0, -1},
-  {ERE, "[^C](\\b|\\B)[^C]", "=~CB", 0, -1},
+  {ERE, "[^C](\\b|\\B)[^C]", "=~CBD", 0, 0},
+  {ERE, "[^C](\\b|\\B)[^C]", "=~\\!", 0, 0},
+  {ERE, "[^C](\\b|\\B)[^C]", "=~CB", 0, 0},
   {ERE, "\\b([A]|[!]|.B)", "A=AC", 0, 0},
   {ERE, "\\b([A]|[!]|.B)", "=AC", 0, 1},
   {ERE, "\\b([A]|[!]|.B)", "!AC", 0, 1},

diff --git a/posix/regcomp.c b/posix/regcomp.c
index 5de5bf725ac66f00400d9d82d6e03b449618ff97..72bf187b14be96f7edcaa98a0639216ebca5ba6a 100644 (file)
--- a/posix/regcomp.c
+++ b/posix/regcomp.c
@@ -1859,7 +1859,7 @@ peek_token (token, input, syntax)
          if (!(syntax & RE_NO_GNU_OPS))
            {
              token->type = ANCHOR;
-             token->opr.ctx_type = INSIDE_WORD;
+             token->opr.ctx_type = NOT_WORD_DELIM;
            }
          break;
        case 'w':
@@ -2349,15 +2349,25 @@ parse_expression (regexp, preg, token, syntax, nest, err)
       break;
     case ANCHOR:
       if ((token->opr.ctx_type
-          & (WORD_DELIM | INSIDE_WORD | WORD_FIRST | WORD_LAST))
+          & (WORD_DELIM | NOT_WORD_DELIM | WORD_FIRST | WORD_LAST))
          && dfa->word_ops_used == 0)
        init_word_char (dfa);
-      if (token->opr.ctx_type == WORD_DELIM)
+      if (token->opr.ctx_type == WORD_DELIM
+          || token->opr.ctx_type == NOT_WORD_DELIM)
        {
          bin_tree_t *tree_first, *tree_last;
-         token->opr.ctx_type = WORD_FIRST;
-         tree_first = re_dfa_add_tree_node (dfa, NULL, NULL, token);
-         token->opr.ctx_type = WORD_LAST;
+         if (token->opr.ctx_type == WORD_DELIM)
+           {
+             token->opr.ctx_type = WORD_FIRST;
+             tree_first = re_dfa_add_tree_node (dfa, NULL, NULL, token);
+             token->opr.ctx_type = WORD_LAST;
+            }
+          else
+            {
+             token->opr.ctx_type = INSIDE_WORD;
+             tree_first = re_dfa_add_tree_node (dfa, NULL, NULL, token);
+             token->opr.ctx_type = INSIDE_NOTWORD;
+            }
          tree_last = re_dfa_add_tree_node (dfa, NULL, NULL, token);
          token->type = OP_ALT;
          tree = re_dfa_add_tree_node (dfa, tree_first, tree_last, token);

diff --git a/posix/regex_internal.h b/posix/regex_internal.h
index 0ccd8d3665fd5db7b9d152c30ccf46bc20d52cb2..18865a7266447e6c0dfe274c2bbb0480ab3ef026 100644 (file)
--- a/posix/regex_internal.h
+++ b/posix/regex_internal.h
@@ -143,18 +143,21 @@ static inline void bitset_mask (bitset dest, const bitset src);
 #define NEXT_NEWLINE_CONSTRAINT 0x0020
 #define PREV_BEGBUF_CONSTRAINT 0x0040
 #define NEXT_ENDBUF_CONSTRAINT 0x0080
-#define DUMMY_CONSTRAINT 0x0100
+#define WORD_DELIM_CONSTRAINT 0x0100
+#define NOT_WORD_DELIM_CONSTRAINT 0x0200
 
 typedef enum
 {
   INSIDE_WORD = PREV_WORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
   WORD_FIRST = PREV_NOTWORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
   WORD_LAST = PREV_WORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
+  INSIDE_NOTWORD = PREV_NOTWORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
   LINE_FIRST = PREV_NEWLINE_CONSTRAINT,
   LINE_LAST = NEXT_NEWLINE_CONSTRAINT,
   BUF_FIRST = PREV_BEGBUF_CONSTRAINT,
   BUF_LAST = NEXT_ENDBUF_CONSTRAINT,
-  WORD_DELIM = DUMMY_CONSTRAINT
+  WORD_DELIM = WORD_DELIM_CONSTRAINT,
+  NOT_WORD_DELIM = NOT_WORD_DELIM_CONSTRAINT
 } re_context_type;
 
 typedef struct

diff --git a/posix/rxspencer/tests b/posix/rxspencer/tests
index a724252d8c9724b37da948291d1a289d5f079c36..a8b6e4baa82085967a8c63bbdfb200ac27164fec 100644 (file)
--- a/posix/rxspencer/tests
+++ b/posix/rxspencer/tests
@@ -526,3 +526,12 @@ a((b+|((c)*)))+d   -       abcd    abcd    c,c,c,c
 (((\b))){0}    -       x       @x      -,-,-
 a(((.*)))b((\2)){0}c   -       abc     abc     @bc,@bc,@bc,-,-
 a(((.*)))b((\1)){0}c   -       axbc    axbc    x,x,x,-,-
+
+\b     &       SaT     @aT
+\b     &       aT      @aT
+a.*\b  &       abT     ab
+\b     &       STSS
+\B     &       abc     @bc
+\B     &       aSbTc
+\B     &       SaT     @SaT
+\B     &       aSTSb   @TSb

diff --git a/posix/tst-rxspencer.c b/posix/tst-rxspencer.c
index cb40421797bf2f8fb3e7151880d25f433006b4f9..3febc01cb201d3482293629c3210abfa1cff31e8 100644 (file)
--- a/posix/tst-rxspencer.c
+++ b/posix/tst-rxspencer.c
@@ -127,14 +127,15 @@ mb_frob_string (const char *str, const char *letters)
 }
 
 /* Like mb_frob_string, but don't replace anything between
-   [: and :], [. and .] or [= and =].  */
+   [: and :], [. and .] or [= and =] or characters escaped
+   with a backslash.  */
 
 static char *
 mb_frob_pattern (const char *str, const char *letters)
 {
   char *ret, *dst;
   const char *src;
-  int in_class = 0;
+  int in_class = 0, escaped = 0;
 
   if (str == NULL)
     return NULL;
@@ -144,7 +145,18 @@ mb_frob_pattern (const char *str, const char *letters)
     return NULL;
 
   for (src = str, dst = ret; *src; ++src)
-    if (!in_class && strchr (letters, *src))
+    if (*src == '\\')
+      {
+       escaped ^= 1;
+       *dst++ = *src;
+      }
+    else if (escaped)
+      {
+       escaped = 0;
+       *dst++ = *src;
+       continue;
+      }
+    else if (!in_class && strchr (letters, *src))
       dst = mb_replace (dst, *src);
     else
       {