nano

commit cbf7e57ed0a0800d35f84573e353b5a1b4af0334
parent e36e829ad064d4f80fbd7b3249d3b33c4ebbf64c
Author: Benno Schulenberg <bensberg@justemail.net>
Date:   Sun,  1 Jan 2017 20:16:18 +0100

search: make a regex with a beginning-of-word anchor work correctly

The search routine begins searching right after the cursor and behaves
as if the line starts there, which means that a beginning-of-word anchor
(\< or \b) will match there also when in fact the cursor is sitting in
the middle of a word.  To prevent finding a false match, verify that
for a regex that starts with a BOW anchor the found match is actually
the start of a word.

This fixes https://savannah.gnu.org/bugs/?45630.

Diffstat:
M
src/search.c
 | 
26
++++++++++++++++++++++++--

1 file changed, 24 insertions(+), 2 deletions(-)
diff --git a/src/search.c b/src/search.c
@@ -38,6 +38,8 @@ static bool history_changed = FALSE;
 #ifdef HAVE_REGEX_H
 static bool regexp_compiled = FALSE;
 	/* Have we compiled any regular expressions? */
+static bool bow_anchored = FALSE;
+	/* Whether a regex starts with a beginning-of-word anchor. */
 
 /* Compile the given regular expression and store it in search_regexp.
  * Return TRUE if the expression is valid, and FALSE otherwise. */
@@ -60,6 +62,10 @@ bool regexp_init(const char *regexp)
 
     regexp_compiled = TRUE;
 
+    /* Remember whether the regex starts with a beginning-of-word anchor. */
+    bow_anchored = (strncmp(regexp, "\\<", 2) == 0 ||
+			strncmp(regexp, "\\b", 2) == 0);
+
     return TRUE;
 }
 
@@ -292,8 +298,24 @@ int findnextstr(const char *needle, bool whole_word_only, size_t *match_len,
 	if (found != NULL) {
 #ifdef HAVE_REGEX_H
 	    /* When doing a regex search, compute the length of the match. */
-	    if (ISSET(USE_REGEXP))
+	    if (ISSET(USE_REGEXP)) {
 		found_len = regmatches[0].rm_eo - regmatches[0].rm_so;
+
+		/* If the regex starts with a BOW anchor, check that the found
+		 * match actually is the start of a word.  If not, continue. */
+		if (bow_anchored && found != fileptr->data) {
+		    size_t before = move_mbleft(fileptr->data, found - fileptr->data);
+
+		    /* If a word char is before the match, skip this match. */
+		    if (is_word_mbchar(fileptr->data + before, FALSE)) {
+			if (ISSET(BACKWARDS_SEARCH))
+			    rev_start = fileptr->data + before;
+			else
+			    rev_start = found + move_mbright(found, 0);
+			continue;
+		    }
+		}
+	    }
 #endif
 #ifndef DISABLE_SPELLER
 	    /* When we're spell checking, a match is only a true match when
@@ -304,7 +326,7 @@ int findnextstr(const char *needle, bool whole_word_only, size_t *match_len,
 		    break;
 		else {
 		    /* Maybe there is a whole word in the rest of the line. */
-		    rev_start = found + 1;
+		    rev_start = found + move_mbright(found, 0);
 		    continue;
 		}
 	    } else