Ticket #2396 (Find File "Whole words" search bug)

when searching in files for non english word with "Whole words" set "on" - then nothig will be found
try search word: "время" and also "time" in next example:

'time'
Время
'Time'
'время'

Signed-off-by: Slava Zanko <slavazanko@gmail.com>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>

lib/search/normal.c

@@ -92,9 +92,12 @@ mc_search__cond_struct_new_init_normal (const char *charset, mc_search_t * lc_mc
         mc_search__normal_translate_to_regex (mc_search_cond->str->str, &mc_search_cond->len);
 
     g_string_free (mc_search_cond->str, TRUE);
-    if (lc_mc_search->whole_words) {
+    if (lc_mc_search->whole_words)
-        g_string_prepend (tmp, "\\b");
+    {
-        g_string_append (tmp, "\\b");
+        /* NOTE: \b as word boundary doesn't allow search
+         * whole words with non-ASCII symbols */
+        g_string_prepend (tmp, "(^|[^\\p{L}\\p{N}_])(");
+        g_string_append (tmp, ")([^\\p{L}\\p{N}_]|$)");
     }
     mc_search_cond->str = tmp;
 

lib/search/regex.c

@@ -58,7 +58,7 @@ typedef enum
 /*** file scope functions ************************************************************************/
 
 static gboolean
-mc_search__regex_str_append_if_special (GString * copy_to, GString * regex_str, gsize * offset)
+mc_search__regex_str_append_if_special (GString * copy_to, const GString * regex_str, gsize * offset)
 {
     char *tmp_regex_str;
     gsize spec_chr_len;
@@ -622,10 +622,25 @@ mc_search__run_regex (mc_search_t * lc_mc_search, const void *user_data,
         {
         case COND__FOUND_OK:
 #ifdef SEARCH_TYPE_GLIB
+            if (lc_mc_search->whole_words)
+            {
+                g_match_info_fetch_pos (lc_mc_search->regex_match_info, 2, &start_pos, &end_pos);
+            }
+            else
+            {
                 g_match_info_fetch_pos (lc_mc_search->regex_match_info, 0, &start_pos, &end_pos);
+            }
 #else /* SEARCH_TYPE_GLIB */
+            if (lc_mc_search->whole_words)
+            {
+                start_pos = lc_mc_search->iovector[4];
+                end_pos = lc_mc_search->iovector[5];
+            }
+            else
+            {
                 start_pos = lc_mc_search->iovector[0];
                 end_pos = lc_mc_search->iovector[1];
+            }
 #endif /* SEARCH_TYPE_GLIB */
             if (found_len)
                 *found_len = end_pos - start_pos;