X-Git-Url: https://code.delx.au/gnu-emacs/blobdiff_plain/7c82f3e23e37cc848a38b1f8be7149fd672a6393..f15c8583198c3d6c26ca0c0a5b6fb019f98d6c3c:/lisp/emacs-lisp/regexp-opt.el

diff --git a/lisp/emacs-lisp/regexp-opt.el b/lisp/emacs-lisp/regexp-opt.el
index e2ae390d19..b0fb23dbcc 100644
--- a/lisp/emacs-lisp/regexp-opt.el
+++ b/lisp/emacs-lisp/regexp-opt.el
@@ -1,10 +1,9 @@
 ;;; regexp-opt.el --- generate efficient regexps to match strings
 
-;; Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
-;;   2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation, Inc.
+;; Copyright (C) 1994-2015 Free Software Foundation, Inc.
 
 ;; Author: Simon Marshall <simon@gnu.org>
-;; Maintainer: FSF
+;; Maintainer: emacs-devel@gnu.org
 ;; Keywords: strings, regexps, extensions
 
 ;; This file is part of GNU Emacs.
@@ -137,14 +136,11 @@ This means the number of non-shy regexp grouping constructs
 
 ;;; Workhorse functions.
 
-(eval-when-compile
-  (require 'cl))
-
 (defun regexp-opt-group (strings &optional paren lax)
   "Return a regexp to match a string in the sorted list STRINGS.
 If PAREN non-nil, output regexp parentheses around returned regexp.
 If LAX non-nil, don't output parentheses if it doesn't require them.
-Merges keywords to avoid backtracking in Emacs' regexp matcher."
+Merges keywords to avoid backtracking in Emacs's regexp matcher."
   ;; The basic idea is to find the shortest common prefix or suffix, remove it
   ;; and recurse.  If there is no prefix, we divide the list into two so that
   ;; \(at least) one half will have at least a one-character common prefix.
@@ -209,9 +205,7 @@ Merges keywords to avoid backtracking in Emacs' regexp matcher."
 		      (regexp-opt-group suffixes t t)
 		      close-group))
 
-	  (let* ((sgnirts (mapcar (lambda (s)
-				    (concat (nreverse (string-to-list s))))
-				  strings))
+	  (let* ((sgnirts (mapcar #'reverse strings))
 		 (xiffus (try-completion "" sgnirts)))
 	    (if (> (length xiffus) 0)
 		;; common suffix: take it and recurse on the prefixes.
@@ -222,8 +216,7 @@ Merges keywords to avoid backtracking in Emacs' regexp matcher."
 			      'string-lessp)))
 		  (concat open-group
 			  (regexp-opt-group prefixes t t)
-			  (regexp-quote
-			   (concat (nreverse (string-to-list xiffus))))
+			  (regexp-quote (nreverse xiffus))
 			  close-group))
 
 	      ;; Otherwise, divide the list into those that start with a
@@ -238,7 +231,8 @@ Merges keywords to avoid backtracking in Emacs' regexp matcher."
 
 
 (defun regexp-opt-charset (chars)
-  "Return a regexp to match a character in CHARS."
+  "Return a regexp to match a character in CHARS.
+CHARS should be a list of characters."
   ;; The basic idea is to find character ranges.  Also we take care in the
   ;; position of character set meta characters in the character set regexp.
   ;;
@@ -249,15 +243,15 @@ Merges keywords to avoid backtracking in Emacs' regexp matcher."
     ;;
     ;; Make a character map but extract character set meta characters.
     (dolist (char chars)
-      (case char
-	(?\]
-	 (setq bracket "]"))
-	(?^
-	 (setq caret "^"))
-	(?-
-	 (setq dash "-"))
-	(otherwise
-	 (aset charmap char t))))
+      (cond
+       ((eq char ?\])
+	(setq bracket "]"))
+       ((eq char ?^)
+	(setq caret "^"))
+       ((eq char ?-)
+	(setq dash "-"))
+       (t
+	(aset charmap char t))))
     ;;
     ;; Make a character set from the map using ranges where applicable.
     (map-char-table
@@ -269,14 +263,14 @@ Merges keywords to avoid backtracking in Emacs' regexp matcher."
 		   (setq charset (format "%s%c-%c" charset start end))
 		 (while (>= end start)
 		   (setq charset (format "%s%c" charset start))
-		   (incf start)))
+		   (setq start (1+ start))))
 	       (setq start (car c) end (cdr c)))
 	   (if (= (1- c) end) (setq end c)
 	     (if (> end (+ start 2))
 	       (setq charset (format "%s%c-%c" charset start end))
 	     (while (>= end start)
 	       (setq charset (format "%s%c" charset start))
-	       (incf start)))
+	       (setq start (1+ start))))
 	     (setq start c end c)))))
      charmap)
     (when (>= end start)
@@ -284,11 +278,13 @@ Merges keywords to avoid backtracking in Emacs' regexp matcher."
 	  (setq charset (format "%s%c-%c" charset start end))
 	(while (>= end start)
 	  (setq charset (format "%s%c" charset start))
-	  (incf start))))
+	  (setq start (1+ start)))))
     ;;
     ;; Make sure a caret is not first and a dash is first or last.
     (if (and (string-equal charset "") (string-equal bracket ""))
-	(concat "[" dash caret "]")
+	(if (string-equal dash "")
+            "\\^"                       ; [^] is not a valid regexp
+          (concat "[" dash caret "]"))
       (concat "[" bracket charset caret dash "]"))))
 
 (provide 'regexp-opt)