--- a/lisp/ChangeLog	Wed Jun 19 09:30:30 2013 -0600
+++ b/lisp/ChangeLog	Fri Jun 21 18:45:06 2013 +0900
@@ -1,3 +1,10 @@
+2013-06-21  Stephen J. Turnbull  <stephen@xemacs.org>
+
+	Fix no-Mule build.
+
+	* paragraphs.el (sentence-end): Use octal, not Unicode, escapes.
+	* (sentence-end-base): Use non-ASCII only in Mule.
+
 2013-06-17  Aidan Kehoe  <kehoea@parhasard.net>
 
 	* process.el (process-synchronize-point): Moved to a label.

--- a/lisp/paragraphs.el	Wed Jun 19 09:30:30 2013 -0600
+++ b/lisp/paragraphs.el	Fri Jun 21 18:45:06 2013 +0900
@@ -180,7 +180,10 @@
   :type '(choice regexp (const :tag "Use default value" nil)))
 (put 'sentence-end 'safe-local-variable 'string-or-null-p)
 
-(defcustom sentence-end-base "[.?!][]\"'”)}]*"
+(defcustom sentence-end-base
+  (concat "[.?!][]\"'"
+	  (if (featurep 'mule) (unicode-to-char #x201D) "")
+	  ")}]*")
   "Regexp matching the basic end of a sentence, not including following space."
   :group 'paragraphs
   :type 'string
@@ -201,14 +204,14 @@
 in between.  See Info node `(elisp)Standard Regexps'."
   (or sentence-end
       ;; We accept non-break space along with space.
-      (concat (if sentence-end-without-period "\\w[ \u00a0][ \u00a0]\\|")
+      (concat (if sentence-end-without-period "\\w[ \240][ \240]\\|")
 	      "\\("
 	      sentence-end-base
               (if sentence-end-double-space
-                  "\\($\\|[ \u00a0]$\\|\t\\|[ \u00a0][ \u00a0]\\)" "\\($\\|[\t \u00a0]\\)")
+                  "\\($\\|[ \240]$\\|\t\\|[ \240][ \240]\\)" "\\($\\|[\t \240]\\)")
               "\\|[" sentence-end-without-space "]+"
 	      "\\)"
-              "[ \u00a0\t\n]*")))
+              "[ \240\t\n]*")))
 
 (defcustom page-delimiter "^\014"
   "*Regexp describing line-beginnings that separate pages."