Fix behavior of string-mark-left-to-right.

* lisp/subr.el (string-mark-left-to-right): Search the entire string
for RTL script, not just the terminating character.  Doc fix.

3 files changed, 35 insertions, 15 deletions

diff --git a/etc/NEWS b/etc/NEWS
index 6f8c125f7f9..1a788e7f6f9 100644
--- a/etc/NEWS
+++ b/etc/NEWS
@@ -1038,11 +1038,15 @@ of function value which looks like (closure ENV ARGS &rest BODY).
 *** New function `special-variable-p' to check whether a variable is
 declared as dynamically bound.
 
-** New function `string-mark-left-to-right' appends a Unicode LRM
-(left-to-right mark) character to a string if it terminates in
-right-to-left script.  This is useful when the buffer has overall
-left-to-right paragraph direction and you need to insert a string
-whose contents (and directionality) are not known in advance.
+** New function `string-mark-left-to-right'.
+Given a string containing right-to-left (RTL) script, this function
+returns another string with a terminating LRM (left-to-right mark)
+character.  If this string is inserted into a buffer, Emacs treats the
+LRM as the end of an RTL segment and displays following text as LTR.
+
+This is useful when the buffer has overall left-to-right (LTR)
+paragraph direction and you need to insert a string whose contents
+(and hence directionality) are not known in advance.
 
 ** pre/post-command-hook are not reset to nil upon error.
 Instead, the offending function is removed.
diff --git a/lisp/ChangeLog b/lisp/ChangeLog
index bde9d54cfa2..b9cf5367bf0 100644
--- a/lisp/ChangeLog
+++ b/lisp/ChangeLog
@@ -1,3 +1,8 @@
+2011-08-12  Chong Yidong  <cyd@stupidchicken.com>
+
+        * subr.el (string-mark-left-to-right): Search the entire string
+        for RTL script, not just the terminating character.  Doc fix.
+
 2011-08-12  Stefan Monnier  <monnier@iro.umontreal.ca>
 
         * progmodes/js.el (js-syntax-propertize, js-syntax-propertize-regexp):
diff --git a/lisp/subr.el b/lisp/subr.el
index a897da1d9ba..a4251b6fee6 100644
--- a/lisp/subr.el
+++ b/lisp/subr.el
@@ -3540,18 +3540,29 @@ to case differences."
                          str2 0 (length str1) ignore-case)))
 
 (defun string-mark-left-to-right (str)
-  "Return a string that can be safely embedded in left-to-right text.
-If STR ends in right-to-left (RTL) script, return a string
-consisting of STR followed by an invisible left-to-right
-mark (LRM) character.  Otherwise, return STR."
+  "Return a string that can be safely inserted in left-to-right text.
+If STR contains right-to-left (RTL) script, return a string
+consisting of STR followed by a terminating invisible
+left-to-right mark (LRM) character.
+
+The LRM character marks the end of an RTL segment, and resets the
+display direction of any subsequent text to left-to-right.
+\(Otherwise, some of that text might be displayed as part of the
+RTL segment, based on the bidirectional display algorithm.)
+
+If STR contains no RTL characters, return STR."
   (unless (stringp str)
     (signal 'wrong-type-argument (list 'stringp str)))
-  (if (and (> (length str) 0)
-           (eq (get-char-code-property (aref str (1- (length str)))
-                                       'bidi-class)
-               'R))
-      (concat str (propertize (string ?\x200e) 'invisible t))
-    str))
+  (let ((len (length str))
+        (n 0)
+        rtl-found)
+    (while (and (not rtl-found) (< n len))
+      (setq rtl-found (memq (get-char-code-property
+                             (aref str n) 'bidi-class) '(R AL RLO))
+            n (1+ n)))
+    (if rtl-found
+        (concat str (propertize (string ?\x200e) 'invisible t))
+      str)))
 
 ;;;; invisibility specs