[Emacs-diffs] trunk r115376: Fix bug #16043 with crashes when displaying new bidi control characters.

[Eli Zaretskii]

------------------------------------------------------------
revno: 115376
revision-id: address@hidden
parent: address@hidden
fixes bug: http://debbugs.gnu.org/16043
committer: Eli Zaretskii <address@hidden>
branch nick: trunk
timestamp: Wed 2013-12-04 18:58:05 +0200
message:
  Fix bug #16043 with crashes when displaying new bidi control characters.
  
   src/bidi.c (bidi_get_type, bidi_get_category): Handle the isolate
   directional control characters.  Update type and category
   determination according to the UBA from Unicode v6.3.
   (bidi_category_t): New category EXPLICIT_FORMATTING.
   src/dispextern.h (bidi_type_t): Update to include new bidirectional
   properties introduced with Unicode v6.3.
  
   admin/unidata/unidata-gen.el (unidata-prop-alist): Update bidi-class
   to include the new isolate-related classes introduced with Unicode
   v6.3.
   (unidata-encode-val): Accept an additional optional argument, a
   warning message to emit when UnicodeData.txt defines bidi-class
   values that are not in unidata-prop-alist.  Add a comment
   explaining what should maintainers do if/when such a warning ever
   appears.
   (unidata-gen-table): Call unidata-encode-val with 3rd arg non-nil
   when generating uni-bidi.el.
modified:
  admin/ChangeLog                changelog-20091113204419-o5vbwnq5f7feedwu-2226
  admin/unidata/unidata-gen.el   
unidatagen.el-20091113204419-o5vbwnq5f7feedwu-8382
  src/ChangeLog                  changelog-20091113204419-o5vbwnq5f7feedwu-1438
  src/bidi.c                     bidi.c-20091231194348-rm8gzug639w0dpq5-1
  src/dispextern.h               
dispextern.h-20091113204419-o5vbwnq5f7feedwu-218

=== modified file 'admin/ChangeLog'
--- a/admin/ChangeLog   2013-12-01 01:37:23 +0000
+++ b/admin/ChangeLog   2013-12-04 16:58:05 +0000
@@ -1,3 +1,16 @@
+2013-12-04  Eli Zaretskii  <address@hidden>
+
+       * unidata/unidata-gen.el (unidata-prop-alist): Update bidi-class
+       to include the new isolate-related classes introduced with Unicode
+       v6.3.
+       (unidata-encode-val): Accept an additional optional argument, a
+       warning message to emit when UnicodeData.txt defines bidi-class
+       values that are not in unidata-prop-alist.  Add a comment
+       explaining what should maintainers do if/when such a warning ever
+       appears.
+       (unidata-gen-table): Call unidata-encode-val with 3rd arg non-nil
+       when generating uni-bidi.el.
+
 2013-12-01  Glenn Morris  <address@hidden>
 
        * unidata/Makefile.in (${DSTDIR}/charprop.el):

=== modified file 'admin/unidata/unidata-gen.el'
--- a/admin/unidata/unidata-gen.el      2013-11-28 20:21:55 +0000
+++ b/admin/unidata/unidata-gen.el      2013-12-04 16:58:05 +0000
@@ -194,8 +194,8 @@
      4 unidata-gen-table-symbol "uni-bidi.el"
      "Unicode bidi class.
 Property value is one of the following symbols:
-  L, LRE, LRO, R, AL, RLE, RLO, PDF, EN, ES, ET,
-  AN, CS, NSM, BN, B, S, WS, ON"
+  L, LRE, LRO, LRI, R, AL, RLE, RLO, RLI, FSI, PDF, PDI,
+  EN, ES, ET, AN, CS, NSM, BN, B, S, WS, ON"
      unidata-describe-bidi-class
      ;; The assignment of default values to blocks of code points
      ;; follows the file DerivedBidiClass.txt from the Unicode
@@ -205,7 +205,8 @@
        (#xFB1D #xFB4F R) (#x10800 #x10FFF R) (#x1E800 #x1EFFF R))
      ;; The order of elements must be in sync with bidi_type_t in
      ;; src/dispextern.h.
-     (L R EN AN BN B AL LRE LRO RLE RLO PDF ES ET CS NSM S WS ON))
+     (L R EN AN BN B AL LRE LRO RLE RLO PDF LRI RLI FSI PDI
+       ES ET CS NSM S WS ON))
     (decomposition
      5 unidata-gen-table-decomposition "uni-decomposition.el"
      "Unicode decomposition mapping.
@@ -397,12 +398,17 @@
 ;; If VAL is one of VALn, just return n.
 ;; Otherwise, VAL-LIST is modified to this:
 ;;   ((nil . 0) (VAL1 . 1) (VAL2 . 2) ... (VAL . n+1))
+;;
+;; WARN is an optional warning to display when the value list is
+;; extended, for property values that need to be in sync with other
+;; parts of Emacs; currently only used for bidi-class.
 
-(defun unidata-encode-val (val-list val)
+(defun unidata-encode-val (val-list val &optional warn)
   (let ((slot (assoc val val-list))
        val-code)
     (if slot
        (cdr slot)
+      (if warn (message warn val))
       (setq val-code (length val-list))
       (nconc val-list (list (cons val val-code)))
       val-code)))
@@ -413,6 +419,16 @@
   (let ((table (make-char-table 'char-code-property-table))
        (prop-idx (unidata-prop-index prop))
        (vec (make-vector 128 0))
+       ;; When this warning is printed, there's a need to make the
+       ;; following changes:
+       ;; (1) update unidata-prop-alist with the new bidi-class values;
+       ;; (2) extend bidi_type_t enumeration on src/dispextern.h to
+       ;;     include the new classes;
+       ;; (3) possibly update the assertion in bidi.c:bidi_check_type; and
+       ;; (4) possibly update the switch cases in
+       ;;     bidi.c:bidi_get_type and bidi.c:bidi_get_category.
+       (bidi-warning "\
+** Found new bidi-class '%s', please update bidi.c and dispextern.h")
        tail elt range val val-code idx slot
        prev-range-data)
     (setq val-list (cons nil (copy-sequence val-list)))
@@ -438,7 +454,9 @@
       (setq elt (car tail) tail (cdr tail))
       (setq range (car elt)
            val (funcall val-func (nth prop-idx elt)))
-      (setq val-code (if val (unidata-encode-val val-list val)))
+      (setq val-code (if val (unidata-encode-val val-list val
+                                                (and (eq prop 'bidi-class)
+                                                     bidi-warning))))
       (if (consp range)
          (when val-code
            (set-char-table-range table range val-code)
@@ -486,7 +504,9 @@
            (setq new-val (funcall val-func (nth prop-idx elt)))
            (if (not (eq val new-val))
                (setq val new-val
-                     val-code (if val (unidata-encode-val val-list val))))
+                     val-code (if val (unidata-encode-val
+                                       val-list val (and (eq prop 'bidi-class)
+                                                         bidi-warning)))))
            (if val-code
                (aset vec (- range start) val-code))
            (setq tail (cdr tail)))

=== modified file 'src/ChangeLog'
--- a/src/ChangeLog     2013-12-04 14:11:33 +0000
+++ b/src/ChangeLog     2013-12-04 16:58:05 +0000
@@ -1,3 +1,13 @@
+2013-12-04  Eli Zaretskii  <address@hidden>
+
+       * bidi.c (bidi_get_type, bidi_get_category): Handle the isolate
+       directional control characters.  Update type and category
+       determination according to the UBA from Unicode v6.3.
+       (bidi_category_t): New category EXPLICIT_FORMATTING.
+
+       * dispextern.h (bidi_type_t): Update to include new bidirectional
+       properties introduced with Unicode v6.3.  (Bug#16043)
+
 2013-12-04  Martin Rudalics  <address@hidden>
 
        * xterm.c (XTflash): Fix coordinate of bottom area to flash

=== modified file 'src/bidi.c'
--- a/src/bidi.c        2013-11-22 16:04:49 +0000
+++ b/src/bidi.c        2013-12-04 16:58:05 +0000
@@ -76,7 +76,8 @@
   UNKNOWN_BC,
   NEUTRAL,
   WEAK,
-  STRONG
+  STRONG,
+  EXPLICIT_FORMATTING
 } bidi_category_t;
 
 /* UAX#9 says to search only for L, AL, or R types of characters, and
@@ -115,13 +116,9 @@
   if (default_type == UNKNOWN_BT)
     emacs_abort ();
 
-  if (override == NEUTRAL_DIR)
-    return default_type;
-
   switch (default_type)
     {
-      /* Although UAX#9 does not tell, it doesn't make sense to
-        override NEUTRAL_B and LRM/RLM characters.  */
+      case WEAK_BN:
       case NEUTRAL_B:
       case LRE:
       case LRO:
@@ -129,20 +126,20 @@
       case RLO:
       case PDF:
        return default_type;
+       /* FIXME: The isolate controls are treated as BN until we add
+          support for UBA v6.3.  */
+      case LRI:
+      case RLI:
+      case FSI:
+      case PDI:
+       return WEAK_BN;
       default:
-       switch (ch)
-         {
-           case LRM_CHAR:
-           case RLM_CHAR:
-             return default_type;
-           default:
-             if (override == L2R) /* X6 */
-               return STRONG_L;
-             else if (override == R2L)
-               return STRONG_R;
-             else
-               emacs_abort (); /* can't happen: handled above */
-         }
+       if (override == L2R)
+         return STRONG_L;
+       else if (override == R2L)
+         return STRONG_R;
+       else
+         return default_type;
     }
 }
 
@@ -163,12 +160,7 @@
       case STRONG_L:
       case STRONG_R:
       case STRONG_AL:
-      case LRE:
-      case LRO:
-      case RLE:
-      case RLO:
        return STRONG;
-      case PDF:                /* ??? really?? */
       case WEAK_EN:
       case WEAK_ES:
       case WEAK_ET:
@@ -176,12 +168,30 @@
       case WEAK_CS:
       case WEAK_NSM:
       case WEAK_BN:
+       /* FIXME */
+      case LRI:
+      case RLI:
+      case FSI:
+      case PDI:
        return WEAK;
       case NEUTRAL_B:
       case NEUTRAL_S:
       case NEUTRAL_WS:
       case NEUTRAL_ON:
        return NEUTRAL;
+      case LRE:
+      case LRO:
+      case RLE:
+      case RLO:
+      case PDF:
+#if 0
+       /* FIXME: This awaits implementation of isolate support.  */
+      case LRI:
+      case RLI:
+      case FSI:
+      case PDI:
+#endif
+       return EXPLICIT_FORMATTING;
       default:
        emacs_abort ();
     }

=== modified file 'src/dispextern.h'
--- a/src/dispextern.h  2013-12-01 22:33:13 +0000
+++ b/src/dispextern.h  2013-12-04 16:58:05 +0000
@@ -1895,6 +1895,10 @@
   RLE,         /* right-to-left embedding */
   RLO,         /* right-to-left override */
   PDF,         /* pop directional format */
+  LRI,         /* left-to-right isolate */
+  RLI,         /* right-to-left isolate */
+  FSI,         /* first strong isolate */
+  PDI,         /* pop directional isolate */
   WEAK_ES,     /* european number separator */
   WEAK_ET,     /* european number terminator */
   WEAK_CS,     /* common separator */