[Emacs-diffs] emacs/lisp/language indian.el

[Kenichi Handa]

CVSROOT:        /cvsroot/emacs
Module name:    emacs
Changes by:     Kenichi Handa <handa>   09/12/25 02:45:47

Modified files:
        lisp/language  : indian.el 

Log message:
        (devanagari-composable-pattern): Fixed to
        handle ZWNJ and ZWJ.  Use it in composition-function-table for
        Devanagari.
        (malayalam-composable-pattern): Fix previous change.

CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/emacs/lisp/language/indian.el?cvsroot=emacs&r1=1.41&r2=1.42

Patches:
Index: indian.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/language/indian.el,v
retrieving revision 1.41
retrieving revision 1.42
diff -u -b -r1.41 -r1.42
--- indian.el   12 Dec 2009 02:28:05 -0000      1.41
+++ indian.el   25 Dec 2009 02:45:47 -0000      1.42
@@ -139,12 +139,34 @@
     regexp))
 
 (defconst devanagari-composable-pattern
+  (let ((table
+        '(("V" . "[\u0904-\u0914\u0960-\u0961\u0972]") ; independent vowel
+          ("C" . "[\u0915-\u0939]")             ; consonant
+          ("R" . "\u0930")                      ; RA
+          ("n" . "\u093C")                      ; NUKTA
+          ("H" . "\u094D")                      ; HALANT
+          ("m" . "\u093F")                      ; vowel sign (pre)
+          ("u" . "[\u0945-\u0948\u0955]")       ; vowel sign (above)
+          ("b" . "[\u0941-\u0944\u0962-\u0963]") ; vowel sign (below)
+          ("p" . "[\u093E\u0940\u0949-\u094C]") ; vowel sign (post)
+          ("A" . "[\u0900-\u0902\u0953-\u0954]") ; vowel modifier (above)
+          ("a" . "\u0903")                      ; vowel modifier (post) 
+          ("S" . "\u0951")                      ; stress sign (above)
+          ("s" . "\u0952")                      ; stress sign (below)
+          ("J" . "\u200D")                      ; ZWJ
+          ("N" . "\u200C")                      ; ZWNJ
+          ("X" . "[\u0900-\u097F]"))))          ; all coverage
+    (indian-compose-regexp
   (concat
-   "\\([अ-औॠॡ][ँं]?\\)\\|[ः।]"
-   "\\|\\("
-   
"\\(?:\\(?:[क-हक़-य़]्\\)?\\(?:[क-हक़-य़]्\\)?\\(?:[क-हक़-य़]्\\)?[क-हक़-य़]्\\)?"
-   "[क-हक़-य़]\\(?:्\\|[ा-्ॢॣ]?[ंँ]?\\)?"
-   "\\)")
+      ;; syllables with an independent vowel, or
+      "\\(?:RH\\)?Vn?m?b?u?p?n?A?s?S?a?\\|"
+      ;; consonant-based syllables, or
+      "\\(?:Cn?J?HJ?\\)*Cn?\\(?:H[NJ]?\\|m?b?u?p?n?A?s?S?a?\\)\\|"
+      ;; special consonant form, or
+      "JHR\\|"
+      ;; any other singleton characters
+      "X")
+     table))
   "Regexp matching a composable sequence of Devanagari characters.")
 
 (defconst tamil-composable-pattern
@@ -165,23 +187,24 @@
   "Regexp matching a composable sequence of Kannada characters.")
 
 (defconst malayalam-composable-pattern
-  (let ((table '(("V" . "[\u0D05-\u0D14\u0D60-\u0D61]") ; independent vowel
+  (let ((table
+        '(("V" . "[\u0D05-\u0D14\u0D60-\u0D61]") ; independent vowel
                 ("C" . "[\u0D15-\u0D39]")              ; consonant 
                 ("m" . "[\u0D46-\u0D48\u0D4A-\u0D4C]") ; prebase matra
-                ("p" . "[\u0D3E-\u0D44\u0D57]") ; postname matra
+          ("p" . "[\u0D3E-\u0D44\u0D57]")        ; postbase matra
                 ("b" . "[\u0D62-\u0D63]")       ; belowbase matra
                 ("a" . "[\u0D02-\u0D03]")       ; abovebase sign
-                ("H" . "്")                   ; virama sign 
+          ("H" . "\u0D4D")                       ; virama sign 
                 ("N" . "\u200D")                ; ZWJ
                 ("J" . "\u200C")                ; ZWNJ
                 ("X" . "[\u0D00-\u0D7F]")))) ; all coverage
     (indian-compose-regexp
      (concat
-      ;; consonant-based syllables
-      "\\(CJ?HJ?\\)*C\\(H[NJ]?\\|m?b?p?a?\\)\\|"
-      ;; syllables with an independent vowel
-      "V\\(J?HC\\)?m?b?p?a?\\|"
-      ;; special consonant form
+      ;; syllables with an independent vowel, or
+      "V\\(?:J?HC\\)?m?b?p?a?\\|"
+      ;; consonant-based syllables, or
+      "\\(?:CJ?HJ?\\)\\{0,4\\}C\\(?:H[NJ]?\\|m?b?p?a?\\)\\|"
+      ;; special consonant form, or
       "JHC\\|"
       ;; any other singleton characters
       "X")
@@ -189,7 +212,7 @@
   "Regexp matching a composable sequence of Malayalam characters.")
 
 (let ((script-regexp-alist
-       `((devanagari . "[\x900-\x97F\x200C\x200D]+")
+       `((devanagari . ,devanagari-composable-pattern)
         (bengali . "[\x980-\x9FF\x200C\x200D]+")
         (gurmukhi . "[\xA00-\xA7F\x200C\x200D]+")
         (gujarati . "[\xA80-\xAFF\x200C\x200D]+")