lisp/language/indian.el

   1 ;;; indian.el --- Indian languages support -*- coding: iso-2022-7bit; -*-
   2
   3 ;; Copyright (C) 1997, 1999, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
   4 ;;   Free Software Foundation, Inc.
   5 ;; Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008
   6 ;;   National Institute of Advanced Industrial Science and Technology (AIST)
   7 ;;   Registration Number H14PRO021
   8
   9 ;; Maintainer:  KAWABATA, Taichi <kawabata@m17n.org>
  10 ;; Keywords:    multilingual, i18n, Indian
  11
  12 ;; This file is part of GNU Emacs.
  13
  14 ;; GNU Emacs is free software; you can redistribute it and/or modify
  15 ;; it under the terms of the GNU General Public License as published by
  16 ;; the Free Software Foundation; either version 3, or (at your option)
  17 ;; any later version.
  18
  19 ;; GNU Emacs is distributed in the hope that it will be useful,
  20 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  21 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  22 ;; GNU General Public License for more details.
  23
  24 ;; You should have received a copy of the GNU General Public License
  25 ;; along with GNU Emacs; see the file COPYING.  If not, write to the
  26 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
  27 ;; Boston, MA 02110-1301, USA.
  28
  29 ;;; Commentary:
  30
  31 ;; This file defines in-is13194 coding system and relationship between
  32 ;; indian-glyph character-set and various CDAC fonts.
  33
  34 ;;; Code:
  35
  36 (make-coding-system
  37  'in-is13194 2 ?D
  38  "8-bit encoding for ASCII (MSB=0) and IS13194-Devanagari (MSB=1)."
  39  '(ascii indian-is13194 nil nil
  40    nil ascii-eol)
  41  `((safe-chars . ,(let ((table (make-char-table 'safe-chars nil)))
  42                     (set-char-table-range table 'indian-is13194 t)
  43                     (dotimes (i 127)
  44                       (aset table i t)
  45                       (aset table (decode-char 'ucs (+ #x900 i)) t))
  46                     table))
  47    (post-read-conversion . in-is13194-post-read-conversion)
  48    (pre-write-conversion . in-is13194-pre-write-conversion)))
  49
  50 (define-coding-system-alias 'devanagari 'in-is13194)
  51
  52 (defvar indian-font-foundry 'cdac
  53   "Font foundry for Indian characters.
  54 Currently supported foundries are `cdac' and `akruti'.")
  55
  56 (defvar indian-script-language-alist
  57   '((devanagari (hindi sanskrit) nil)
  58     (bengali (bengali assamese) nil)
  59     (gurmukhi (punjabi) nil)
  60     (gujarati (gujarati) nil)
  61     (oriya (oriya) nil)
  62     (tamil (tamil) nil)
  63     (telugu (telugu) nil)
  64     (kannada (kannada) nil)
  65     (malayalam (malayalam) nil))
  66   "Alist of Indian scripts vs the corresponding language list and font foundry.
  67 Each element has this form:
  68
  69   (SCRIPT LANGUAGE-LIST FONT-FOUNDRY)
  70
  71 SCRIPT is one of Indian script names.
  72
  73 LANGUAGE-LIST is a list of Indian langauge names SCRIPT is used for.
  74 The list is in the priority order.
  75
  76 FONT-FOUNDRY is a font foundry representing a group of Indian
  77 fonts.  If the value is nil, the value of `indian-font-foundry'
  78 is used.")
  79
  80 (defconst indian-font-char-index-table
  81   '(                                    ; for which language(s)
  82     ;; CDAC fonts
  83     (#x0000 . cdac:dv-ttsurekh)         ; hindi, etc
  84     (#x0100 . cdac:sd-ttsurekh)         ; sanskrit
  85     (#x0200 . cdac:bn-ttdurga)          ; bengali
  86     (#x0300 . cdac:tm-ttvalluvar)       ; tamil
  87     (#x0400 . cdac:tl-tthemalatha)      ; telugu
  88     (#x0500 . cdac:as-ttdurga)          ; assamese
  89     (#x0600 . cdac:or-ttsarala)         ; oriya
  90     (#x0700 . cdac:kn-ttuma)            ; kannada
  91     (#x0800 . cdac:ml-ttkarthika)       ; malayalam
  92     (#x0900 . cdac:gj-ttavantika)       ; gujarati
  93     (#x0A00 . cdac:pn-ttamar)           ; punjabi
  94
  95     ;; AKRUTI fonts
  96     (#x0B00 . akruti:dev)               ; hindi, etc
  97     (#x0C00 . akruti:bng)               ; bengali
  98     (#x0D00 . akruti:pnj)               ; punjabi
  99     (#x0E00 . akruti:guj)               ; gujarati
 100     (#x0F00 . akruti:ori)               ; oriya
 101     (#x1000 . akruti:tml)               ; tamil
 102     (#x1100 . akruti:tlg)               ; telugu
 103     (#x1200 . akruti:knd)               ; kannada
 104     (#x1300 . akruti:mal)               ; malayalam
 105     )
 106   "Alist of indices of `indian-glyph' character vs Indian font identifiers.
 107 Each element has this form: (INDEX . FONT-IDENTIFIER)
 108
 109 INDEX is an index number of the first character in the charset
 110 `indian-glyph' assigned for glyphs in the font specified by
 111 FONT-IDENTIFIER.  Currently FONT-IDENTIFIERs are defined for CDAC
 112 and AKRUTI font groups.")
 113
 114 (defun indian-font-char (index font-identifier)
 115   "Return character of charset `indian-glyph' made from glyph index INDEX.
 116 FONT-IDENTIFIER is an identifier of an Indian font listed in the
 117 variable `indian-font-char-index-table'.  It specifies which
 118 font INDEX is for."
 119   (if (or (< index 0) (> index 255))
 120       (error "Invalid glyph index: %d" index))
 121   (let ((start (car (rassq font-identifier indian-font-char-index-table))))
 122     (if (not start)
 123         (error "Unknown font identifier: %s" font-identifier))
 124     (setq index (+ start index))
 125     (make-char 'indian-glyph (+ (/ index 96) 32) (+ (% index 96) 32))))
 126
 127 ;; Return a range of characters (cons of min and max character) of the
 128 ;; charset `indian-glyph' for displaying SCRIPT in LANGUAGE by a font
 129 ;; of FOUNDRY.
 130
 131 (defun indian-font-char-range (font-identifier)
 132   (cons (indian-font-char 0 font-identifier)
 133         (indian-font-char 255 font-identifier)))
 134
 135 (defvar indian-script-table
 136   '[
 137     devanagari
 138     sanskrit
 139     bengali
 140     tamil
 141     telugu
 142     assamese
 143     oriya
 144     kannada
 145     malayalam
 146     gujarati
 147     punjabi
 148     ]
 149   "Vector of Indian script names.")
 150
 151 (let ((len (length indian-script-table))
 152       (i 0))
 153   (while (< i len)
 154     (put (aref indian-script-table i) 'indian-glyph-code-offset (* 256 i))
 155     (setq i (1+ i))))
 156
 157 (defvar indian-default-script 'devanagari
 158   "Default script for Indian languages.
 159 Each Indian language environment sets this value
 160 to one of `indian-script-table' (which see).
 161 The default value is `devanagari'.")
 162
 163 (define-ccl-program ccl-encode-indian-glyph-font
 164   `(0
 165     ;; Shorten (r1 = (((((r1 - 32) * 96) + r2) - 32) % 256))
 166     (r1 = ((((r1 * 96) + r2) - ,(+ (* 32 96) 32)) % 256))))
 167
 168 (setq font-ccl-encoder-alist
 169       (cons (cons "-CDAC" 'ccl-encode-indian-glyph-font)
 170             font-ccl-encoder-alist))
 171
 172 (setq font-ccl-encoder-alist
 173       (cons (cons "-AKRUTI" 'ccl-encode-indian-glyph-font)
 174             font-ccl-encoder-alist))
 175
 176 (provide 'indian)
 177
 178 ;;; arch-tag: 83aa8fc7-7ee2-4364-a6e5-498f5e3b8c2f
 179 ;;; indian.el ends here