lisp/language/indian.el

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