lisp/language/china-util.el

   1 ;;; china-util.el --- utilities for Chinese
   2
   3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
   4 ;; Licensed to the Free Software Foundation.
   5
   6 ;; Keywords: mule, multilingual, Chinese
   7
   8 ;; This file is part of GNU Emacs.
   9
  10 ;; GNU Emacs is free software; you can redistribute it and/or modify
  11 ;; it under the terms of the GNU General Public License as published by
  12 ;; the Free Software Foundation; either version 2, or (at your option)
  13 ;; any later version.
  14
  15 ;; GNU Emacs is distributed in the hope that it will be useful,
  16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18 ;; GNU General Public License for more details.
  19
  20 ;; You should have received a copy of the GNU General Public License
  21 ;; along with GNU Emacs; see the file COPYING.  If not, write to the
  22 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  23 ;; Boston, MA 02111-1307, USA.
  24
  25 ;;; Code:
  26
  27 ;;;###autoload
  28 (defun setup-chinese-gb-environment ()
  29   "Setup multilingual environment (MULE) for Chinese GB2312 users."
  30   (interactive)
  31   (set-language-environment "Chinese-GB"))
  32
  33 ;;;###autoload
  34 (defun setup-chinese-big5-environment ()
  35   "Setup multilingual environment (MULE) for Chinese Big5 users."
  36   (interactive)
  37   (set-language-environment "Chinese-BIG5"))
  38
  39 ;;;###autoload
  40 (defun setup-chinese-cns-environment ()
  41   "Setup multilingual environment (MULE) for Chinese CNS11643 family users."
  42   (interactive)
  43   (set-language-environment "Chinese-CNS"))
  44
  45 ;; Hz/ZW encoding stuffs
  46
  47 ;; HZ is an encoding method for Chinese character set GB2312 used
  48 ;; widely in Internet.  It is very similar to 7-bit environment of
  49 ;; ISO-2022.  The difference is that HZ uses the sequence "~{" and
  50 ;; "~}" for designating GB2312 and ASCII respectively, hence, it
  51 ;; doesn't uses ESC (0x1B) code.
  52
  53 ;; ZW is another encoding method for Chinese character set GB2312.  It
  54 ;; encodes Chinese characters line by line by starting each line with
  55 ;; the sequence "zW".  It also uses only 7-bit as HZ.
  56
  57 ;; ISO-2022 escape sequence to designate GB2312.
  58 (defvar iso2022-gb-designation "\e$A")
  59 ;; HZ escape sequence to designate GB2312.
  60 (defvar hz-gb-designnation "~{")
  61 ;; ISO-2022 escape sequence to designate ASCII.
  62 (defvar iso2022-ascii-designation "\e(B")
  63 ;; HZ escape sequence to designate ASCII.
  64 (defvar hz-ascii-designnation "~}")
  65 ;; Regexp of ZW sequence to start GB2312.
  66 (defvar zw-start-gb "^zW")
  67 ;; Regexp for start of GB2312 in an encoding mixture of HZ and ZW.
  68 (defvar hz/zw-start-gb
  69   (concat hz-gb-designnation "\\|" zw-start-gb "\\|[^\0-\177]"))
  70
  71 (defvar decode-hz-line-continuation nil
  72   "Flag to tell if we should care line continuation convention of Hz.")
  73
  74 (defconst hz-set-msb-table
  75   (let ((str (make-string 127 0))
  76         (i 0))
  77     (while (< i 33)
  78       (aset str i i)
  79       (setq i (1+ i)))
  80     (while (< i 127)
  81       (aset str i (+ i 128))
  82       (setq i (1+ i)))
  83     str))
  84
  85 ;;;###autoload
  86 (defun decode-hz-region (beg end)
  87   "Decode HZ/ZW encoded text in the current region.
  88 Return the length of resulting text."
  89   (interactive "r")
  90   (save-excursion
  91     (save-restriction
  92       (let (pos ch)
  93         (narrow-to-region beg end)
  94
  95         ;; We, at first, convert HZ/ZW to `euc-china',
  96         ;; then decode it.
  97
  98         ;; "~\n" -> "\n", "~~" -> "~"
  99         (goto-char (point-min))
 100         (while (search-forward "~" nil t)
 101           (setq ch (following-char))
 102           (if (or (= ch ?\n) (= ch ?~)) (delete-char -1)))
 103
 104         ;; "^zW...\n" -> Chinese GB2312
 105         ;; "~{...~}"  -> Chinese GB2312
 106         (goto-char (point-min))
 107         (setq beg nil)
 108         (while (re-search-forward hz/zw-start-gb nil t)
 109           (setq pos (match-beginning 0)
 110                 ch (char-after pos))
 111           ;; Record the first position to start conversion.
 112           (or beg (setq beg pos))
 113           (end-of-line)
 114           (setq end (point))
 115           (if (>= ch 128)               ; 8bit GB2312
 116               nil
 117             (goto-char pos)
 118             (delete-char 2)
 119             (setq end (- end 2))
 120             (if (= ch ?z)                       ; ZW -> euc-china
 121                 (progn
 122                   (translate-region (point) end hz-set-msb-table)
 123                   (goto-char end))
 124               (if (search-forward hz-ascii-designnation
 125                                   (if decode-hz-line-continuation nil end)
 126                                   t)
 127                   (delete-char -2))
 128               (setq end (point))
 129               (translate-region pos (point) hz-set-msb-table))))
 130         (if beg
 131             (decode-coding-region beg end 'euc-china)))
 132       (- (point-max) (point-min)))))
 133
 134 ;;;###autoload
 135 (defun decode-hz-buffer ()
 136   "Decode HZ/ZW encoded text in the current buffer."
 137   (interactive)
 138   (decode-hz-region (point-min) (point-max)))
 139
 140 ;;;###autoload
 141 (defun encode-hz-region (beg end)
 142   "Encode the text in the current region to HZ.
 143 Return the length of resulting text."
 144   (interactive "r")
 145   (save-excursion
 146     (save-restriction
 147       (narrow-to-region beg end)
 148
 149       ;; "~" -> "~~"
 150       (goto-char (point-min))
 151       (while (search-forward "~" nil t) (insert ?~))
 152
 153       ;; Chinese GB2312 -> "~{...~}"
 154       (goto-char (point-min))
 155       (if (re-search-forward "\\cc" nil t)
 156           (let (pos)
 157             (goto-char (setq pos (match-beginning 0)))
 158             (encode-coding-region pos (point-max) 'iso-2022-7bit)
 159             (goto-char pos)
 160             (while (search-forward iso2022-gb-designation nil t)
 161               (delete-char -3)
 162               (insert hz-gb-designnation))
 163             (goto-char pos)
 164             (while (search-forward iso2022-ascii-designation nil t)
 165               (delete-char -3)
 166               (insert hz-ascii-designnation))))
 167       (- (point-max) (point-min)))))
 168
 169 ;;;###autoload
 170 (defun encode-hz-buffer ()
 171   "Encode the text in the current buffer to HZ."
 172   (interactive)
 173   (encode-hz-region (point-min) (point-max)))
 174
 175 ;;
 176 (provide 'china-util)
 177
 178 ;;; china-util.el ends here