1 ;;; code-pages.el --- coding systems for assorted codepages -*-coding: utf-8;-*-
3 ;; Copyright (C) 2001, 2002 Free Software Foundation, Inc.
5 ;; Author: Dave Love <fx@gnu.org>
8 ;; This file is part of GNU Emacs.
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)
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.
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.
27 ;; Definitions of miscellaneous 8-bit coding systems based on ASCII
28 ;; (we can't cope properly with EBCDIC, for instance), mainly for PC
29 ;; `code pages'. They are decoded into Latin-1 and mule-unicode
30 ;; charsets rather than (lossily) into single iso8859 charsets à la
31 ;; codepage.el. The utility `cp-make-coding-system' derives them from
34 ;; Those covered are: cp437, cp737, cp720, cp775, cp850, cp851, cp852,
35 ;; cp855, cp857, cp860, cp861, cp862, cp863, cp864, cp865, cp866,
36 ;; cp869, cp874, cp1125, windows-1250, windows-1251, windows-1252,
37 ;; windows-1253, windows-1254, windows-1255, windows-1256,
38 ;; windows-1257, windows-1258, next, koi8-u, iso-8859-6,
39 ;; iso-8859-10, iso-8859-11, iso-8859-16, koi8-t, georgian-ps. This
40 ;; is meant to include all the single-byte ones relevant to GNU (used
41 ;; in glibc-defined locales); we don't yet get all the multibyte ones
44 ;; Note that various of these can clash with definitions in
45 ;; codepage.el; we try to avoid damage from that. A few CPs from
46 ;; codepage.el (770, 773, 774) aren't covered (in the absence of
47 ;; translation tables to Unicode).
49 ;; Compile this, to avoid loading `ccl' at runtime.
51 ;; Although the tables used here aren't very big, it might be worth
52 ;; splitting the file and autoloading the coding systems if/when my
53 ;; (or similar) autoloading code is installed.
57 (defun cp-make-translation-table (v)
58 "Return a translation table made from 128-long vector V.
59 V comprises characters encodable by mule-utf-8."
60 (let ((encoding-vector (make-vector 256 0)))
62 (aset encoding-vector i i))
64 (aset encoding-vector (+ i 128) (aref v i)))
65 ;; Add equivalent characters to the encoder so that we can unify
67 (let* ((tab (make-translation-table-from-vector encoding-vector))
68 ;; Translation table used for encoding:
69 (encode-table (char-table-extra-slot tab 0)))
70 (map-char-table (lambda (c v)
72 (let ((c1 (aref encode-table v)))
73 (if c1 ; we encode that unicode
74 (aset encode-table c c1)))))
75 ucs-mule-to-mule-unicode)
78 (defun cp-valid-codes (v)
79 "Derive a valid-codes list for translation vector V.
80 See `make-coding-system'."
82 (i 128) ; index into v
83 (start 0) ; start of a valid range
84 (end 127)) ; end of a valid range
86 (if (aref v (- i 128)) ; start or extend range
89 (unless start (setq start i)))
91 (push (cons start end) pairs))
94 (if start (push (cons start end) pairs))
97 (defun cp-fix-safe-chars (cs)
98 "Remove `char-coding-system-table' entries from previous definition of CS.
99 CS is a base coding system or alias."
100 (when (coding-system-p cs)
101 (let ((chars (coding-system-get cs 'safe-chars)))
104 (if (and v (not (eq v t)))
105 (aset char-coding-system-table
107 (remq cs (aref char-coding-system-table k)))))
110 ;; Fix things that have been, or might be done by codepage.el.
111 (eval-after-load "codepage"
114 (dolist (cs '(cp857 cp861 cp1253 cp852 cp866 cp437 cp855 cp869 cp775
115 cp862 cp864 cp1250 cp863 cp865 cp1251 cp737 cp1257 cp850
117 (cp-fix-safe-chars cs))
119 ;; Semi-dummy version for the stuff in codepage.el which we don't
120 ;; define here. (Used by mule-diag.)
121 (defun cp-supported-codepages ()
122 "Return an alist of supported codepages.
124 Each association in the alist has the form (NNN . CHARSET), where NNN is the
125 codepage number, and CHARSET is the MULE charset which is the closest match
126 for the character set supported by that codepage.
128 A codepage NNN is supported if a variable called `cpNNN-decode-table' exists,
129 is a vector, and has a charset property."
130 '(("774" . latin-iso8859-4) ("770" . latin-iso8859-4)
131 ("773" . latin-iso8859-4)))
133 ;; A version which doesn't override the coding systems set up by this
134 ;; file. It could still be used for the few missing ones from
136 (defun codepage-setup (codepage)
137 "Create a coding system cpCODEPAGE to support the IBM codepage CODEPAGE.
139 These coding systems are meant for encoding and decoding 8-bit non-ASCII
140 characters used by the IBM codepages, typically in conjunction with files
141 read/written by MS-DOS software, or for display on the MS-DOS terminal."
143 (let ((completion-ignore-case t)
144 (candidates (cp-supported-codepages)))
145 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
146 nil t nil nil "437"))))
147 (let ((cp (format "cp%s" codepage)))
148 (unless (coding-system-p (intern cp))
149 (cp-make-coding-systems-for-codepage
150 cp (cp-charset-for-codepage cp) (cp-offset-for-codepage cp))))))
153 ;; For `non-iso-charset-alist'. Do this after redefining
154 ;; `cp-supported-codepages', which is called through loading
158 ;; Macro to allow ccl compilation at byte-compile time, avoiding
161 (defmacro cp-make-coding-system (name v &optional doc-string mnemonic)
162 "Make coding system NAME for and 8-bit, extended-ASCII character set.
163 V is a 128-long vector of characters to translate the upper half of
164 the charactert set. DOC-STRING and MNEMONIC are used as the
165 corresponding args of `make-coding-system'. If MNEMONIC isn't given,
167 (let* ((encoder (intern (format "encode-%s" name)))
168 (decoder (intern (format "decode-%s" name)))
174 (if (r1 < 128) ;; ASCII
175 (r0 = ,(charset-id 'ascii))
177 (r0 = ,(charset-id 'eight-bit-control))
178 (r0 = ,(charset-id 'eight-bit-graphic))))
179 (translate-character ,decoder r0 r1)
180 ;; Allow fragmentation on decoding -- relevant for
181 ;; Cyrillic, Greek and, possibly Arabic and Hebrew.
182 (translate-character utf-translation-table-for-decode r0 r1)
183 (write-multibyte-character r0 r1)
189 (read-multibyte-character r0 r1)
190 (translate-character ,encoder r0 r1)
191 (if (r0 != ,(charset-id 'eight-bit-graphic))
192 (if (r0 != ,(charset-id 'eight-bit-control))
194 (write-repeat r1)))))))
195 `(let ((translation-table (cp-make-translation-table ,v))
196 (codes (cp-valid-codes ,v)))
197 (define-translation-table ',decoder translation-table)
198 (define-translation-table ',encoder
199 (char-table-extra-slot translation-table 0))
200 (cp-fix-safe-chars ',name)
202 ',name 4 ,(or mnemonic ?*)
203 (or ,doc-string (format "%s encoding" ',name))
204 (cons ,ccl-decoder ,ccl-encoder)
205 (list (cons 'safe-chars (get ',encoder 'translation-table))
206 (cons 'valid-codes codes)
207 (cons 'mime-charset ',name)
208 ;; For Quail translation. Fixme: this should really be
209 ;; a separate table that only translates the coding
210 ;; system's safe-chars.
211 (cons 'translation-table-for-input ,ucs-mule-to-mule-unicode)))
215 (let (l) ; code range
216 (dolist (elt (reverse codes))
220 non-iso-charset-alist))))
223 ;; These tables were mostly derived by running somthing like
224 ;; `recode -f cpxxx/..utf-8' on a binary file filled by
225 ;; `(dotimes (i 128) (insert ?? ?\\ (+ 128 i) ?\n))' and then
226 ;; exchanging the ?\� entries for nil. iconv was used instead in some
229 ;; Fixme: Do better for mode-line mnemonics?
231 (cp-make-coding-system
362 (cp-make-coding-system
492 (coding-system-put 'cp737 'mime-charset nil) ; not in IANA list
494 (cp-make-coding-system
625 (cp-make-coding-system
756 (cp-make-coding-system
887 (cp-make-coding-system
1018 (cp-make-coding-system
1149 (cp-make-coding-system
1280 (cp-make-coding-system
1411 (cp-make-coding-system
1542 (cp-make-coding-system
1673 (cp-make-coding-system
1804 (cp-make-coding-system
1935 (cp-make-coding-system
2066 (cp-make-coding-system
2199 (cp-make-coding-system
2330 (cp-make-coding-system
2461 (cp-make-coding-system
2593 (cp-make-coding-system
2725 (cp-make-coding-system
2856 (cp-make-coding-system
2988 (cp-make-coding-system
3120 (cp-make-coding-system
3252 (cp-make-coding-system
3384 (cp-make-coding-system
3515 (cp-make-coding-system
3646 (cp-make-coding-system
3776 "NeXTstep encoding." ?N)
3778 (cp-make-coding-system
3779 koi8-t ; used by glibc for tg_TJ
3908 "Unicode-based KOI8-T encoding for Cyrillic")
3909 (coding-system-put 'koi8-t 'mime-charset nil) ; not in the IANA list
3911 ;; Online final ISO draft:
3913 ;; http://www.evertype.com/standards/iso8859/fdis8859-16-en.pdf
3915 ;; Equivalent National Standard:
3916 ;; Romanian Standard SR 14111:1998, Romanian Standards Institution
3921 ;; "This set of coded graphic characters is intended for use in data and
3922 ;; text processing applications and also for information interchange. The
3923 ;; set contains graphic characters used for general purpose applications in
3924 ;; typical office environments in at least the following languages:
3925 ;; Albanian, Croatian, English, Finnish, French, German, Hungarian, Irish
3926 ;; Gaelic (new orthography), Italian, Latin, Polish, Romanian, and
3927 ;; Slovenian. This set of coded graphic characters may be regarded as a
3928 ;; version of an 8-bit code according to ISO/IEC 2022 or ISO/IEC 4873 at
3929 ;; level 1." [ISO 8859-16:2001(E), p. 1]
3931 ;; This charset is suitable for use in MIME text body parts.
3933 ;; ISO 8859-16 was primarily designed for single-byte encoding the Romanian
3934 ;; language. The UTF-8 charset is the preferred and in today's MIME software
3935 ;; more widely implemented encoding suitable for Romanian.
3936 (cp-make-coding-system
3937 iso-latin-10 ; consistent with, e.g. Latin-1
3938 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3939 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4036 "Unicode-based encoding for Latin-10 (MIME: ISO-8859-16)"
4038 (coding-system-put 'iso-latin-10 'mime-charset 'iso-8859-16)
4039 (define-coding-system-alias 'iso-8859-16 'iso-latin-10)
4040 (define-coding-system-alias 'latin-10 'iso-latin-10)
4042 ;; Unicode-based alternative which has the possible advantage of
4043 ;; having its relative sparseness specified.
4044 (cp-make-coding-system
4045 ;; The base system uses arabic-iso-8bit, but that's not a MIME charset.
4047 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4048 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4100 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4101 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4102 nil nil nil nil nil nil nil nil nil nil nil]
4103 "Unicode-based Arabic ISO/IEC 8859-6 (MIME: ISO-8859-6)"
4105 (define-coding-system-alias 'arabic-iso-8bit 'iso-8859-6)
4107 (cp-make-coding-system
4109 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4110 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4207 "Unicode-based encoding for Latin-6 (MIME: ISO-8859-10)")
4208 (coding-system-put 'iso-latin-6 'mime-charset 'iso-8859-10)
4209 (define-coding-system-alias 'iso-8859-10 'iso-latin-6)
4210 (define-coding-system-alias 'latin-6 'iso-latin-6)
4212 ;; used by lt_LT, lv_LV, mi_NZ
4213 (cp-make-coding-system
4215 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4216 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4314 "Unicode-based encoding for Latin-7 (MIME: ISO-8859-13)"
4315 ?l) ;; Lithuanian/Latvian
4316 (coding-system-put 'iso-latin-7 'mime-charset 'iso-8859-13)
4317 (define-coding-system-alias 'iso-8859-13 'iso-latin-7)
4318 (define-coding-system-alias 'latin-7 'iso-latin-7)
4320 (cp-make-coding-system
4321 georgian-ps ; used by glibc for ka_GE
4451 (coding-system-put 'georgian-ps 'mime-charset nil) ; not in IANA list
4453 ;; From http://www.microsoft.com/globaldev/reference/oem/720.htm
4454 (cp-make-coding-system
4584 (coding-system-put 'cp720 'mime-charset nil) ; not in IANA list
4586 ;; http://oss.software.ibm.com/cvs/icu/charset/data/ucm/ibm-1125_P100-2000.ucm
4587 (cp-make-coding-system
4717 (define-coding-system-alias 'ruscii 'cp1125)
4718 ;; Original name for cp1125, says Serhii Hlodin <hlodin@lutsk.bank.gov.ua>
4719 (define-coding-system-alias 'cp866u 'cp1125)
4721 (cp-make-coding-system
4724 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4725 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4726 ? ?ก ?ข ?ฃ ?ค ?ฅ ?ฆ ?ง ?จ ?ฉ ?ช ?ซ ?ฌ ?ญ ?ฎ ?ฏ
4727 ?ฐ ?ฑ ?ฒ ?ณ ?ด ?ต ?ถ ?ท ?ธ ?น ?บ ?ป ?ผ ?ฝ ?พ ?ฟ
4728 ?ภ ?ม ?ย ?ร ?ฤ ?ล ?ฦ ?ว ?ศ ?ษ ?ส ?ห ?ฬ ?อ ?ฮ ?ฯ
4729 ?ะ ?ั ?า ?ำ ?ิ ?ี ?ึ ?ื ?ุ ?ู ?ฺ nil nil nil nil ?฿
4730 ?เ ?แ ?โ ?ใ ?ไ ?ๅ ?ๆ ?็ ?่ ?้ ?๊ ?๋ ?์ ?ํ ?๎ ?๏
4731 ?๐ ?๑ ?๒ ?๓ ?๔ ?๕ ?๖ ?๗ ?๘ ?๙ ?๚ ?๛ nil nil nil nil
4733 "ISO-8859-11. This is `thai-tis620' with the addition of no-break-space.")
4736 (let ((w (intern (format "windows-125%d" i)))
4737 (c (intern (format "cp125%d" i))))
4738 (define-coding-system-alias c w)
4739 ;; Compatibility with codepage.el, though cp... are not the
4741 (push (assoc w non-iso-charset-alist) non-iso-charset-alist)))
4743 ;; Use Unicode font under Windows. Jason Rumney fecit.
4744 (if (and (fboundp 'w32-add-charset-info)
4745 (not (boundp 'w32-unicode-charset-defined)))
4746 (w32-add-charset-info "iso10646-1" 'w32-charset-ansi t))
4748 (provide 'code-pages)
4750 ;;; code-pages.el ends here