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 (require 'mule-diag) ; for non-iso-charset-alist
59 (defun cp-make-translation-table (v)
60 "Return a translation table made from 128-long vector V.
61 V comprises characters encodable by mule-utf-8."
62 (let ((encoding-vector (make-vector 256 0)))
64 (aset encoding-vector i i))
66 (aset encoding-vector (+ i 128) (aref v i)))
67 ;; Add equivalent characters to the encoder so that we can unify
69 (let* ((tab (make-translation-table-from-vector encoding-vector))
70 ;; Translation table used for encoding:
71 (encode-table (char-table-extra-slot tab 0)))
72 (map-char-table (lambda (c v)
74 (let ((c1 (aref encode-table v)))
75 (if c1 ; we encode that unicode
76 (aset encode-table c c1)))))
77 ucs-mule-to-mule-unicode)
80 (defun cp-valid-codes (v)
81 "Derive a valid-codes list for translation vector V.
82 See `make-coding-system'."
84 (i 128) ; index into v
85 (start 0) ; start of a valid range
86 (end 127)) ; end of a valid range
88 (if (aref v (- i 128)) ; start or extend range
91 (unless start (setq start i)))
93 (push (cons start end) pairs))
96 (if start (push (cons start end) pairs))
99 (defun cp-fix-safe-chars (cs)
100 "Remove `char-coding-system-table' entries from previous definition of CS.
101 CS is a base coding system or alias."
102 (when (coding-system-p cs)
103 (let ((chars (coding-system-get cs 'safe-chars)))
106 (if (and v (not (eq v t)))
107 (aset char-coding-system-table
109 (remq cs (aref char-coding-system-table k)))))
112 ;; Fix things that have been, or might be done by codepage.el.
113 (eval-after-load "codepage"
116 (dolist (cs '(cp857 cp861 cp1253 cp852 cp866 cp437 cp855 cp869 cp775
117 cp862 cp864 cp1250 cp863 cp865 cp1251 cp737 cp1257 cp850
119 (cp-fix-safe-chars cs))
121 ;; Semi-dummy version for the stuff in codepage.el which we don't
122 ;; define here. (Used by mule-diag.)
123 (defun cp-supported-codepages ()
124 "Return an alist of supported codepages.
126 Each association in the alist has the form (NNN . CHARSET), where NNN is the
127 codepage number, and CHARSET is the MULE charset which is the closest match
128 for the character set supported by that codepage.
130 A codepage NNN is supported if a variable called `cpNNN-decode-table' exists,
131 is a vector, and has a charset property."
132 '(("774" . latin-iso8859-4) ("770" . latin-iso8859-4)
133 ("773" . latin-iso8859-4)))
135 ;; A version which doesn't override the coding systems set up by this
136 ;; file. It could still be used for the few missing ones from
138 (defun codepage-setup (codepage)
139 "Create a coding system cpCODEPAGE to support the IBM codepage CODEPAGE.
141 These coding systems are meant for encoding and decoding 8-bit non-ASCII
142 characters used by the IBM codepages, typically in conjunction with files
143 read/written by MS-DOS software, or for display on the MS-DOS terminal."
145 (let ((completion-ignore-case t)
146 (candidates (cp-supported-codepages)))
147 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
148 nil t nil nil "437"))))
149 (let ((cp (format "cp%s" codepage)))
150 (unless (coding-system-p (intern cp))
151 (cp-make-coding-systems-for-codepage
152 cp (cp-charset-for-codepage cp) (cp-offset-for-codepage cp))))))
155 ;; Macro to allow ccl compilation at byte-compile time, avoiding
158 (defmacro cp-make-coding-system (name v &optional doc-string mnemonic)
159 "Make coding system NAME for and 8-bit, extended-ASCII character set.
160 V is a 128-long vector of characters to translate the upper half of
161 the charactert set. DOC-STRING and MNEMONIC are used as the
162 corresponding args of `make-coding-system'. If MNEMONIC isn't given,
164 (let* ((encoder (intern (format "encode-%s" name)))
165 (decoder (intern (format "decode-%s" name)))
171 (if (r1 < 128) ;; ASCII
172 (r0 = ,(charset-id 'ascii))
174 (r0 = ,(charset-id 'eight-bit-control))
175 (r0 = ,(charset-id 'eight-bit-graphic))))
176 (translate-character ,decoder r0 r1)
177 ;; Allow fragmentation on decoding -- relevant for
178 ;; Cyrillic, Greek and, possibly Arabic and Hebrew.
179 (translate-character utf-translation-table-for-decode r0 r1)
180 (write-multibyte-character r0 r1)
186 (read-multibyte-character r0 r1)
187 (translate-character ,encoder r0 r1)
188 (write-repeat r1)))))))
189 `(let ((translation-table (cp-make-translation-table ,v))
190 (codes (cp-valid-codes ,v)))
191 (define-translation-table ',decoder translation-table)
192 (define-translation-table ',encoder
193 (char-table-extra-slot translation-table 0))
194 (cp-fix-safe-chars ',name)
196 ',name 4 ,(or mnemonic ?*)
197 (or ,doc-string (format "%s encoding" ',name))
198 (cons ,ccl-decoder ,ccl-encoder)
199 (list (cons 'safe-chars (get ',encoder 'translation-table))
200 (cons 'valid-codes codes)
201 (cons 'mime-charset ',name)
202 ;; For Quail translation. Fixme: this should really be
203 ;; a separate table that only translates the coding
204 ;; system's safe-chars.
205 (cons 'translation-table-for-input ,ucs-mule-to-mule-unicode)))
209 (let (l) ; code range
210 (dolist (elt (reverse codes))
214 non-iso-charset-alist))))
217 ;; These tables were mostly derived by running somthing like
218 ;; `recode -f cpxxx/..utf-8' on a binary file filled by
219 ;; `(dotimes (i 128) (insert ?? ?\\ (+ 128 i) ?\n))' and then
220 ;; exchanging the ?\� entries for nil. iconv was used instead in some
223 ;; Fixme: Do better for mode-line mnemonics?
225 (cp-make-coding-system
356 (cp-make-coding-system
486 (coding-system-put 'cp737 'mime-charset nil) ; not in IANA list
488 (cp-make-coding-system
619 (cp-make-coding-system
750 (cp-make-coding-system
881 (cp-make-coding-system
1012 (cp-make-coding-system
1143 (cp-make-coding-system
1274 (cp-make-coding-system
1405 (cp-make-coding-system
1536 (cp-make-coding-system
1667 (cp-make-coding-system
1798 (cp-make-coding-system
1929 (cp-make-coding-system
2060 (cp-make-coding-system
2193 (cp-make-coding-system
2324 (cp-make-coding-system
2455 (cp-make-coding-system
2587 (cp-make-coding-system
2719 (cp-make-coding-system
2850 (cp-make-coding-system
2982 (cp-make-coding-system
3114 (cp-make-coding-system
3246 (cp-make-coding-system
3378 (cp-make-coding-system
3509 (cp-make-coding-system
3640 (cp-make-coding-system
3770 "NeXTstep encoding." ?N)
3772 (cp-make-coding-system
3773 koi8-t ; used by glibc for tg_TJ
3902 "Unicode-based KOI8-T encoding for Cyrillic")
3903 (coding-system-put 'koi8-t 'mime-charset nil) ; not in the IANA list
3905 ;; Online final ISO draft:
3907 ;; http://www.evertype.com/standards/iso8859/fdis8859-16-en.pdf
3909 ;; Equivalent National Standard:
3910 ;; Romanian Standard SR 14111:1998, Romanian Standards Institution
3915 ;; "This set of coded graphic characters is intended for use in data and
3916 ;; text processing applications and also for information interchange. The
3917 ;; set contains graphic characters used for general purpose applications in
3918 ;; typical office environments in at least the following languages:
3919 ;; Albanian, Croatian, English, Finnish, French, German, Hungarian, Irish
3920 ;; Gaelic (new orthography), Italian, Latin, Polish, Romanian, and
3921 ;; Slovenian. This set of coded graphic characters may be regarded as a
3922 ;; version of an 8-bit code according to ISO/IEC 2022 or ISO/IEC 4873 at
3923 ;; level 1." [ISO 8859-16:2001(E), p. 1]
3925 ;; This charset is suitable for use in MIME text body parts.
3927 ;; ISO 8859-16 was primarily designed for single-byte encoding the Romanian
3928 ;; language. The UTF-8 charset is the preferred and in today's MIME software
3929 ;; more widely implemented encoding suitable for Romanian.
3930 (cp-make-coding-system
3931 iso-latin-10 ; consistent with, e.g. Latin-1
3932 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3933 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4030 "Unicode-based encoding for Latin-10 (MIME: ISO-8859-16)"
4032 (coding-system-put 'iso-latin-10 'mime-charset 'iso-8859-16)
4033 (define-coding-system-alias 'iso-8859-16 'iso-latin-10)
4034 (define-coding-system-alias 'latin-10 'iso-latin-10)
4036 ;; Unicode-based alternative which has the possible advantage of
4037 ;; having its relative sparseness specified.
4038 (cp-make-coding-system
4039 ;; The base system uses arabic-iso-8bit, but that's not a MIME charset.
4041 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4042 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4094 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4095 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4096 nil nil nil nil nil nil nil nil nil nil nil]
4097 "Unicode-based Arabic ISO/IEC 8859-6 (MIME: ISO-8859-6)"
4099 (define-coding-system-alias 'arabic-iso-8bit 'iso-8859-6)
4101 (cp-make-coding-system
4103 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4104 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4201 "Unicode-based encoding for Latin-6 (MIME: ISO-8859-10)")
4202 (coding-system-put 'iso-latin-6 'mime-charset 'iso-8859-10)
4203 (define-coding-system-alias 'iso-8859-10 'iso-latin-6)
4204 (define-coding-system-alias 'latin-6 'iso-latin-6)
4206 ;; used by lt_LT, lv_LV, mi_NZ
4207 (cp-make-coding-system
4209 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4210 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4308 "Unicode-based encoding for Latin-7 (MIME: ISO-8859-13)"
4309 ?l) ;; Lithuanian/Latvian
4310 (coding-system-put 'iso-latin-7 'mime-charset 'iso-8859-13)
4311 (define-coding-system-alias 'iso-8859-13 'iso-latin-7)
4312 (define-coding-system-alias 'latin-7 'iso-latin-7)
4314 (cp-make-coding-system
4315 georgian-ps ; used by glibc for ka_GE
4445 (coding-system-put 'georgian-ps 'mime-charset nil) ; not in IANA list
4447 ;; From http://www.microsoft.com/globaldev/reference/oem/720.htm
4448 (cp-make-coding-system
4578 (coding-system-put 'cp720 'mime-charset nil) ; not in IANA list
4580 ;; http://oss.software.ibm.com/cvs/icu/charset/data/ucm/ibm-1125_P100-2000.ucm
4581 (cp-make-coding-system
4711 (define-coding-system-alias 'ruscii 'cp1125)
4712 ;; Original name for cp1125, says Serhii Hlodin <hlodin@lutsk.bank.gov.ua>
4713 (define-coding-system-alias 'cp866u 'cp1125)
4715 (cp-make-coding-system
4718 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4719 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4720 ? ?ก ?ข ?ฃ ?ค ?ฅ ?ฆ ?ง ?จ ?ฉ ?ช ?ซ ?ฌ ?ญ ?ฎ ?ฏ
4721 ?ฐ ?ฑ ?ฒ ?ณ ?ด ?ต ?ถ ?ท ?ธ ?น ?บ ?ป ?ผ ?ฝ ?พ ?ฟ
4722 ?ภ ?ม ?ย ?ร ?ฤ ?ล ?ฦ ?ว ?ศ ?ษ ?ส ?ห ?ฬ ?อ ?ฮ ?ฯ
4723 ?ะ ?ั ?า ?ำ ?ิ ?ี ?ึ ?ื ?ุ ?ู ?ฺ nil nil nil nil ?฿
4724 ?เ ?แ ?โ ?ใ ?ไ ?ๅ ?ๆ ?็ ?่ ?้ ?๊ ?๋ ?์ ?ํ ?๎ ?๏
4725 ?๐ ?๑ ?๒ ?๓ ?๔ ?๕ ?๖ ?๗ ?๘ ?๙ ?๚ ?๛ nil nil nil nil
4727 "ISO-8859-11. This is `thai-tis620' with the addition of no-break-space.")
4730 (let ((w (intern (format "windows-125%d" i)))
4731 (c (intern (format "cp125%d" i))))
4732 (define-coding-system-alias c w)
4733 ;; Compatibility with codepage.el, though cp... are not the
4735 (push (assoc w non-iso-charset-alist) non-iso-charset-alist)))
4737 ;; Use Unicode font under Windows. Jason Rumney fecit.
4738 (if (and (fboundp 'w32-add-charset-info)
4739 (not (boundp 'w32-unicode-charset-defined)))
4740 (w32-add-charset-info "iso10646-1" 'w32-charset-ansi t))
4742 (provide 'code-pages)
4744 ;;; code-pages.el ends here