1 ;;; code-pages.el --- coding systems for assorted codepages -*-coding: utf-8;-*-
3 ;; Copyright (C) 2001, 2002, 2003, 2004, 2005 Free Software Foundation, Inc.
4 ;; Copyright (C) 2004, 2005
5 ;; National Institute of Advanced Industrial Science and Technology (AIST)
6 ;; Registration Number H14PRO021
8 ;; Author: Dave Love <fx@gnu.org>
11 ;; This file is part of GNU Emacs.
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)
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.
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., 59 Temple Place - Suite 330,
26 ;; Boston, MA 02111-1307, USA.
30 ;; Definitions of miscellaneous 8-bit coding systems based on ASCII
31 ;; (we can't cope properly with EBCDIC, for instance), mainly for PC
32 ;; `code pages'. They are decoded into Latin-1 and mule-unicode
33 ;; charsets rather than (lossily) into single iso8859 charsets à la
34 ;; codepage.el. The utility `cp-make-coding-system' derives them from
37 ;; Those covered are: cp437, cp737, cp720, cp775, cp850, cp851, cp852,
38 ;; cp855, cp857, cp860, cp861, cp862, cp863, cp864, cp865, cp866,
39 ;; cp869, cp874, cp1125, windows-1250, windows-1253, windows-1254,
40 ;; windows-1255, windows-1256, windows-1257, windows-1258, next,
41 ;; iso-8859-6, iso-8859-10, iso-8859-11, iso-8859-16, koi8-t,
42 ;; georgian-ps. This is meant to include all the single-byte ones
43 ;; relevant to GNU (used in glibc-defined locales); we don't yet get
44 ;; all the multibyte ones in base Emacs.
46 ;; Note that various of these can clash with definitions in
47 ;; codepage.el; we try to avoid damage from that. A few CPs from
48 ;; codepage.el (770, 773, 774) aren't covered (in the absence of
49 ;; translation tables to Unicode).
51 ;; Compile this, to avoid loading `ccl' at runtime.
53 ;; Although the tables used here aren't very big, it might be worth
54 ;; splitting the file and autoloading the coding systems if/when my
55 ;; (or similar) autoloading code is installed.
59 ;; The defsubsts here are just so that language files can use
60 ;; `cp-make-coding-system' and not require functions from this file
63 (defsubst cp-make-translation-table
(v)
64 "Return a translation table made from 128-long vector V.
65 V comprises characters encodable by mule-utf-8."
66 (let ((encoding-vector (make-vector 256 0)))
68 (aset encoding-vector i i
))
70 (aset encoding-vector
(+ i
128) (aref v i
)))
71 ;; Add equivalent characters to the encoder so that we can unify
73 (let* ((tab (make-translation-table-from-vector encoding-vector
))
74 ;; Translation table used for encoding:
75 (encode-table (char-table-extra-slot tab
0)))
76 (map-char-table (lambda (c v
)
78 (let ((c1 (aref encode-table v
)))
79 (if c1
; we encode that unicode
80 (aset encode-table c c1
)))))
81 ucs-mule-to-mule-unicode
)
84 (defsubst cp-valid-codes
(v)
85 "Derive a valid-codes list for translation vector V.
86 See `make-coding-system'."
88 (i 128) ; index into v
89 (start 0) ; start of a valid range
90 (end 127)) ; end of a valid range
92 (if (aref v
(- i
128)) ; start or extend range
95 (unless start
(setq start i
)))
97 (push (cons start end
) pairs
))
100 (if start
(push (cons start end
) pairs
))
103 ;; Fix things that have been, or might be, done by codepage.el.
104 (eval-after-load "codepage"
107 ;; Semi-dummy version for the stuff in codepage.el which we don't
108 ;; define here. (Used by mule-diag.)
109 (defun cp-supported-codepages ()
110 "Return an alist of supported codepages.
112 Each association in the alist has the form (NNN . CHARSET), where NNN is the
113 codepage number, and CHARSET is the MULE charset which is the closest match
114 for the character set supported by that codepage.
116 A codepage NNN is supported if a variable called `cpNNN-decode-table' exists,
117 is a vector, and has a charset property."
118 '(("774" . latin-iso8859-4
) ("770" . latin-iso8859-4
)
119 ("773" . latin-iso8859-4
)))
121 ;; A version which doesn't override the coding systems set up by this
122 ;; file. It could still be used for the few missing ones from
124 (defun codepage-setup (codepage)
125 "Create a coding system cpCODEPAGE to support the IBM codepage CODEPAGE.
127 These coding systems are meant for encoding and decoding 8-bit non-ASCII
128 characters used by the IBM codepages, typically in conjunction with files
129 read/written by MS-DOS software, or for display on the MS-DOS terminal."
131 (let ((completion-ignore-case t
)
132 (candidates (cp-supported-codepages)))
133 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
134 nil t nil nil
"437"))))
135 (let ((cp (format "cp%s" codepage
)))
136 (unless (coding-system-p (intern cp
))
137 (cp-make-coding-systems-for-codepage
138 cp
(cp-charset-for-codepage cp
) (cp-offset-for-codepage cp
))))))
141 ;; Macro to allow ccl compilation at byte-compile time, avoiding
144 (defmacro cp-make-coding-system
(name v
&optional doc-string mnemonic
)
145 "Make coding system NAME for and 8-bit, extended-ASCII character set.
146 V is a 128-long vector of characters to translate the upper half of
147 the character set. DOC-STRING and MNEMONIC are used as the
148 corresponding args of `make-coding-system'. If MNEMONIC isn't given,
150 Return an updated `non-iso-charset-alist'."
151 (let* ((encoder (intern (format "encode-%s" name
)))
152 (decoder (intern (format "decode-%s" name
)))
158 (if (r1 < 128) ;; ASCII
159 (r0 = ,(charset-id 'ascii
))
161 (r0 = ,(charset-id 'eight-bit-control
))
162 (r0 = ,(charset-id 'eight-bit-graphic
))))
163 (translate-character ,decoder r0 r1
)
164 ;; Allow fragmentation on decoding -- relevant for
165 ;; Cyrillic, Greek and, possibly Arabic and Hebrew.
166 (translate-character utf-translation-table-for-decode r0 r1
)
167 (write-multibyte-character r0 r1
)
173 (read-multibyte-character r0 r1
)
174 (translate-character ,encoder r0 r1
)
175 (if (r0 != ,(charset-id 'ascii
))
176 (if (r0 != ,(charset-id 'eight-bit-graphic
))
177 (if (r0 != ,(charset-id 'eight-bit-control
))
179 (write-repeat r1
)))))))
180 `(let ((translation-table (cp-make-translation-table ,v
))
181 (codes (cp-valid-codes ,v
)))
182 (define-translation-table ',decoder translation-table
)
183 (define-translation-table ',encoder
184 (char-table-extra-slot translation-table
0))
186 ',name
4 ,(or mnemonic ?
*)
187 (or ,doc-string
(format "%s encoding" ',name
))
188 (cons ,ccl-decoder
,ccl-encoder
)
189 (list (cons 'safe-chars
(get ',encoder
'translation-table
))
190 (cons 'valid-codes codes
)
191 (cons 'mime-charset
',name
)
192 ;; For Quail translation. Fixme: this should really be
193 ;; a separate table that only translates the coding
194 ;; system's safe-chars.
195 (cons 'translation-table-for-input
'ucs-mule-to-mule-unicode
)))
196 (let ((slot (assq ',name non-iso-charset-alist
))
197 (elt (list nil
; charset list
199 (let (l) ; code range
200 (dolist (elt (reverse codes
))
205 (push (cons ',name elt
) non-iso-charset-alist
)
207 non-iso-charset-alist
)))))
209 (eval-when-compile (defvar non-iso-charset-alist
))
211 ;; These tables were mostly derived by running somthing like
212 ;; `recode -f cpxxx/..utf-8' on a binary file filled by
213 ;; `(dotimes (i 128) (insert ?? ?\\ (+ 128 i) ?\n))' and then
214 ;; exchanging the ?\� entries for nil. iconv was used instead in some
217 ;; Fixme: Do better for mode-line mnemonics?
219 (cp-make-coding-system
350 (cp-make-coding-system
480 (coding-system-put 'cp737
'mime-charset nil
) ; not in IANA list
482 (cp-make-coding-system
613 (cp-make-coding-system
744 (cp-make-coding-system
875 (cp-make-coding-system
1006 (cp-make-coding-system
1137 (cp-make-coding-system
1268 (cp-make-coding-system
1399 (cp-make-coding-system
1530 (cp-make-coding-system
1661 (cp-make-coding-system
1792 (cp-make-coding-system
1923 (cp-make-coding-system
2054 (cp-make-coding-system
2187 (cp-make-coding-system
2318 (cp-make-coding-system
2449 ;;;###autoload(autoload-coding-system 'windows-1250 '(require 'code-pages))
2450 (cp-make-coding-system
2581 ;;;###autoload(autoload-coding-system 'windows-1253 '(require 'code-pages))
2582 (cp-make-coding-system
2714 ;;;###autoload(autoload-coding-system 'windows-1254 '(require 'code-pages))
2715 (cp-make-coding-system
2847 ;;;###autoload(autoload-coding-system 'windows-1255 '(require 'code-pages))
2848 (cp-make-coding-system
2980 ;;;###autoload(autoload-coding-system 'windows-1256 '(require 'code-pages))
2981 (cp-make-coding-system
3113 ;;;###autoload(autoload-coding-system 'windows-1257 '(require 'code-pages))
3114 (cp-make-coding-system
3245 ;;;###autoload(autoload-coding-system 'windows-1258 '(require 'code-pages))
3246 (cp-make-coding-system
3377 (cp-make-coding-system
3507 "NeXTstep encoding." ?N
)
3509 (cp-make-coding-system
3510 koi8-t
; used by glibc for tg_TJ
3639 "Unicode-based KOI8-T encoding for Cyrillic")
3640 (coding-system-put 'koi8-t
'mime-charset nil
) ; not in the IANA list
3641 (define-coding-system-alias 'cyrillic-koi8-t
'koi8-t
)
3643 ;; Online final ISO draft:
3645 ;; http://www.evertype.com/standards/iso8859/fdis8859-16-en.pdf
3647 ;; Equivalent National Standard:
3648 ;; Romanian Standard SR 14111:1998, Romanian Standards Institution
3653 ;; "This set of coded graphic characters is intended for use in data and
3654 ;; text processing applications and also for information interchange. The
3655 ;; set contains graphic characters used for general purpose applications in
3656 ;; typical office environments in at least the following languages:
3657 ;; Albanian, Croatian, English, Finnish, French, German, Hungarian, Irish
3658 ;; Gaelic (new orthography), Italian, Latin, Polish, Romanian, and
3659 ;; Slovenian. This set of coded graphic characters may be regarded as a
3660 ;; version of an 8-bit code according to ISO/IEC 2022 or ISO/IEC 4873 at
3661 ;; level 1." [ISO 8859-16:2001(E), p. 1]
3663 ;; This charset is suitable for use in MIME text body parts.
3665 ;; ISO 8859-16 was primarily designed for single-byte encoding the Romanian
3666 ;; language. The UTF-8 charset is the preferred and in today's MIME software
3667 ;; more widely implemented encoding suitable for Romanian.
3668 ;;;###autoload(autoload-coding-system 'iso-8859-16 '(require 'code-pages))
3669 (cp-make-coding-system
3670 iso-latin-10
; consistent with, e.g. Latin-1
3671 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3672 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3769 "Unicode-based encoding for Latin-10 (MIME: ISO-8859-16)"
3771 (coding-system-put 'iso-latin-10
'mime-charset
'iso-8859-16
)
3772 (define-coding-system-alias 'iso-8859-16
'iso-latin-10
)
3773 (define-coding-system-alias 'latin-10
'iso-latin-10
)
3775 ;; Unicode-based alternative which has the possible advantage of
3776 ;; having its relative sparseness specified.
3777 (cp-make-coding-system
3778 ;; The base system uses arabic-iso-8bit, but that's not a MIME charset.
3780 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3781 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3833 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3834 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3835 nil nil nil nil nil nil nil nil nil nil nil
]
3836 "Unicode-based Arabic ISO/IEC 8859-6 (MIME: ISO-8859-6)"
3838 (define-coding-system-alias 'arabic-iso-8bit
'iso-8859-6
)
3840 ;;;###autoload(autoload-coding-system 'iso-8859-10 '(require 'code-pages))
3841 (cp-make-coding-system
3843 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3844 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3941 "Unicode-based encoding for Latin-6 (MIME: ISO-8859-10)")
3942 (coding-system-put 'iso-latin-6
'mime-charset
'iso-8859-10
)
3943 (define-coding-system-alias 'iso-8859-10
'iso-latin-6
)
3944 (define-coding-system-alias 'latin-6
'iso-latin-6
)
3946 ;; used by lt_LT, lv_LV, mi_NZ
3947 ;;;###autoload(autoload-coding-system 'iso-8859-13 '(require 'code-pages))
3948 (cp-make-coding-system
3950 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3951 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4049 "Unicode-based encoding for Latin-7 (MIME: ISO-8859-13)"
4050 ?l
) ;; Lithuanian/Latvian
4051 (coding-system-put 'iso-latin-7
'mime-charset
'iso-8859-13
)
4052 (define-coding-system-alias 'iso-8859-13
'iso-latin-7
)
4053 (define-coding-system-alias 'latin-7
'iso-latin-7
)
4055 ;; Fixme: check on the C1 characters which libiconv includes. They
4056 ;; are reproduced below, but are probably wrong. I can't find an
4057 ;; official definition of georgian-ps.
4058 ;;;###autoload(autoload-coding-system 'georgian-ps '(require 'code-pages))
4059 (cp-make-coding-system
4060 georgian-ps
; used by glibc for ka_GE
4190 (coding-system-put 'georgian-ps
'mime-charset nil
) ; not in IANA list
4192 ;; From http://www.microsoft.com/globaldev/reference/oem/720.htm
4193 (cp-make-coding-system
4323 (coding-system-put 'cp720
'mime-charset nil
) ; not in IANA list
4325 ;; http://oss.software.ibm.com/cvs/icu/charset/data/ucm/ibm-1125_P100-2000.ucm
4326 (cp-make-coding-system
4456 (define-coding-system-alias 'ruscii
'cp1125
)
4457 ;; Original name for cp1125, says Serhii Hlodin <hlodin@lutsk.bank.gov.ua>
4458 (define-coding-system-alias 'cp866u
'cp1125
)
4459 (coding-system-put 'cp1125
'mime-charset nil
)
4461 ;; Suggested by Anton Zinoviev <anton@lml.bas.bg>: Bulgarian DOS
4462 ;; codepage. Table at
4463 ;; <URL:http://czyborra.com/charsets/bulgarian-mik.txt.gz>.
4464 (cp-make-coding-system
4466 [?А ?Б ?В ?Г ?Д ?Е ?Ж ?З ?И ?Й ?К ?Л ?М ?Н ?О ?П ?Р ?С ?Т ?У ?Ф ?Х ?Ц
4467 ?Ч ?Ш ?Щ ?Ъ ?Ы ?Ь ?Э ?Ю ?Я ?а ?б ?в ?г ?д ?е ?ж ?з ?и ?й ?к ?л ?м ?н
4468 ?о ?п ?р ?с ?т ?у ?ф ?х ?ц ?ч ?ш ?щ ?ъ ?ы ?ь ?э ?ю ?я ?└ ?┴ ?┬ ?├ ?─
4469 ?┼ ?╣ ?║ ?╚ ?╔ ?╩ ?╦ ?╠ ?═ ?╬ ?┐ ?░ ?▒ ?▓ ?│ ?┤ ?№ ?§ ?╗ ?╝ ?┘ ?┌ ?█
4470 ?▄ ?▌ ?▐ ?▀ ?α ?β ?Γ ?π ?Σ ?σ ?μ ?τ ?Φ ?Θ ?Ω ?δ ?∞ ?∅ ?∈ ?∩ ?≡ ?± ?≥
4471 ?≤ ?⌠ ?⌡ ?÷ ?≈ ?° ?∙ ?· ?√ ?ⁿ ?² ?■ ?
])
4472 (coding-system-put 'mik
'mime-charset nil
)
4474 ;; Suggested by Anton Zinoviev <anton@lml.bas.bg>: similar to CP1251
4475 ;; and used for some non-Slavic Cyrillic languages. Table found at
4476 ;; <URL:ftp://ftp.logic.ru/pub/logic/linux/cyr-asian/PT154>. See also
4477 ;; <URL:http://lists.w3.org/Archives/Public/ietf-charsets/2002AprJun/0092.html,
4478 ;; which suggests it's used in an Asian Cyrillic context.
4479 ;;;###autoload(autoload-coding-system 'pt154 '(require 'code-pages))
4480 (cp-make-coding-system
4482 [?Җ ?Ғ ?Ӯ ?ғ ?„ ?… ?Ҷ ?Ү ?Ҳ ?ү ?Ҡ ?Ӣ ?Ң ?Қ ?Һ ?Ҹ ?җ ?‘ ?’ ?“ ?” ?• ?–
4483 ?— ?ҳ ?ҷ ?ҡ ?ӣ ?ң ?қ ?һ ?ҹ ? ?Ў ?ў ?Ј ?Ө ?Ҙ ?Ұ ?§ ?Ё ?© ?Ә ?\« ?¬ ?ӯ
4484 ?® ?Ҝ ?° ?ұ ?І ?і ?ҙ ?ө ?¶ ?· ?ё ?№ ?ә ?\» ?ј ?Ҫ ?ҫ ?ҝ ?А ?Б ?В ?Г ?Д
4485 ?Е ?Ж ?З ?И ?Й ?К ?Л ?М ?Н ?О ?П ?Р ?С ?Т ?У ?Ф ?Х ?Ц ?Ч ?Ш ?Щ ?Ъ ?Ы
4486 ?Ь ?Э ?Ю ?Я ?а ?б ?в ?г ?д ?е ?ж ?з ?и ?й ?к ?л ?м ?н ?о ?п ?р ?с ?т
4487 ?у ?ф ?х ?ц ?ч ?ш ?щ ?ъ ?ы ?ь ?э ?ю ?я
])
4489 ;;;###autoload(autoload-coding-system 'iso-8859-11 '(require 'code-pages))
4490 (cp-make-coding-system
4492 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4493 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4494 ? ?ก ?ข ?ฃ ?ค ?ฅ ?ฆ ?ง ?จ ?ฉ ?ช ?ซ ?ฌ ?ญ ?ฎ ?ฏ
4495 ?ฐ ?ฑ ?ฒ ?ณ ?ด ?ต ?ถ ?ท ?ธ ?น ?บ ?ป ?ผ ?ฝ ?พ ?ฟ
4496 ?ภ ?ม ?ย ?ร ?ฤ ?ล ?ฦ ?ว ?ศ ?ษ ?ส ?ห ?ฬ ?อ ?ฮ ?ฯ
4497 ?ะ ?ั ?า ?ำ ?ิ ?ี ?ึ ?ื ?ุ ?ู ?ฺ nil nil nil nil ?฿
4498 ?เ ?แ ?โ ?ใ ?ไ ?ๅ ?ๆ ?็ ?่ ?้ ?๊ ?๋ ?์ ?ํ ?๎ ?๏
4499 ?๐ ?๑ ?๒ ?๓ ?๔ ?๕ ?๖ ?๗ ?๘ ?๙ ?๚ ?๛ nil nil nil nil
]
4500 "ISO-8859-11. This is `thai-tis620' with the addition of no-break-space.")
4503 (let ((w (intern (format "windows-125%d" i
)))
4504 (c (intern (format "cp125%d" i
))))
4505 ;; Define cp125* as aliases for all windows-125*, so on Windows
4506 ;; we can just concat "cp" to the ANSI codepage we get from the system
4507 ;; and not have to worry about whether it should be "cp" or "windows-".
4508 (when (coding-system-p w
)
4509 (define-coding-system-alias c w
)
4510 ;; Compatibility with codepage.el, though cp... are not the
4512 (if (not (assq c non-iso-charset-alist
))
4513 (let ((slot (assq w non-iso-charset-alist
)))
4515 (push (cons c
(cdr slot
)) non-iso-charset-alist
)))))))
4517 (provide 'code-pages
)
4519 ;;; arch-tag: 8b6e3c73-b271-4198-866d-ea6d0ceff1b2
4520 ;;; code-pages.el ends here