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 "This is an obsolete function.
99 It exists just for backward compatibility, and it does nothing.")
100 (make-obsolete 'cp-fix-safe-chars
101 "Unnecessary function. Calling it has no effect."
104 ;; Fix things that have been, or might be, done by codepage.el.
105 (eval-after-load "codepage"
108 ;; Semi-dummy version for the stuff in codepage.el which we don't
109 ;; define here. (Used by mule-diag.)
110 (defun cp-supported-codepages ()
111 "Return an alist of supported codepages.
113 Each association in the alist has the form (NNN . CHARSET), where NNN is the
114 codepage number, and CHARSET is the MULE charset which is the closest match
115 for the character set supported by that codepage.
117 A codepage NNN is supported if a variable called `cpNNN-decode-table' exists,
118 is a vector, and has a charset property."
119 '(("774" . latin-iso8859-4
) ("770" . latin-iso8859-4
)
120 ("773" . latin-iso8859-4
)))
122 ;; A version which doesn't override the coding systems set up by this
123 ;; file. It could still be used for the few missing ones from
125 (defun codepage-setup (codepage)
126 "Create a coding system cpCODEPAGE to support the IBM codepage CODEPAGE.
128 These coding systems are meant for encoding and decoding 8-bit non-ASCII
129 characters used by the IBM codepages, typically in conjunction with files
130 read/written by MS-DOS software, or for display on the MS-DOS terminal."
132 (let ((completion-ignore-case t
)
133 (candidates (cp-supported-codepages)))
134 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
135 nil t nil nil
"437"))))
136 (let ((cp (format "cp%s" codepage
)))
137 (unless (coding-system-p (intern cp
))
138 (cp-make-coding-systems-for-codepage
139 cp
(cp-charset-for-codepage cp
) (cp-offset-for-codepage cp
))))))
142 ;; For `non-iso-charset-alist'. Do this after redefining
143 ;; `cp-supported-codepages', which is called through loading
147 ;; Macro to allow ccl compilation at byte-compile time, avoiding
150 (defmacro cp-make-coding-system
(name v
&optional doc-string mnemonic
)
151 "Make coding system NAME for and 8-bit, extended-ASCII character set.
152 V is a 128-long vector of characters to translate the upper half of
153 the charactert set. DOC-STRING and MNEMONIC are used as the
154 corresponding args of `make-coding-system'. If MNEMONIC isn't given,
156 (let* ((encoder (intern (format "encode-%s" name
)))
157 (decoder (intern (format "decode-%s" name
)))
163 (if (r1 < 128) ;; ASCII
164 (r0 = ,(charset-id 'ascii
))
166 (r0 = ,(charset-id 'eight-bit-control
))
167 (r0 = ,(charset-id 'eight-bit-graphic
))))
168 (translate-character ,decoder r0 r1
)
169 ;; Allow fragmentation on decoding -- relevant for
170 ;; Cyrillic, Greek and, possibly Arabic and Hebrew.
171 (translate-character utf-translation-table-for-decode r0 r1
)
172 (write-multibyte-character r0 r1
)
178 (read-multibyte-character r0 r1
)
179 (translate-character ,encoder r0 r1
)
180 (if (r0 != ,(charset-id 'ascii
))
181 (if (r0 != ,(charset-id 'eight-bit-graphic
))
182 (if (r0 != ,(charset-id 'eight-bit-control
))
184 (write-repeat r1
)))))))
185 `(let ((translation-table (cp-make-translation-table ,v
))
186 (codes (cp-valid-codes ,v
)))
187 (define-translation-table ',decoder translation-table
)
188 (define-translation-table ',encoder
189 (char-table-extra-slot translation-table
0))
191 ',name
4 ,(or mnemonic ?
*)
192 (or ,doc-string
(format "%s encoding" ',name
))
193 (cons ,ccl-decoder
,ccl-encoder
)
194 (list (cons 'safe-chars
(get ',encoder
'translation-table
))
195 (cons 'valid-codes codes
)
196 (cons 'mime-charset
',name
)
197 ;; For Quail translation. Fixme: this should really be
198 ;; a separate table that only translates the coding
199 ;; system's safe-chars.
200 (cons 'translation-table-for-input
,ucs-mule-to-mule-unicode
)))
204 (let (l) ; code range
205 (dolist (elt (reverse codes
))
209 non-iso-charset-alist
))))
212 ;; These tables were mostly derived by running somthing like
213 ;; `recode -f cpxxx/..utf-8' on a binary file filled by
214 ;; `(dotimes (i 128) (insert ?? ?\\ (+ 128 i) ?\n))' and then
215 ;; exchanging the ?\� entries for nil. iconv was used instead in some
218 ;; Fixme: Do better for mode-line mnemonics?
220 (cp-make-coding-system
351 (cp-make-coding-system
481 (coding-system-put 'cp737
'mime-charset nil
) ; not in IANA list
483 (cp-make-coding-system
614 (cp-make-coding-system
745 (cp-make-coding-system
876 (cp-make-coding-system
1007 (cp-make-coding-system
1138 (cp-make-coding-system
1269 (cp-make-coding-system
1400 (cp-make-coding-system
1531 (cp-make-coding-system
1662 (cp-make-coding-system
1793 (cp-make-coding-system
1924 (cp-make-coding-system
2055 (cp-make-coding-system
2188 (cp-make-coding-system
2319 (cp-make-coding-system
2450 (cp-make-coding-system
2582 (cp-make-coding-system
2714 (cp-make-coding-system
2845 (cp-make-coding-system
2977 (cp-make-coding-system
3109 (cp-make-coding-system
3241 (cp-make-coding-system
3373 (cp-make-coding-system
3504 (cp-make-coding-system
3635 (cp-make-coding-system
3765 "NeXTstep encoding." ?N
)
3767 (cp-make-coding-system
3768 koi8-t
; used by glibc for tg_TJ
3897 "Unicode-based KOI8-T encoding for Cyrillic")
3898 (coding-system-put 'koi8-t
'mime-charset nil
) ; not in the IANA list
3900 ;; Online final ISO draft:
3902 ;; http://www.evertype.com/standards/iso8859/fdis8859-16-en.pdf
3904 ;; Equivalent National Standard:
3905 ;; Romanian Standard SR 14111:1998, Romanian Standards Institution
3910 ;; "This set of coded graphic characters is intended for use in data and
3911 ;; text processing applications and also for information interchange. The
3912 ;; set contains graphic characters used for general purpose applications in
3913 ;; typical office environments in at least the following languages:
3914 ;; Albanian, Croatian, English, Finnish, French, German, Hungarian, Irish
3915 ;; Gaelic (new orthography), Italian, Latin, Polish, Romanian, and
3916 ;; Slovenian. This set of coded graphic characters may be regarded as a
3917 ;; version of an 8-bit code according to ISO/IEC 2022 or ISO/IEC 4873 at
3918 ;; level 1." [ISO 8859-16:2001(E), p. 1]
3920 ;; This charset is suitable for use in MIME text body parts.
3922 ;; ISO 8859-16 was primarily designed for single-byte encoding the Romanian
3923 ;; language. The UTF-8 charset is the preferred and in today's MIME software
3924 ;; more widely implemented encoding suitable for Romanian.
3925 (cp-make-coding-system
3926 iso-latin-10
; consistent with, e.g. Latin-1
3927 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3928 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4025 "Unicode-based encoding for Latin-10 (MIME: ISO-8859-16)"
4027 (coding-system-put 'iso-latin-10
'mime-charset
'iso-8859-16
)
4028 (define-coding-system-alias 'iso-8859-16
'iso-latin-10
)
4029 (define-coding-system-alias 'latin-10
'iso-latin-10
)
4031 ;; Unicode-based alternative which has the possible advantage of
4032 ;; having its relative sparseness specified.
4033 (cp-make-coding-system
4034 ;; The base system uses arabic-iso-8bit, but that's not a MIME charset.
4036 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4037 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4089 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4090 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4091 nil nil nil nil nil nil nil nil nil nil nil
]
4092 "Unicode-based Arabic ISO/IEC 8859-6 (MIME: ISO-8859-6)"
4094 (define-coding-system-alias 'arabic-iso-8bit
'iso-8859-6
)
4096 (cp-make-coding-system
4098 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4099 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4196 "Unicode-based encoding for Latin-6 (MIME: ISO-8859-10)")
4197 (coding-system-put 'iso-latin-6
'mime-charset
'iso-8859-10
)
4198 (define-coding-system-alias 'iso-8859-10
'iso-latin-6
)
4199 (define-coding-system-alias 'latin-6
'iso-latin-6
)
4201 ;; used by lt_LT, lv_LV, mi_NZ
4202 (cp-make-coding-system
4204 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4205 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4303 "Unicode-based encoding for Latin-7 (MIME: ISO-8859-13)"
4304 ?l
) ;; Lithuanian/Latvian
4305 (coding-system-put 'iso-latin-7
'mime-charset
'iso-8859-13
)
4306 (define-coding-system-alias 'iso-8859-13
'iso-latin-7
)
4307 (define-coding-system-alias 'latin-7
'iso-latin-7
)
4309 ;; Fixme: check on the C1 characters which libiconv includes. They
4310 ;; are reproduced below, but are probably wrong. I can't find an
4311 ;; official definition of georgian-ps.
4312 (cp-make-coding-system
4313 georgian-ps
; used by glibc for ka_GE
4443 (coding-system-put 'georgian-ps
'mime-charset nil
) ; not in IANA list
4445 ;; From http://www.microsoft.com/globaldev/reference/oem/720.htm
4446 (cp-make-coding-system
4576 (coding-system-put 'cp720
'mime-charset nil
) ; not in IANA list
4578 ;; http://oss.software.ibm.com/cvs/icu/charset/data/ucm/ibm-1125_P100-2000.ucm
4579 (cp-make-coding-system
4709 (define-coding-system-alias 'ruscii
'cp1125
)
4710 ;; Original name for cp1125, says Serhii Hlodin <hlodin@lutsk.bank.gov.ua>
4711 (define-coding-system-alias 'cp866u
'cp1125
)
4712 (coding-system-put 'cp1125
'mime-charset nil
)
4714 ;; Suggested by Anton Zinoviev <anton@lml.bas.bg>: Bulgarian DOS
4715 ;; codepage. Table at
4716 ;; <URL:http://czyborra.com/charsets/bulgarian-mik.txt.gz>.
4717 (cp-make-coding-system
4719 [?А ?Б ?В ?Г ?Д ?Е ?Ж ?З ?И ?Й ?К ?Л ?М ?Н ?О ?П ?Р ?С ?Т ?У ?Ф ?Х ?Ц
4720 ?Ч ?Ш ?Щ ?Ъ ?Ы ?Ь ?Э ?Ю ?Я ?а ?б ?в ?г ?д ?е ?ж ?з ?и ?й ?к ?л ?м ?н
4721 ?о ?п ?р ?с ?т ?у ?ф ?х ?ц ?ч ?ш ?щ ?ъ ?ы ?ь ?э ?ю ?я ?└ ?┴ ?┬ ?├ ?─
4722 ?┼ ?╣ ?║ ?╚ ?╔ ?╩ ?╦ ?╠ ?═ ?╬ ?┐ ?░ ?▒ ?▓ ?│ ?┤ ?№ ?§ ?╗ ?╝ ?┘ ?┌ ?█
4723 ?▄ ?▌ ?▐ ?▀ ?α ?β ?Γ ?π ?Σ ?σ ?μ ?τ ?Φ ?Θ ?Ω ?δ ?∞ ?∅ ?∈ ?∩ ?≡ ?± ?≥
4724 ?≤ ?⌠ ?⌡ ?÷ ?≈ ?° ?∙ ?· ?√ ?ⁿ ?² ?■ ?
])
4725 (coding-system-put 'mik
'mime-charset nil
)
4727 ;; Suggested by Anton Zinoviev <anton@lml.bas.bg>: similar to CP1251
4728 ;; and used for some non-Slavic Cyrillic languages. Table found at
4729 ;; <URL:ftp://ftp.logic.ru/pub/logic/linux/cyr-asian/PT154>. See also
4730 ;; <URL:http://lists.w3.org/Archives/Public/ietf-charsets/2002AprJun/0092.html,
4731 ;; which suggests it's used in an Asian Cyrillic context.
4732 (cp-make-coding-system
4734 [?Җ ?Ғ ?Ӯ ?ғ ?„ ?… ?Ҷ ?Ү ?Ҳ ?ү ?Ҡ ?Ӣ ?Ң ?Қ ?Һ ?Ҹ ?җ ?‘ ?’ ?“ ?” ?• ?–
4735 ?— ?ҳ ?ҷ ?ҡ ?ӣ ?ң ?қ ?һ ?ҹ ? ?Ў ?ў ?Ј ?Ө ?Ҙ ?Ұ ?§ ?Ё ?© ?Ә ?\« ?¬ ?ӯ
4736 ?® ?Ҝ ?° ?ұ ?І ?і ?ҙ ?ө ?¶ ?· ?ё ?№ ?ә ?» ?ј ?Ҫ ?ҫ ?ҝ ?А ?Б ?В ?Г ?Д
4737 ?Е ?Ж ?З ?И ?Й ?К ?Л ?М ?Н ?О ?П ?Р ?С ?Т ?У ?Ф ?Х ?Ц ?Ч ?Ш ?Щ ?Ъ ?Ы
4738 ?Ь ?Э ?Ю ?Я ?а ?б ?в ?г ?д ?е ?ж ?з ?и ?й ?к ?л ?м ?н ?о ?п ?р ?с ?т
4739 ?у ?ф ?х ?ц ?ч ?ш ?щ ?ъ ?ы ?ь ?э ?ю ?я
])
4741 (cp-make-coding-system
4743 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4744 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4745 ? ?ก ?ข ?ฃ ?ค ?ฅ ?ฆ ?ง ?จ ?ฉ ?ช ?ซ ?ฌ ?ญ ?ฎ ?ฏ
4746 ?ฐ ?ฑ ?ฒ ?ณ ?ด ?ต ?ถ ?ท ?ธ ?น ?บ ?ป ?ผ ?ฝ ?พ ?ฟ
4747 ?ภ ?ม ?ย ?ร ?ฤ ?ล ?ฦ ?ว ?ศ ?ษ ?ส ?ห ?ฬ ?อ ?ฮ ?ฯ
4748 ?ะ ?ั ?า ?ำ ?ิ ?ี ?ึ ?ื ?ุ ?ู ?ฺ nil nil nil nil ?฿
4749 ?เ ?แ ?โ ?ใ ?ไ ?ๅ ?ๆ ?็ ?่ ?้ ?๊ ?๋ ?์ ?ํ ?๎ ?๏
4750 ?๐ ?๑ ?๒ ?๓ ?๔ ?๕ ?๖ ?๗ ?๘ ?๙ ?๚ ?๛ nil nil nil nil
]
4751 "ISO-8859-11. This is `thai-tis620' with the addition of no-break-space.")
4754 (let ((w (intern (format "windows-125%d" i
)))
4755 (c (intern (format "cp125%d" i
))))
4756 (define-coding-system-alias c w
)
4757 ;; Compatibility with codepage.el, though cp... are not the
4759 (push (assoc w non-iso-charset-alist
) non-iso-charset-alist
)))
4761 ;; Use Unicode font under Windows. Jason Rumney fecit.
4762 (if (and (fboundp 'w32-add-charset-info
)
4763 (not (boundp 'w32-unicode-charset-defined
)))
4764 (w32-add-charset-info "iso10646-1" 'w32-charset-ansi t
))
4766 (provide 'code-pages
)
4768 ;;; code-pages.el ends here