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 ;; Macro to allow ccl compilation at byte-compile time, avoiding
145 (defmacro cp-make-coding-system
(name v
&optional doc-string mnemonic
)
146 "Make coding system NAME for and 8-bit, extended-ASCII character set.
147 V is a 128-long vector of characters to translate the upper half of
148 the charactert set. DOC-STRING and MNEMONIC are used as the
149 corresponding args of `make-coding-system'. If MNEMONIC isn't given,
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
)))
199 (let (l) ; code range
200 (dolist (elt (reverse codes
))
204 non-iso-charset-alist
))))
207 ;; These tables were mostly derived by running somthing like
208 ;; `recode -f cpxxx/..utf-8' on a binary file filled by
209 ;; `(dotimes (i 128) (insert ?? ?\\ (+ 128 i) ?\n))' and then
210 ;; exchanging the ?\� entries for nil. iconv was used instead in some
213 ;; Fixme: Do better for mode-line mnemonics?
215 (cp-make-coding-system
346 (cp-make-coding-system
476 (coding-system-put 'cp737
'mime-charset nil
) ; not in IANA list
478 (cp-make-coding-system
609 (cp-make-coding-system
740 (cp-make-coding-system
871 (cp-make-coding-system
1002 (cp-make-coding-system
1133 (cp-make-coding-system
1264 (cp-make-coding-system
1395 (cp-make-coding-system
1526 (cp-make-coding-system
1657 (cp-make-coding-system
1788 (cp-make-coding-system
1919 (cp-make-coding-system
2050 (cp-make-coding-system
2183 (cp-make-coding-system
2314 (cp-make-coding-system
2445 (cp-make-coding-system
2577 (cp-make-coding-system
2709 (cp-make-coding-system
2840 (cp-make-coding-system
2972 (cp-make-coding-system
3104 (cp-make-coding-system
3236 (cp-make-coding-system
3368 (cp-make-coding-system
3499 (cp-make-coding-system
3630 (cp-make-coding-system
3760 "NeXTstep encoding." ?N
)
3762 (cp-make-coding-system
3763 koi8-t
; used by glibc for tg_TJ
3892 "Unicode-based KOI8-T encoding for Cyrillic")
3893 (coding-system-put 'koi8-t
'mime-charset nil
) ; not in the IANA list
3894 (define-coding-system-alias 'cyrillic-koi8-t
'koi8-t
)
3896 ;; Online final ISO draft:
3898 ;; http://www.evertype.com/standards/iso8859/fdis8859-16-en.pdf
3900 ;; Equivalent National Standard:
3901 ;; Romanian Standard SR 14111:1998, Romanian Standards Institution
3906 ;; "This set of coded graphic characters is intended for use in data and
3907 ;; text processing applications and also for information interchange. The
3908 ;; set contains graphic characters used for general purpose applications in
3909 ;; typical office environments in at least the following languages:
3910 ;; Albanian, Croatian, English, Finnish, French, German, Hungarian, Irish
3911 ;; Gaelic (new orthography), Italian, Latin, Polish, Romanian, and
3912 ;; Slovenian. This set of coded graphic characters may be regarded as a
3913 ;; version of an 8-bit code according to ISO/IEC 2022 or ISO/IEC 4873 at
3914 ;; level 1." [ISO 8859-16:2001(E), p. 1]
3916 ;; This charset is suitable for use in MIME text body parts.
3918 ;; ISO 8859-16 was primarily designed for single-byte encoding the Romanian
3919 ;; language. The UTF-8 charset is the preferred and in today's MIME software
3920 ;; more widely implemented encoding suitable for Romanian.
3921 (cp-make-coding-system
3922 iso-latin-10
; consistent with, e.g. Latin-1
3923 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
3924 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4021 "Unicode-based encoding for Latin-10 (MIME: ISO-8859-16)"
4023 (coding-system-put 'iso-latin-10
'mime-charset
'iso-8859-16
)
4024 (define-coding-system-alias 'iso-8859-16
'iso-latin-10
)
4025 (define-coding-system-alias 'latin-10
'iso-latin-10
)
4027 ;; Unicode-based alternative which has the possible advantage of
4028 ;; having its relative sparseness specified.
4029 (cp-make-coding-system
4030 ;; The base system uses arabic-iso-8bit, but that's not a MIME charset.
4032 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4033 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4085 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4086 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4087 nil nil nil nil nil nil nil nil nil nil nil
]
4088 "Unicode-based Arabic ISO/IEC 8859-6 (MIME: ISO-8859-6)"
4090 (define-coding-system-alias 'arabic-iso-8bit
'iso-8859-6
)
4092 (cp-make-coding-system
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
4192 "Unicode-based encoding for Latin-6 (MIME: ISO-8859-10)")
4193 (coding-system-put 'iso-latin-6
'mime-charset
'iso-8859-10
)
4194 (define-coding-system-alias 'iso-8859-10
'iso-latin-6
)
4195 (define-coding-system-alias 'latin-6
'iso-latin-6
)
4197 ;; used by lt_LT, lv_LV, mi_NZ
4198 (cp-make-coding-system
4200 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4201 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4299 "Unicode-based encoding for Latin-7 (MIME: ISO-8859-13)"
4300 ?l
) ;; Lithuanian/Latvian
4301 (coding-system-put 'iso-latin-7
'mime-charset
'iso-8859-13
)
4302 (define-coding-system-alias 'iso-8859-13
'iso-latin-7
)
4303 (define-coding-system-alias 'latin-7
'iso-latin-7
)
4305 ;; Fixme: check on the C1 characters which libiconv includes. They
4306 ;; are reproduced below, but are probably wrong. I can't find an
4307 ;; official definition of georgian-ps.
4308 (cp-make-coding-system
4309 georgian-ps
; used by glibc for ka_GE
4439 (coding-system-put 'georgian-ps
'mime-charset nil
) ; not in IANA list
4441 ;; From http://www.microsoft.com/globaldev/reference/oem/720.htm
4442 (cp-make-coding-system
4572 (coding-system-put 'cp720
'mime-charset nil
) ; not in IANA list
4574 ;; http://oss.software.ibm.com/cvs/icu/charset/data/ucm/ibm-1125_P100-2000.ucm
4575 (cp-make-coding-system
4705 (define-coding-system-alias 'ruscii
'cp1125
)
4706 ;; Original name for cp1125, says Serhii Hlodin <hlodin@lutsk.bank.gov.ua>
4707 (define-coding-system-alias 'cp866u
'cp1125
)
4708 (coding-system-put 'cp1125
'mime-charset nil
)
4710 ;; Suggested by Anton Zinoviev <anton@lml.bas.bg>: Bulgarian DOS
4711 ;; codepage. Table at
4712 ;; <URL:http://czyborra.com/charsets/bulgarian-mik.txt.gz>.
4713 (cp-make-coding-system
4715 [?А ?Б ?В ?Г ?Д ?Е ?Ж ?З ?И ?Й ?К ?Л ?М ?Н ?О ?П ?Р ?С ?Т ?У ?Ф ?Х ?Ц
4716 ?Ч ?Ш ?Щ ?Ъ ?Ы ?Ь ?Э ?Ю ?Я ?а ?б ?в ?г ?д ?е ?ж ?з ?и ?й ?к ?л ?м ?н
4717 ?о ?п ?р ?с ?т ?у ?ф ?х ?ц ?ч ?ш ?щ ?ъ ?ы ?ь ?э ?ю ?я ?└ ?┴ ?┬ ?├ ?─
4718 ?┼ ?╣ ?║ ?╚ ?╔ ?╩ ?╦ ?╠ ?═ ?╬ ?┐ ?░ ?▒ ?▓ ?│ ?┤ ?№ ?§ ?╗ ?╝ ?┘ ?┌ ?█
4719 ?▄ ?▌ ?▐ ?▀ ?α ?β ?Γ ?π ?Σ ?σ ?μ ?τ ?Φ ?Θ ?Ω ?δ ?∞ ?∅ ?∈ ?∩ ?≡ ?± ?≥
4720 ?≤ ?⌠ ?⌡ ?÷ ?≈ ?° ?∙ ?· ?√ ?ⁿ ?² ?■ ?
])
4721 (coding-system-put 'mik
'mime-charset nil
)
4723 ;; Suggested by Anton Zinoviev <anton@lml.bas.bg>: similar to CP1251
4724 ;; and used for some non-Slavic Cyrillic languages. Table found at
4725 ;; <URL:ftp://ftp.logic.ru/pub/logic/linux/cyr-asian/PT154>. See also
4726 ;; <URL:http://lists.w3.org/Archives/Public/ietf-charsets/2002AprJun/0092.html,
4727 ;; which suggests it's used in an Asian Cyrillic context.
4728 (cp-make-coding-system
4730 [?Җ ?Ғ ?Ӯ ?ғ ?„ ?… ?Ҷ ?Ү ?Ҳ ?ү ?Ҡ ?Ӣ ?Ң ?Қ ?Һ ?Ҹ ?җ ?‘ ?’ ?“ ?” ?• ?–
4731 ?— ?ҳ ?ҷ ?ҡ ?ӣ ?ң ?қ ?һ ?ҹ ? ?Ў ?ў ?Ј ?Ө ?Ҙ ?Ұ ?§ ?Ё ?© ?Ә ?\« ?¬ ?ӯ
4732 ?® ?Ҝ ?° ?ұ ?І ?і ?ҙ ?ө ?¶ ?· ?ё ?№ ?ә ?» ?ј ?Ҫ ?ҫ ?ҝ ?А ?Б ?В ?Г ?Д
4733 ?Е ?Ж ?З ?И ?Й ?К ?Л ?М ?Н ?О ?П ?Р ?С ?Т ?У ?Ф ?Х ?Ц ?Ч ?Ш ?Щ ?Ъ ?Ы
4734 ?Ь ?Э ?Ю ?Я ?а ?б ?в ?г ?д ?е ?ж ?з ?и ?й ?к ?л ?м ?н ?о ?п ?р ?с ?т
4735 ?у ?ф ?х ?ц ?ч ?ш ?щ ?ъ ?ы ?ь ?э ?ю ?я
])
4737 (cp-make-coding-system
4739 [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4740 nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
4741 ? ?ก ?ข ?ฃ ?ค ?ฅ ?ฆ ?ง ?จ ?ฉ ?ช ?ซ ?ฌ ?ญ ?ฎ ?ฏ
4742 ?ฐ ?ฑ ?ฒ ?ณ ?ด ?ต ?ถ ?ท ?ธ ?น ?บ ?ป ?ผ ?ฝ ?พ ?ฟ
4743 ?ภ ?ม ?ย ?ร ?ฤ ?ล ?ฦ ?ว ?ศ ?ษ ?ส ?ห ?ฬ ?อ ?ฮ ?ฯ
4744 ?ะ ?ั ?า ?ำ ?ิ ?ี ?ึ ?ื ?ุ ?ู ?ฺ nil nil nil nil ?฿
4745 ?เ ?แ ?โ ?ใ ?ไ ?ๅ ?ๆ ?็ ?่ ?้ ?๊ ?๋ ?์ ?ํ ?๎ ?๏
4746 ?๐ ?๑ ?๒ ?๓ ?๔ ?๕ ?๖ ?๗ ?๘ ?๙ ?๚ ?๛ nil nil nil nil
]
4747 "ISO-8859-11. This is `thai-tis620' with the addition of no-break-space.")
4750 (let ((w (intern (format "windows-125%d" i
)))
4751 (c (intern (format "cp125%d" i
))))
4752 (define-coding-system-alias c w
)
4753 ;; Compatibility with codepage.el, though cp... are not the
4755 (push (assoc w non-iso-charset-alist
) non-iso-charset-alist
)))
4757 ;; Use Unicode font under Windows. Jason Rumney fecit.
4758 (if (and (fboundp 'w32-add-charset-info
)
4759 (not (boundp 'w32-unicode-charset-defined
)))
4760 (w32-add-charset-info "iso10646-1" 'w32-charset-ansi t
))
4762 (provide 'code-pages
)
4764 ;;; code-pages.el ends here