| 1 | -*-mode: text; coding: utf-8;-*- |
| 2 | |
| 3 | Copyright (C) 2002-2014 Free Software Foundation, Inc. |
| 4 | See the end of the file for license conditions. |
| 5 | |
| 6 | Problems, fixmes and other unicode-related issues |
| 7 | ------------------------------------------------------------- |
| 8 | |
| 9 | Notes by fx to record various things of variable importance. handa |
| 10 | needs to check them -- don't take too seriously, especially with |
| 11 | regard to completeness. |
| 12 | |
| 13 | * SINGLE_BYTE_CHAR_P returns true for Latin-1 characters, which has |
| 14 | undesirable effects. E.g.: |
| 15 | (multibyte-string-p (let ((s "x")) (aset s 0 ?£) s)) => nil |
| 16 | (multibyte-string-p (concat [?£])) => nil |
| 17 | (text-char-description ?£) => "M-#" |
| 18 | |
| 19 | These examples are all fixed by the change of 2002-10-14, but |
| 20 | there still exist questionable SINGLE_BYTE_CHAR_P in the |
| 21 | code (keymap.c and print.c). |
| 22 | |
| 23 | * Rationalize character syntax and its relationship to the Unicode |
| 24 | database. (Applies mainly to symbol an punctuation syntax.) |
| 25 | |
| 26 | * Fontset handling and customization needs work. We want to relate |
| 27 | fonts to scripts, probably based on the Unicode blocks. The |
| 28 | presence of small-repertoire 10646-encoded fonts in XFree 4 is a |
| 29 | pain, not currently worked round. |
| 30 | |
| 31 | With the change on 2002-07-26, multiple fonts can be |
| 32 | specified in a fontset for a specific range of characters. |
| 33 | Each range can also be specified by script. Before using |
| 34 | ISO10646 fonts, Emacs checks their repertories to avoid such |
| 35 | fonts that don't have a glyph for a specific character. |
| 36 | |
| 37 | fx has worked on fontset customization, but was stymied by |
| 38 | basic problems with the way the default face is dealt with |
| 39 | (and something else, I think). This needs revisiting. |
| 40 | |
| 41 | * Work is also needed on charset and coding system priorities. |
| 42 | |
| 43 | * The relevant bits of latin1-disp.el need porting (and probably |
| 44 | re-naming/updating). See also cyril-util.el. |
| 45 | |
| 46 | * Quail files need more work now the encoding is largely irrelevant. |
| 47 | |
| 48 | * What to do with the old coding categories stuff? |
| 49 | |
| 50 | * The preferred-coding-system property of charsets should probably be |
| 51 | junked unless it can be made more useful now. |
| 52 | |
| 53 | * find-multibyte-characters needs looking at. |
| 54 | |
| 55 | * Implement Korean cp949/UHC, BIG5-HKSCS and any other important missing |
| 56 | charsets. |
| 57 | |
| 58 | * Lazy-load tables for unify-charset somehow? |
| 59 | |
| 60 | Actually, Emacs clears out all charset maps and unify-map just |
| 61 | before dumping, and they are loaded again on demand by the |
| 62 | dumped emacs. But, those maps (char tables) generated while |
| 63 | temacs is running can't be removed from the dumped emacs. |
| 64 | |
| 65 | * iso-2022 charsets get unified on i/o. |
| 66 | |
| 67 | With the change on 2003-01-06, decoding routines put `charset' |
| 68 | property to decoded text, and iso-2022 encoder pay attention |
| 69 | to it. Thus, for instance, reading and writing by |
| 70 | iso-2022-7bit preserve the original designation sequences. |
| 71 | The property name `preferred-charset' may be better? |
| 72 | |
| 73 | We may have to utilize this property to decide a font. |
| 74 | |
| 75 | * Revisit locale processing: look at treating the language and |
| 76 | charset parts separately. (Language should affect things like |
| 77 | spelling and calendar, but that's not a Unicode issue.) |
| 78 | |
| 79 | * Handle Unicode combining characters usefully, e.g. diacritics, and |
| 80 | handle more scripts specifically (à la Devanagari). There are |
| 81 | issues with canonicalization. |
| 82 | |
| 83 | * We need tabular input methods, e.g. for maths symbols. (Not |
| 84 | specific to Unicode.) |
| 85 | |
| 86 | * Need multibyte text in menus, e.g. for the above. (Not specific to |
| 87 | Unicode -- see Emacs etc/TODO, but now mostly works with gtk.) |
| 88 | |
| 89 | * There's currently no support for Unicode normalization. |
| 90 | |
| 91 | * Populate char-width-table correctly for Unicode characters and |
| 92 | worry about what happens when double-width charsets covering |
| 93 | non-CJK characters are unified. |
| 94 | |
| 95 | * There are type errors lurking, e.g. in |
| 96 | Fcheck_coding_systems_region. Define ENABLE_CHECKING to find them. |
| 97 | |
| 98 | * Old auto-save files, and similar files, such as Gnus drafts, |
| 99 | containing non-ASCII characters probably won't be re-read correctly. |
| 100 | |
| 101 | |
| 102 | Source file encoding |
| 103 | -------------------- |
| 104 | |
| 105 | Most Emacs source files are encoded in UTF-8 (or in ASCII, which is a |
| 106 | subset), but there are a few exceptions, listed below. Perhaps |
| 107 | someday many of these files will be converted to UTF-8, for |
| 108 | convenience when using tools like 'grep -r', but this might need |
| 109 | nontrivial changes to the build process. |
| 110 | |
| 111 | * chinese-big5 |
| 112 | |
| 113 | These are verbatim copies of files taken from external sources. |
| 114 | They haven't been converted to UTF-8. |
| 115 | |
| 116 | leim/CXTERM-DIC/4Corner.tit |
| 117 | leim/CXTERM-DIC/ARRAY30.tit |
| 118 | leim/CXTERM-DIC/ECDICT.tit |
| 119 | leim/CXTERM-DIC/ETZY.tit |
| 120 | leim/CXTERM-DIC/PY-b5.tit |
| 121 | leim/CXTERM-DIC/Punct-b5.tit |
| 122 | leim/CXTERM-DIC/QJ-b5.tit |
| 123 | leim/CXTERM-DIC/ZOZY.tit |
| 124 | leim/MISC-DIC/CTLau-b5.html |
| 125 | leim/MISC-DIC/cangjie-table.b5 |
| 126 | |
| 127 | * chinese-iso-8bit |
| 128 | |
| 129 | These are verbatim copies of files taken from external sources. |
| 130 | They haven't been converted to UTF-8. |
| 131 | |
| 132 | leim/CXTERM-DIC/CCDOSPY.tit |
| 133 | leim/CXTERM-DIC/Punct.tit |
| 134 | leim/CXTERM-DIC/QJ.tit |
| 135 | leim/CXTERM-DIC/SW.tit |
| 136 | leim/CXTERM-DIC/TONEPY.tit |
| 137 | leim/MISC-DIC/pinyin.map |
| 138 | leim/MISC-DIC/CTLau.html |
| 139 | leim/MISC-DIC/ziranma.cin |
| 140 | |
| 141 | * cp850 |
| 142 | |
| 143 | This file contains non-ASCII characters in unibyte strings. When |
| 144 | editing a keyboard layout it's more convenient to see 'é' than |
| 145 | '\202', and the MS-DOS compiler requires the single byte if a |
| 146 | backslash escape is not being used. |
| 147 | |
| 148 | src/msdos.c |
| 149 | |
| 150 | * iso-2022-cn-ext |
| 151 | |
| 152 | This file is externally generated from leim/MISC-DIC/cangjie-table.b5 |
| 153 | by Big5->CNS converter. It hasn't been converted to UTF-8. |
| 154 | |
| 155 | leim/MISC-DIC/cangjie-table.cns |
| 156 | |
| 157 | * iso-latin-2 |
| 158 | |
| 159 | These files are processed by csplain, a program that requires |
| 160 | Latin-2 input. In 2012 the csplain maintainers started |
| 161 | recommending UTF-8, but these files haven't been converted yet. |
| 162 | |
| 163 | etc/refcards/cs-dired-ref.tex |
| 164 | etc/refcards/cs-refcard.tex |
| 165 | etc/refcards/cs-survival.tex |
| 166 | etc/refcards/sk-dired-ref.tex |
| 167 | etc/refcards/sk-refcard.tex |
| 168 | etc/refcards/sk-survival.tex |
| 169 | |
| 170 | * japanese-iso-8bit |
| 171 | |
| 172 | SKK-JISYO.L is a verbatim copy of a file taken from an external source. |
| 173 | It hasn't been converted to UTF-8. |
| 174 | |
| 175 | leim/SKK-DIC/SKK-JISYO.L |
| 176 | |
| 177 | * japanese-shift-jis |
| 178 | |
| 179 | This is a verbatim copy of a file taken from an external source. |
| 180 | It hasn't been converted to UTF-8. |
| 181 | |
| 182 | admin/charsets/mapfiles/cns2ucsdkw.txt |
| 183 | |
| 184 | * iso-2022-7bit |
| 185 | |
| 186 | This file switches between CJK charsets, which is not encoded in UTF-8. |
| 187 | |
| 188 | etc/HELLO |
| 189 | |
| 190 | Each of these files contains just one CJK charset, but Emacs |
| 191 | currently has no easy way to specify set-charset-priority on a |
| 192 | per-file basis, so converting any of these files to UTF-8 might |
| 193 | change the file's appearance when viewed by an Emacs that is |
| 194 | operating in some other language environment. |
| 195 | |
| 196 | etc/tutorials/TUTORIAL.ja |
| 197 | leim/quail/cyril-jis.el |
| 198 | leim/quail/hanja-jis.el |
| 199 | leim/quail/japanese.el |
| 200 | leim/quail/py-punct.el |
| 201 | leim/quail/pypunct-b5.el |
| 202 | lisp/international/ja-dic-cnv.el |
| 203 | lisp/international/ja-dic-utl.el |
| 204 | lisp/international/kinsoku.el |
| 205 | lisp/international/kkc.el |
| 206 | lisp/international/titdic-cnv.el |
| 207 | lisp/language/japan-util.el |
| 208 | lisp/language/japanese.el |
| 209 | lisp/term/x-win.el |
| 210 | |
| 211 | * utf-8-emacs |
| 212 | |
| 213 | These files contain characters that cannot be encoded in UTF-8. |
| 214 | |
| 215 | leim/quail/tibetan.el |
| 216 | leim/quail/ethiopic.el |
| 217 | lisp/international/titdic-cnv.el |
| 218 | lisp/language/tibetan.el |
| 219 | lisp/language/tibet-util.el |
| 220 | lisp/language/ind-util.el |
| 221 | |
| 222 | \f |
| 223 | This file is part of GNU Emacs. |
| 224 | |
| 225 | GNU Emacs is free software: you can redistribute it and/or modify |
| 226 | it under the terms of the GNU General Public License as published by |
| 227 | the Free Software Foundation, either version 3 of the License, or |
| 228 | (at your option) any later version. |
| 229 | |
| 230 | GNU Emacs is distributed in the hope that it will be useful, |
| 231 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 232 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 233 | GNU General Public License for more details. |
| 234 | |
| 235 | You should have received a copy of the GNU General Public License |
| 236 | along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. |