Commit | Line | Data |
---|---|---|
18acc655 TTN |
1 | ;;; latexenc.el --- guess correct coding system in LaTeX files |
2 | ||
3 | ;; Copyright (C) 2005 Free Software Foundation, Inc. | |
4 | ||
5 | ;; Author: Arne J\e,Ax\e(Brgensen <arne@arnested.dk> | |
6 | ;; Keywords: mule, coding system, latex | |
7 | ||
8 | ;; This file is part of GNU Emacs. | |
9 | ||
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) | |
13 | ;; any later version. | |
14 | ||
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. | |
19 | ||
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. | |
24 | ||
25 | ;;; Commentary: | |
26 | ||
27 | ;; This code tries to guess the correct coding system of a LaTeX file. | |
28 | ||
29 | ;; First it searches for a \inputencoding{...} or | |
30 | ;; \usepackage[...]{inputenc} line in the file and looks up the ... in | |
31 | ;; `latex-inputenc-coding-alist' to find the corresponding coding | |
32 | ;; system. | |
33 | ||
34 | ;; If this fails it will search for AUCTeX's TeX-master or tex-mode's | |
35 | ;; tex-main-file variable in the local variables section and visit | |
36 | ;; that file to get the coding system from the master file. This check | |
37 | ;; can be disabled by setting `latexenc-dont-use-TeX-master-flag' to | |
38 | ;; t. | |
39 | ||
40 | ;; If we have still not found a coding system we will try to use the | |
41 | ;; standard tex-mode's `tex-guess-main-file' and get the coding system | |
42 | ;; from the main file. This check can be disabled by setting | |
43 | ;; `latexenc-dont-use-tex-guess-main-file-flag' to t. | |
44 | ||
45 | ;; The functionality is enabled by adding the function | |
46 | ;; `latexenc-find-file-coding-system' to `file-coding-system-alist' | |
47 | ;; like this | |
48 | ||
49 | ;; (add-to-list 'file-coding-system-alist | |
50 | ;; '("\\.tex\\|\\.ltx\\|\\.dtx\\|\\.drv\\'" . latexenc-find-file-coding-system)) | |
51 | ||
52 | ;;; Code: | |
53 | ||
54 | ;;;###autoload | |
55 | (defcustom latex-inputenc-coding-alist | |
56 | '(("ansinew" . windows-1252) ; MS Windows ANSI encoding, extension of Latin-1 | |
57 | ("applemac" . mac-roman) | |
58 | ("ascii" . us-ascii) | |
59 | ("cp1250" . windows-1250) ; MS Windows encoding, codepage 1250 | |
60 | ("cp1252" . windows-1252) ; synonym of ansinew | |
61 | ("cp1257" . cp1257) | |
62 | ("cp437de" . cp437) ; IBM code page 437 (German version): 225 is \ss | |
63 | ("cp437" . cp437) ; IBM code page 437: 225 is \beta | |
64 | ("cp850" . cp850) ; IBM code page 850 | |
65 | ("cp852" . cp852) ; IBM code page 852 | |
66 | ;; ("cp858" . undecided) ; IBM code page 850 but with a euro symbol | |
67 | ("cp865" . cp865) ; IBM code page 865 | |
68 | ;; The DECMultinational charaterset used by the OpenVMS system | |
69 | ;; ("decmulti" . undecided) | |
70 | ("latin1" . iso-8859-1) | |
71 | ("latin2" . iso-8859-2) | |
72 | ("latin3" . iso-8859-3) | |
73 | ("latin4" . iso-8859-4) | |
74 | ("latin5" . iso-8859-5) | |
75 | ("latin9" . iso-8859-15) | |
76 | ;; ("latin10" . undecided) | |
77 | ;; ("macce" . undecided) ; Apple Central European | |
78 | ("next" . next) ; The Next encoding | |
79 | ("utf8" . utf-8) | |
80 | ("utf8x" . utf-8)) ; used by the Unicode LaTeX package | |
81 | "Mapping from encoding names used by LaTeX's \"inputenc.sty\" to Emacs coding systems. | |
82 | Used by the function `latexenc-find-file-coding-system'." | |
83 | :group 'files | |
84 | :group 'mule | |
85 | :type '(alist :key-type (string :tag "LaTeX input encoding") | |
86 | :value-type (coding-system :tag "Coding system"))) | |
87 | ||
88 | ;;;###autoload | |
89 | (defun latexenc-inputenc-to-coding-system (inputenc) | |
90 | "Return the corresponding coding-system for the specified input encoding. | |
91 | Return nil if no matching coding system can be found." | |
92 | (cdr (assoc inputenc latex-inputenc-coding-alist))) | |
93 | ||
94 | ;;;###autoload | |
95 | (defun latexenc-coding-system-to-inputenc (cs) | |
96 | "Return the corresponding input encoding for the specified coding system. | |
97 | Return nil if no matching input encoding can be found." | |
98 | (let (result) | |
99 | (catch 'result | |
100 | (dolist (elem latex-inputenc-coding-alist result) | |
101 | (let ((elem-cs (cdr elem))) | |
102 | (when (and (coding-system-p elem-cs) | |
103 | (coding-system-p cs) | |
104 | (eq (coding-system-base cs) (coding-system-base elem-cs))) | |
105 | (setq result (car elem)) | |
106 | (throw 'result result))))))) | |
107 | ||
108 | (defvar latexenc-dont-use-TeX-master-flag nil | |
109 | "Non-nil means don't follow TeX-master to find the coding system.") | |
110 | ||
111 | (defvar latexenc-dont-use-tex-guess-main-file-flag nil | |
112 | "Non-nil means don't use tex-guessmain-file to find the coding system.") | |
113 | ||
114 | ;;;###autoload | |
115 | (defun latexenc-find-file-coding-system (arg-list) | |
116 | "Determine the coding system of a LaTeX file if it uses \"inputenc.sty\". | |
117 | The mapping from LaTeX's \"inputenc.sty\" encoding names to Emacs | |
118 | coding system names is determined from `latex-inputenc-coding-alist'." | |
119 | (if (eq (car arg-list) 'insert-file-contents) | |
120 | (save-excursion | |
121 | ;; try to find the coding system in this file | |
122 | (goto-char (point-min)) | |
4b0fcb5d LK |
123 | (if (catch 'cs |
124 | (let ((case-fold-search nil)) | |
125 | (while (search-forward "inputenc" nil t) | |
126 | (goto-char (match-beginning 0)) | |
127 | (beginning-of-line) | |
128 | (if (or (looking-at "[^%\n]*\\\\usepackage\\[\\([^]]*\\)\\]{\\([^}]*,\\)?inputenc\\(,[^}]*\\)?}") | |
129 | (looking-at "[^%\n]*\\\\inputencoding{\\([^}]*\\)}")) | |
130 | (throw 'cs t) | |
131 | (goto-char (match-end 0)))))) | |
132 | (let* ((match (match-string 1)) | |
133 | (sym (intern match))) | |
134 | (when (latexenc-inputenc-to-coding-system match) | |
135 | (setq sym (latexenc-inputenc-to-coding-system match))) | |
136 | (when (coding-system-p sym) | |
137 | sym | |
138 | (if (and (require 'code-pages nil t) (coding-system-p sym)) | |
139 | sym | |
140 | 'undecided))) | |
18acc655 TTN |
141 | ;; else try to find it in the master/main file |
142 | (let (latexenc-main-file) | |
143 | ;; is there a TeX-master or tex-main-file in the local variable section | |
144 | (unless latexenc-dont-use-TeX-master-flag | |
145 | (goto-char (point-max)) | |
4b0fcb5d LK |
146 | (search-backward "\n\^L" (max (- (point-max) 3000) (point-min)) 'move) |
147 | (search-forward "Local Variables:" nil t) | |
148 | (when (re-search-forward "^%+ *\\(TeX-master\\|tex-main-file\\): *\"\\(.+\\)\"" nil t) | |
18acc655 TTN |
149 | (let ((file (concat (file-name-directory (nth 1 arg-list)) (match-string 2)))) |
150 | (if (file-exists-p file) | |
151 | (setq latexenc-main-file file) | |
152 | (if (boundp 'TeX-default-extension) | |
153 | (when (file-exists-p (concat file "." TeX-default-extension)) | |
154 | (setq latexenc-main-file (concat file "." TeX-default-extension))) | |
155 | (dolist (ext '("drv" "dtx" "ltx" "tex")) | |
156 | (if (file-exists-p (concat file "." ext)) | |
157 | (setq latexenc-main-file (concat file "." ext))))))))) | |
158 | ;; try tex-modes tex-guess-main-file | |
159 | (when (and (not latexenc-dont-use-tex-guess-main-file-flag) | |
160 | (not latexenc-main-file)) | |
161 | (when (fboundp 'tex-guess-main-file) | |
162 | (let ((tex-start-of-header "\\\\document\\(style\\|class\\)") | |
163 | (default-directory (file-name-directory (nth 1 arg-list)))) | |
164 | (setq latexenc-main-file (tex-guess-main-file))))) | |
165 | ;; if we found a master/main file get the coding system from it | |
166 | (if (and latexenc-main-file | |
167 | (file-readable-p latexenc-main-file)) | |
168 | (let* ((latexenc-dont-use-tex-guess-main-file-flag t) | |
169 | (latexenc-dont-use-TeX-master-flag t) | |
170 | (latexenc-main-buffer (find-file-noselect latexenc-main-file t))) | |
171 | (or (buffer-local-value 'coding-system-for-write latexenc-main-buffer) | |
172 | (buffer-local-value 'buffer-file-coding-system latexenc-main-buffer))) | |
173 | 'undecided)))) | |
174 | 'undecided)) | |
175 | ||
176 | (provide 'latexenc) | |
177 | ||
2183c76d | 178 | ;; arch-tag: f971bc3e-1fec-4609-8f2f-73dd41ab22e1 |
18acc655 | 179 | ;;; latexenc.el ends here |