Commit | Line | Data |
---|---|---|
a8581027 | 1 | ;;; format.el --- read and save files in multiple formats |
b578f267 | 2 | |
c84cf2bf | 3 | ;; Copyright (c) 1994, 1995, 1997 Free Software Foundation |
029894b9 | 4 | |
7865eac6 | 5 | ;; Author: Boris Goldowsky <boris@gnu.ai.mit.edu> |
029894b9 BG |
6 | |
7 | ;; This file is part of GNU Emacs. | |
8 | ||
9 | ;; GNU Emacs is free software; you can redistribute it and/or modify | |
10 | ;; it under the terms of the GNU General Public License as published by | |
11 | ;; the Free Software Foundation; either version 2, or (at your option) | |
12 | ;; any later version. | |
b578f267 | 13 | |
029894b9 BG |
14 | ;; GNU Emacs is distributed in the hope that it will be useful, |
15 | ;; but WITHOUT ANY WARRANTY; without even the implied warranty of | |
16 | ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
17 | ;; GNU General Public License for more details. | |
b578f267 | 18 | |
029894b9 | 19 | ;; You should have received a copy of the GNU General Public License |
b578f267 EN |
20 | ;; along with GNU Emacs; see the file COPYING. If not, write to the |
21 | ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
22 | ;; Boston, MA 02111-1307, USA. | |
029894b9 BG |
23 | |
24 | ;;; Commentary: | |
b578f267 EN |
25 | |
26 | ;; This file defines a unified mechanism for saving & loading files stored | |
27 | ;; in different formats. `format-alist' contains information that directs | |
029894b9 BG |
28 | ;; Emacs to call an encoding or decoding function when reading or writing |
29 | ;; files that match certain conditions. | |
30 | ;; | |
b578f267 EN |
31 | ;; When a file is visited, its format is determined by matching the |
32 | ;; beginning of the file against regular expressions stored in | |
33 | ;; `format-alist'. If this fails, you can manually translate the buffer | |
34 | ;; using `format-decode-buffer'. In either case, the formats used are | |
35 | ;; listed in the variable `buffer-file-format', and become the default | |
36 | ;; format for saving the buffer. To save a buffer in a different format, | |
37 | ;; change this variable, or use `format-write-file'. | |
029894b9 BG |
38 | ;; |
39 | ;; Auto-save files are normally created in the same format as the visited | |
b578f267 EN |
40 | ;; file, but the variable `auto-save-file-format' can be set to a |
41 | ;; particularly fast or otherwise preferred format to be used for | |
42 | ;; auto-saving (or nil to do no encoding on auto-save files, but then you | |
43 | ;; risk losing any text-properties in the buffer). | |
029894b9 | 44 | ;; |
b578f267 EN |
45 | ;; You can manually translate a buffer into or out of a particular format |
46 | ;; with the functions `format-encode-buffer' and `format-decode-buffer'. | |
47 | ;; To translate just the region use the functions `format-encode-region' | |
48 | ;; and `format-decode-region'. | |
029894b9 | 49 | ;; |
b578f267 EN |
50 | ;; You can define a new format by writing the encoding and decoding |
51 | ;; functions, and adding an entry to `format-alist'. See enriched.el for | |
52 | ;; an example of how to implement a file format. There are various | |
53 | ;; functions defined in this file that may be useful for writing the | |
54 | ;; encoding and decoding functions: | |
55 | ;; * `format-annotate-region' and `format-deannotate-region' allow a | |
56 | ;; single alist of information to be used for encoding and decoding. | |
57 | ;; The alist defines a correspondence between strings in the file | |
58 | ;; ("annotations") and text-properties in the buffer. | |
029894b9 BG |
59 | ;; * `format-replace-strings' is similarly useful for doing simple |
60 | ;; string->string translations in a reversible manner. | |
61 | ||
b578f267 EN |
62 | ;;; Code: |
63 | ||
029894b9 BG |
64 | (put 'buffer-file-format 'permanent-local t) |
65 | ||
4c8a3b19 | 66 | (defvar format-alist |
8d0dd9d2 | 67 | '((text/enriched "Extended MIME text/enriched format." |
029894b9 BG |
68 | "Content-[Tt]ype:[ \t]*text/enriched" |
69 | enriched-decode enriched-encode t enriched-mode) | |
c84cf2bf | 70 | (plain "ISO 8859-1 standard format, no text properties." |
029894b9 BG |
71 | ;; Plain only exists so that there is an obvious neutral choice in |
72 | ;; the completion list. | |
c84cf2bf RS |
73 | nil nil nil nil nil) |
74 | (ibm "IBM Code Page 850 (DOS)" | |
75 | "1\\(^\\)" | |
271af233 | 76 | "recode -f ibm-pc:latin1" "recode -f latin1:ibm-pc" t nil) |
c84cf2bf RS |
77 | (mac "Apple Macintosh" |
78 | "1\\(^\\)" | |
271af233 | 79 | "recode -f mac:latin1" "recode -f latin1:mac" t nil) |
c84cf2bf RS |
80 | (hp "HP Roman8" |
81 | "1\\(^\\)" | |
271af233 | 82 | "recode -f roman8:latin1" "recode -f latin1:roman8" t nil) |
c84cf2bf RS |
83 | (TeX "TeX (encoding)" |
84 | "1\\(^\\)" | |
85 | iso-tex2iso iso-iso2tex t nil) | |
86 | (gtex "German TeX (encoding)" | |
87 | "1\\(^\\)" | |
88 | iso-gtex2iso iso-iso2gtex t nil) | |
89 | (html "HTML (encoding)" | |
90 | "1\\(^\\)" | |
271af233 | 91 | "recode -f html:latin1" "recode -f latin1:html" t nil) |
c84cf2bf RS |
92 | (rot13 "rot13" |
93 | "1\\(^\\)" | |
94 | "tr a-mn-z n-za-m" "tr a-mn-z n-za-m" t nil) | |
95 | (duden "Duden Ersatzdarstellung" | |
96 | "1\\(^\\)" | |
97 | "diac" iso-iso2duden t nil) | |
98 | (de646 "German ASCII (ISO 646)" | |
99 | "1\\(^\\)" | |
271af233 | 100 | "recode -f iso646-ge:latin1" "recode -f latin1:iso646-ge" t nil) |
c84cf2bf RS |
101 | (denet "net German" |
102 | "1\\(^\\)" | |
103 | iso-german iso-cvt-read-only t nil) | |
104 | (esnet "net Spanish" | |
105 | "1\\(^\\)" | |
8d0dd9d2 | 106 | iso-spanish iso-cvt-read-only t nil)) |
029894b9 BG |
107 | "List of information about understood file formats. |
108 | Elements are of the form \(NAME DOC-STR REGEXP FROM-FN TO-FN MODIFY MODE-FN). | |
d1ae737b | 109 | |
029894b9 | 110 | NAME is a symbol, which is stored in `buffer-file-format'. |
d1ae737b | 111 | |
029894b9 BG |
112 | DOC-STR should be a single line providing more information about the |
113 | format. It is currently unused, but in the future will be shown to | |
114 | the user if they ask for more information. | |
d1ae737b | 115 | |
029894b9 BG |
116 | REGEXP is a regular expression to match against the beginning of the file; |
117 | it should match only files in that format. | |
d1ae737b | 118 | |
029894b9 BG |
119 | FROM-FN is called to decode files in that format; it gets two args, BEGIN |
120 | and END, and can make any modifications it likes, returning the new | |
121 | end. It must make sure that the beginning of the file no longer | |
122 | matches REGEXP, or else it will get called again. | |
d1ae737b RS |
123 | Alternatively, FROM-FN can be a string, which specifies a shell command |
124 | (including options) to be used as a filter to perform the conversion. | |
125 | ||
d1abb42f RS |
126 | TO-FN is called to encode a region into that format; it is passed three |
127 | arguments: BEGIN, END, and BUFFER. BUFFER is the original buffer that | |
128 | the data being written came from, which the function could use, for | |
129 | example, to find the values of local variables. TO-FN should either | |
130 | return a list of annotations like `write-region-annotate-functions', | |
131 | or modify the region and return the new end. | |
d1ae737b RS |
132 | Alternatively, TO-FN can be a string, which specifies a shell command |
133 | (including options) to be used as a filter to perform the conversion. | |
134 | ||
d1abb42f RS |
135 | MODIFY, if non-nil, means the TO-FN wants to modify the region. If nil, |
136 | TO-FN will not make any changes but will instead return a list of | |
137 | annotations. | |
d1ae737b | 138 | |
029894b9 BG |
139 | MODE-FN, if specified, is called when visiting a file with that format.") |
140 | ||
141 | ;;; Basic Functions (called from Lisp) | |
142 | ||
c84cf2bf RS |
143 | (defun format-encode-run-method (method from to &optional buffer) |
144 | "Translate using function or shell script METHOD the text from FROM to TO. | |
145 | If METHOD is a string, it is a shell command; | |
146 | otherwise, it should be a Lisp function. | |
147 | BUFFER should be the buffer that the output originally came from." | |
148 | (if (stringp method) | |
149 | (save-current-buffer | |
150 | (set-buffer buffer) | |
952936e8 RS |
151 | (with-output-to-temp-buffer "*Format Errors*" |
152 | (shell-command-on-region from to method t nil standard-output)) | |
c84cf2bf RS |
153 | (point)) |
154 | (funcall method from to buffer))) | |
155 | ||
156 | (defun format-decode-run-method (method from to &optional buffer) | |
157 | "Decode using function or shell script METHOD the text from FROM to TO. | |
158 | If METHOD is a string, it is a shell command; | |
159 | otherwise, it should be a Lisp function." | |
160 | (if (stringp method) | |
161 | (progn | |
952936e8 RS |
162 | (with-output-to-temp-buffer "*Format Errors*" |
163 | (shell-command-on-region from to method t nil standard-output)) | |
c84cf2bf RS |
164 | (point)) |
165 | (funcall method from to))) | |
166 | ||
d1abb42f | 167 | (defun format-annotate-function (format from to orig-buf) |
029894b9 BG |
168 | "Returns annotations for writing region as FORMAT. |
169 | FORMAT is a symbol naming one of the formats defined in `format-alist', | |
170 | it must be a single symbol, not a list like `buffer-file-format'. | |
d1abb42f RS |
171 | FROM and TO delimit the region to be operated on in the current buffer. |
172 | ORIG-BUF is the original buffer that the data came from. | |
029894b9 BG |
173 | This function works like a function on `write-region-annotate-functions': |
174 | it either returns a list of annotations, or returns with a different buffer | |
175 | current, which contains the modified text to write. | |
176 | ||
177 | For most purposes, consider using `format-encode-region' instead." | |
178 | ;; This function is called by write-region (actually build-annotations) | |
179 | ;; for each element of buffer-file-format. | |
180 | (let* ((info (assq format format-alist)) | |
181 | (to-fn (nth 4 info)) | |
182 | (modify (nth 5 info))) | |
183 | (if to-fn | |
184 | (if modify | |
185 | ;; To-function wants to modify region. Copy to safe place. | |
186 | (let ((copy-buf (get-buffer-create " *Format Temp*"))) | |
187 | (copy-to-buffer copy-buf from to) | |
188 | (set-buffer copy-buf) | |
189 | (format-insert-annotations write-region-annotations-so-far from) | |
c84cf2bf | 190 | (format-encode-run-method to-fn (point-min) (point-max) orig-buf) |
029894b9 BG |
191 | nil) |
192 | ;; Otherwise just call function, it will return annotations. | |
d1abb42f | 193 | (funcall to-fn from to orig-buf))))) |
029894b9 BG |
194 | |
195 | (defun format-decode (format length &optional visit-flag) | |
196 | ;; This function is called by insert-file-contents whenever a file is read. | |
197 | "Decode text from any known FORMAT. | |
198 | FORMAT is a symbol appearing in `format-alist' or a list of such symbols, | |
199 | or nil, in which case this function tries to guess the format of the data by | |
200 | matching against the regular expressions in `format-alist'. After a match is | |
201 | found and the region decoded, the alist is searched again from the beginning | |
202 | for another match. | |
203 | ||
204 | Second arg LENGTH is the number of characters following point to operate on. | |
205 | If optional third arg VISIT-FLAG is true, set `buffer-file-format' | |
206 | to the list of formats used, and call any mode functions defined for those | |
207 | formats. | |
208 | ||
209 | Returns the new length of the decoded region. | |
210 | ||
211 | For most purposes, consider using `format-decode-region' instead." | |
212 | (let ((mod (buffer-modified-p)) | |
213 | (begin (point)) | |
214 | (end (+ (point) length))) | |
215 | (if (null format) | |
216 | ;; Figure out which format it is in, remember list in `format'. | |
217 | (let ((try format-alist)) | |
218 | (while try | |
219 | (let* ((f (car try)) | |
220 | (regexp (nth 2 f)) | |
221 | (p (point))) | |
222 | (if (and regexp (looking-at regexp) | |
223 | (< (match-end 0) (+ begin length))) | |
224 | (progn | |
225 | (setq format (cons (car f) format)) | |
226 | ;; Decode it | |
c84cf2bf RS |
227 | (if (nth 3 f) |
228 | (setq end (format-decode-run-method (nth 3 f) begin end))) | |
029894b9 BG |
229 | ;; Call visit function if required |
230 | (if (and visit-flag (nth 6 f)) (funcall (nth 6 f) 1)) | |
231 | ;; Safeguard against either of the functions changing pt. | |
232 | (goto-char p) | |
233 | ;; Rewind list to look for another format | |
234 | (setq try format-alist)) | |
235 | (setq try (cdr try)))))) | |
236 | ;; Deal with given format(s) | |
237 | (or (listp format) (setq format (list format))) | |
238 | (let ((do format) f) | |
239 | (while do | |
240 | (or (setq f (assq (car do) format-alist)) | |
241 | (error "Unknown format" (car do))) | |
242 | ;; Decode: | |
c84cf2bf RS |
243 | (if (nth 3 f) |
244 | (setq end (format-decode-run-method (nth 3 f) begin end))) | |
029894b9 BG |
245 | ;; Call visit function if required |
246 | (if (and visit-flag (nth 6 f)) (funcall (nth 6 f) 1)) | |
247 | (setq do (cdr do))))) | |
248 | (if visit-flag | |
249 | (setq buffer-file-format format)) | |
250 | (set-buffer-modified-p mod) | |
251 | ;; Return new length of region | |
252 | (- end begin))) | |
253 | ||
254 | ;;; | |
255 | ;;; Interactive functions & entry points | |
256 | ;;; | |
257 | ||
258 | (defun format-decode-buffer (&optional format) | |
259 | "Translate the buffer from some FORMAT. | |
260 | If the format is not specified, this function attempts to guess. | |
261 | `buffer-file-format' is set to the format used, and any mode-functions | |
262 | for the format are called." | |
263 | (interactive | |
264 | (list (format-read "Translate buffer from format (default: guess): "))) | |
265 | (save-excursion | |
266 | (goto-char (point-min)) | |
267 | (format-decode format (buffer-size) t))) | |
268 | ||
269 | (defun format-decode-region (from to &optional format) | |
270 | "Decode the region from some format. | |
271 | Arg FORMAT is optional; if omitted the format will be determined by looking | |
272 | for identifying regular expressions at the beginning of the region." | |
273 | (interactive | |
274 | (list (region-beginning) (region-end) | |
275 | (format-read "Translate region from format (default: guess): "))) | |
276 | (save-excursion | |
277 | (goto-char from) | |
278 | (format-decode format (- to from) nil))) | |
279 | ||
280 | (defun format-encode-buffer (&optional format) | |
281 | "Translate the buffer into FORMAT. | |
282 | FORMAT defaults to `buffer-file-format'. It is a symbol naming one of the | |
283 | formats defined in `format-alist', or a list of such symbols." | |
284 | (interactive | |
285 | (list (format-read (format "Translate buffer to format (default %s): " | |
286 | buffer-file-format)))) | |
287 | (format-encode-region (point-min) (point-max) format)) | |
288 | ||
289 | (defun format-encode-region (beg end &optional format) | |
290 | "Translate the region into some FORMAT. | |
291 | FORMAT defaults to `buffer-file-format', it is a symbol naming | |
292 | one of the formats defined in `format-alist', or a list of such symbols." | |
293 | (interactive | |
294 | (list (region-beginning) (region-end) | |
295 | (format-read (format "Translate region to format (default %s): " | |
296 | buffer-file-format)))) | |
297 | (if (null format) (setq format buffer-file-format)) | |
298 | (if (symbolp format) (setq format (list format))) | |
299 | (save-excursion | |
300 | (goto-char end) | |
301 | (let ((cur-buf (current-buffer)) | |
302 | (end (point-marker))) | |
303 | (while format | |
304 | (let* ((info (assq (car format) format-alist)) | |
305 | (to-fn (nth 4 info)) | |
306 | (modify (nth 5 info)) | |
307 | result) | |
308 | (if to-fn | |
309 | (if modify | |
c84cf2bf RS |
310 | (setq end (format-encode-run-method to-fn beg end |
311 | (current-buffer))) | |
d1abb42f RS |
312 | (format-insert-annotations |
313 | (funcall to-fn beg end (current-buffer))))) | |
029894b9 BG |
314 | (setq format (cdr format))))))) |
315 | ||
316 | (defun format-write-file (filename format) | |
317 | "Write current buffer into a FILE using some FORMAT. | |
318 | Makes buffer visit that file and sets the format as the default for future | |
319 | saves. If the buffer is already visiting a file, you can specify a directory | |
320 | name as FILE, to write a file of the same old name in that directory." | |
321 | (interactive | |
322 | ;; Same interactive spec as write-file, plus format question. | |
323 | (let* ((file (if buffer-file-name | |
324 | (read-file-name "Write file: " | |
325 | nil nil nil nil) | |
326 | (read-file-name "Write file: " | |
327 | (cdr (assq 'default-directory | |
328 | (buffer-local-variables))) | |
329 | nil nil (buffer-name)))) | |
330 | (fmt (format-read (format "Write file `%s' in format: " | |
331 | (file-name-nondirectory file))))) | |
332 | (list file fmt))) | |
333 | (setq buffer-file-format format) | |
334 | (write-file filename)) | |
335 | ||
ce6af52b KH |
336 | (defun format-find-file (filename format) |
337 | "Find the file FILE using data format FORMAT. | |
338 | If FORMAT is nil then do not do any format conversion." | |
339 | (interactive | |
340 | ;; Same interactive spec as write-file, plus format question. | |
341 | (let* ((file (read-file-name "Find file: ")) | |
342 | (fmt (format-read (format "Read file `%s' in format: " | |
343 | (file-name-nondirectory file))))) | |
344 | (list file fmt))) | |
345 | (let ((format-alist nil)) | |
346 | (find-file filename)) | |
347 | (if format | |
348 | (format-decode-buffer format))) | |
349 | ||
350 | (defun format-insert-file (filename format &optional beg end) | |
351 | "Insert the contents of file FILE using data format FORMAT. | |
352 | If FORMAT is nil then do not do any format conversion. | |
353 | The optional third and fourth arguments BEG and END specify | |
354 | the part of the file to read. | |
355 | ||
356 | The return value is like the value of `insert-file-contents': | |
357 | a list (ABSOLUTE-FILE-NAME . SIZE)." | |
358 | (interactive | |
359 | ;; Same interactive spec as write-file, plus format question. | |
360 | (let* ((file (read-file-name "Find file: ")) | |
361 | (fmt (format-read (format "Read file `%s' in format: " | |
362 | (file-name-nondirectory file))))) | |
363 | (list file fmt))) | |
364 | (let (value size) | |
365 | (let ((format-alist nil)) | |
366 | (setq value (insert-file-contents filename nil beg end)) | |
367 | (setq size (nth 1 value))) | |
368 | (if format | |
c84cf2bf | 369 | (setq size (format-decode format size) |
ce6af52b KH |
370 | value (cons (car value) size))) |
371 | value)) | |
372 | ||
029894b9 BG |
373 | (defun format-read (&optional prompt) |
374 | "Read and return the name of a format. | |
375 | Return value is a list, like `buffer-file-format'; it may be nil. | |
376 | Formats are defined in `format-alist'. Optional arg is the PROMPT to use." | |
377 | (let* ((table (mapcar (lambda (x) (list (symbol-name (car x)))) | |
378 | format-alist)) | |
379 | (ans (completing-read (or prompt "Format: ") table nil t))) | |
380 | (if (not (equal "" ans)) (list (intern ans))))) | |
381 | ||
382 | ||
383 | ;;; | |
384 | ;;; Below are some functions that may be useful in writing encoding and | |
385 | ;;; decoding functions for use in format-alist. | |
386 | ;;; | |
387 | ||
388 | (defun format-replace-strings (alist &optional reverse beg end) | |
389 | "Do multiple replacements on the buffer. | |
390 | ALIST is a list of (from . to) pairs, which should be proper arguments to | |
391 | `search-forward' and `replace-match' respectively. | |
392 | Optional 2nd arg REVERSE, if non-nil, means the pairs are (to . from), so that | |
393 | you can use the same list in both directions if it contains only literal | |
394 | strings. | |
395 | Optional args BEGIN and END specify a region of the buffer to operate on." | |
396 | (save-excursion | |
397 | (save-restriction | |
398 | (or beg (setq beg (point-min))) | |
399 | (if end (narrow-to-region (point-min) end)) | |
400 | (while alist | |
401 | (let ((from (if reverse (cdr (car alist)) (car (car alist)))) | |
402 | (to (if reverse (car (cdr alist)) (cdr (car alist))))) | |
403 | (goto-char beg) | |
404 | (while (search-forward from nil t) | |
405 | (goto-char (match-beginning 0)) | |
406 | (insert to) | |
407 | (set-text-properties (- (point) (length to)) (point) | |
408 | (text-properties-at (point))) | |
409 | (delete-region (point) (+ (point) (- (match-end 0) | |
410 | (match-beginning 0))))) | |
411 | (setq alist (cdr alist))))))) | |
412 | ||
413 | ;;; Some list-manipulation functions that we need. | |
414 | ||
415 | (defun format-delq-cons (cons list) | |
416 | "Remove the given CONS from LIST by side effect, | |
417 | and return the new LIST. Since CONS could be the first element | |
418 | of LIST, write `\(setq foo \(format-delq-cons element foo))' to be sure of | |
419 | changing the value of `foo'." | |
420 | (if (eq cons list) | |
421 | (cdr list) | |
422 | (let ((p list)) | |
423 | (while (not (eq (cdr p) cons)) | |
424 | (if (null p) (error "format-delq-cons: not an element.")) | |
425 | (setq p (cdr p))) | |
426 | ;; Now (cdr p) is the cons to delete | |
427 | (setcdr p (cdr cons)) | |
428 | list))) | |
429 | ||
430 | (defun format-make-relatively-unique (a b) | |
431 | "Delete common elements of lists A and B, return as pair. | |
432 | Compares using `equal'." | |
433 | (let* ((acopy (copy-sequence a)) | |
434 | (bcopy (copy-sequence b)) | |
435 | (tail acopy)) | |
436 | (while tail | |
437 | (let ((dup (member (car tail) bcopy)) | |
438 | (next (cdr tail))) | |
439 | (if dup (setq acopy (format-delq-cons tail acopy) | |
440 | bcopy (format-delq-cons dup bcopy))) | |
441 | (setq tail next))) | |
442 | (cons acopy bcopy))) | |
443 | ||
444 | (defun format-common-tail (a b) | |
445 | "Given two lists that have a common tail, return it. | |
446 | Compares with `equal', and returns the part of A that is equal to the | |
447 | equivalent part of B. If even the last items of the two are not equal, | |
448 | returns nil." | |
449 | (let ((la (length a)) | |
450 | (lb (length b))) | |
451 | ;; Make sure they are the same length | |
452 | (if (> la lb) | |
453 | (setq a (nthcdr (- la lb) a)) | |
454 | (setq b (nthcdr (- lb la) b)))) | |
455 | (while (not (equal a b)) | |
456 | (setq a (cdr a) | |
457 | b (cdr b))) | |
458 | a) | |
459 | ||
460 | (defun format-reorder (items order) | |
461 | "Arrange ITEMS to following partial ORDER. | |
462 | Elements of ITEMS equal to elements of ORDER will be rearranged to follow the | |
463 | ORDER. Unmatched items will go last." | |
464 | (if order | |
465 | (let ((item (member (car order) items))) | |
466 | (if item | |
467 | (cons (car item) | |
468 | (format-reorder (format-delq-cons item items) | |
469 | (cdr order))) | |
470 | (format-reorder items (cdr order)))) | |
471 | items)) | |
472 | ||
473 | (put 'face 'format-list-valued t) ; These text-properties take values | |
474 | (put 'unknown 'format-list-valued t) ; that are lists, the elements of which | |
475 | ; should be considered separately. | |
476 | ; See format-deannotate-region and | |
477 | ; format-annotate-region. | |
478 | ||
479 | ;;; | |
480 | ;;; Decoding | |
481 | ;;; | |
482 | ||
483 | (defun format-deannotate-region (from to translations next-fn) | |
484 | "Translate annotations in the region into text properties. | |
485 | This sets text properties between FROM to TO as directed by the | |
486 | TRANSLATIONS and NEXT-FN arguments. | |
487 | ||
488 | NEXT-FN is a function that searches forward from point for an annotation. | |
489 | It should return a list of 4 elements: \(BEGIN END NAME POSITIVE). BEGIN and | |
490 | END are buffer positions bounding the annotation, NAME is the name searched | |
491 | for in TRANSLATIONS, and POSITIVE should be non-nil if this annotation marks | |
492 | the beginning of a region with some property, or nil if it ends the region. | |
493 | NEXT-FN should return nil if there are no annotations after point. | |
494 | ||
495 | The basic format of the TRANSLATIONS argument is described in the | |
496 | documentation for the `format-annotate-region' function. There are some | |
497 | additional things to keep in mind for decoding, though: | |
498 | ||
499 | When an annotation is found, the TRANSLATIONS list is searched for a | |
500 | text-property name and value that corresponds to that annotation. If the | |
501 | text-property has several annotations associated with it, it will be used only | |
502 | if the other annotations are also in effect at that point. The first match | |
503 | found whose annotations are all present is used. | |
504 | ||
505 | The text property thus determined is set to the value over the region between | |
506 | the opening and closing annotations. However, if the text-property name has a | |
507 | non-nil `format-list-valued' property, then the value will be consed onto the | |
508 | surrounding value of the property, rather than replacing that value. | |
509 | ||
510 | There are some special symbols that can be used in the \"property\" slot of | |
511 | the TRANSLATIONS list: PARAMETER and FUNCTION \(spelled in uppercase). | |
512 | Annotations listed under the pseudo-property PARAMETER are considered to be | |
513 | arguments of the immediately surrounding annotation; the text between the | |
514 | opening and closing parameter annotations is deleted from the buffer but saved | |
515 | as a string. The surrounding annotation should be listed under the | |
516 | pseudo-property FUNCTION. Instead of inserting a text-property for this | |
517 | annotation, the function listed in the VALUE slot is called to make whatever | |
518 | changes are appropriate. The function's first two arguments are the START and | |
519 | END locations, and the rest of the arguments are any PARAMETERs found in that | |
520 | region. | |
521 | ||
522 | Any annotations that are found by NEXT-FN but not defined by TRANSLATIONS | |
523 | are saved as values of the `unknown' text-property \(which is list-valued). | |
524 | The TRANSLATIONS list should usually contain an entry of the form | |
525 | \(unknown \(nil format-annotate-value)) | |
526 | to write these unknown annotations back into the file." | |
527 | (save-excursion | |
528 | (save-restriction | |
529 | (narrow-to-region (point-min) to) | |
530 | (goto-char from) | |
531 | (let (next open-ans todo loc unknown-ans) | |
532 | (while (setq next (funcall next-fn)) | |
533 | (let* ((loc (nth 0 next)) | |
534 | (end (nth 1 next)) | |
535 | (name (nth 2 next)) | |
536 | (positive (nth 3 next)) | |
537 | (found nil)) | |
538 | ||
539 | ;; Delete the annotation | |
540 | (delete-region loc end) | |
22828206 RS |
541 | (cond |
542 | ;; Positive annotations are stacked, remembering location | |
543 | (positive (setq open-ans (cons `(,name ((,loc . nil))) open-ans))) | |
544 | ;; It is a negative annotation: | |
545 | ;; Close the top annotation & add its text property. | |
546 | ;; If the file's nesting is messed up, the close might not match | |
547 | ;; the top thing on the open-annotations stack. | |
548 | ;; If no matching annotation is open, just ignore the close. | |
549 | ((not (assoc name open-ans)) | |
550 | (message "Extra closing annotation (%s) in file" name)) | |
551 | ;; If one is open, but not on the top of the stack, close | |
552 | ;; the things in between as well. Set `found' when the real | |
553 | ;; one is closed. | |
554 | (t | |
555 | (while (not found) | |
556 | (let* ((top (car open-ans)) ; first on stack: should match. | |
557 | (top-name (car top)) ; text property name | |
558 | (top-extents (nth 1 top)) ; property regions | |
559 | (params (cdr (cdr top))) ; parameters | |
560 | (aalist translations) | |
561 | (matched nil)) | |
562 | (if (equal name top-name) | |
563 | (setq found t) | |
564 | (message "Improper nesting in file.")) | |
565 | ;; Look through property names in TRANSLATIONS | |
566 | (while aalist | |
567 | (let ((prop (car (car aalist))) | |
568 | (alist (cdr (car aalist)))) | |
569 | ;; And look through values for each property | |
570 | (while alist | |
571 | (let ((value (car (car alist))) | |
572 | (ans (cdr (car alist)))) | |
573 | (if (member top-name ans) | |
574 | ;; This annotation is listed, but still have to | |
575 | ;; check if multiple annotations are satisfied | |
576 | (if (member nil (mapcar (lambda (r) | |
577 | (assoc r open-ans)) | |
578 | ans)) | |
579 | nil ; multiple ans not satisfied | |
580 | ;; If there are multiple annotations going | |
581 | ;; into one text property, split up the other | |
582 | ;; annotations so they apply individually to | |
583 | ;; the other regions. | |
584 | (setcdr (car top-extents) loc) | |
585 | (let ((to-split ans) this-one extents) | |
586 | (while to-split | |
587 | (setq this-one | |
588 | (assoc (car to-split) open-ans) | |
589 | extents (nth 1 this-one)) | |
590 | (if (not (eq this-one top)) | |
591 | (setcar (cdr this-one) | |
592 | (format-subtract-regions | |
593 | extents top-extents))) | |
594 | (setq to-split (cdr to-split)))) | |
595 | ;; Set loop variables to nil so loop | |
596 | ;; will exit. | |
597 | (setq alist nil aalist nil matched t | |
598 | ;; pop annotation off stack. | |
599 | open-ans (cdr open-ans)) | |
600 | (let ((extents top-extents) | |
601 | (start (car (car top-extents))) | |
602 | (loc (cdr (car top-extents)))) | |
603 | (while extents | |
604 | (cond | |
605 | ;; Check for pseudo-properties | |
606 | ((eq prop 'PARAMETER) | |
607 | ;; A parameter of the top open ann: | |
608 | ;; delete text and use as arg. | |
609 | (if open-ans | |
610 | ;; (If nothing open, discard). | |
611 | (setq open-ans | |
612 | (cons | |
613 | (append (car open-ans) | |
614 | (list | |
615 | (buffer-substring | |
616 | start loc))) | |
617 | (cdr open-ans)))) | |
618 | (delete-region start loc)) | |
619 | ((eq prop 'FUNCTION) | |
620 | ;; Not a property, but a function. | |
621 | (let ((rtn | |
622 | (apply value start loc params))) | |
623 | (if rtn (setq todo (cons rtn todo))))) | |
624 | (t | |
625 | ;; Normal property/value pair | |
626 | (setq todo | |
627 | (cons (list start loc prop value) | |
628 | todo)))) | |
629 | (setq extents (cdr extents) | |
630 | start (car (car extents)) | |
631 | loc (cdr (car extents)))))))) | |
632 | (setq alist (cdr alist)))) | |
633 | (setq aalist (cdr aalist))) | |
634 | (if (not matched) | |
029894b9 BG |
635 | ;; Didn't find any match for the annotation: |
636 | ;; Store as value of text-property `unknown'. | |
22828206 RS |
637 | (let ((extents top-extents) |
638 | (start (car (car top-extents))) | |
0b140219 | 639 | (loc (or (cdr (car top-extents)) loc))) |
22828206 RS |
640 | (while extents |
641 | (setq open-ans (cdr open-ans) | |
642 | todo (cons (list start loc 'unknown top-name) | |
643 | todo) | |
644 | unknown-ans (cons name unknown-ans) | |
645 | extents (cdr extents) | |
646 | start (car (car extents)) | |
647 | loc (cdr (car extents)))))))))))) | |
029894b9 BG |
648 | |
649 | ;; Once entire file has been scanned, add the properties. | |
650 | (while todo | |
651 | (let* ((item (car todo)) | |
652 | (from (nth 0 item)) | |
653 | (to (nth 1 item)) | |
654 | (prop (nth 2 item)) | |
655 | (val (nth 3 item))) | |
22828206 RS |
656 | |
657 | (if (numberp val) ; add to ambient value if numeric | |
658 | (format-property-increment-region from to prop val 0) | |
659 | (put-text-property | |
029894b9 | 660 | from to prop |
22828206 | 661 | (cond ((get prop 'format-list-valued) ; value gets consed onto |
029894b9 BG |
662 | ; list-valued properties |
663 | (let ((prev (get-text-property from prop))) | |
664 | (cons val (if (listp prev) prev (list prev))))) | |
22828206 | 665 | (t val))))) ; normally, just set to val. |
029894b9 | 666 | (setq todo (cdr todo))) |
22828206 | 667 | |
029894b9 BG |
668 | (if unknown-ans |
669 | (message "Unknown annotations: %s" unknown-ans)))))) | |
670 | ||
22828206 RS |
671 | (defun format-subtract-regions (minu subtra) |
672 | "Remove the regions in SUBTRAHEND from the regions in MINUEND. A region | |
673 | is a dotted pair (from . to). Both parameters are lists of regions. Each | |
674 | list must contain nonoverlapping, noncontiguous regions, in descending | |
675 | order. The result is also nonoverlapping, noncontiguous, and in descending | |
676 | order. The first element of MINUEND can have a cdr of nil, indicating that | |
677 | the end of that region is not yet known." | |
678 | (let* ((minuend (copy-alist minu)) | |
679 | (subtrahend (copy-alist subtra)) | |
680 | (m (car minuend)) | |
681 | (s (car subtrahend)) | |
682 | results) | |
683 | (while (and minuend subtrahend) | |
684 | (cond | |
685 | ;; The minuend starts after the subtrahend ends; keep it. | |
686 | ((> (car m) (cdr s)) | |
687 | (setq results (cons m results) | |
688 | minuend (cdr minuend) | |
689 | m (car minuend))) | |
690 | ;; The minuend extends beyond the end of the subtrahend. Chop it off. | |
691 | ((or (null (cdr m)) (> (cdr m) (cdr s))) | |
692 | (setq results (cons (cons (1+ (cdr s)) (cdr m)) results)) | |
693 | (setcdr m (cdr s))) | |
694 | ;; The subtrahend starts after the minuend ends; throw it away. | |
695 | ((< (cdr m) (car s)) | |
696 | (setq subtrahend (cdr subtrahend) s (car subtrahend))) | |
697 | ;; The subtrahend extends beyond the end of the minuend. Chop it off. | |
698 | (t ;(<= (cdr m) (cdr s))) | |
699 | (if (>= (car m) (car s)) | |
700 | (setq minuend (cdr minuend) m (car minuend)) | |
701 | (setcdr m (1- (car s))) | |
702 | (setq subtrahend (cdr subtrahend) s (car subtrahend)))))) | |
703 | (nconc (nreverse results) minuend))) | |
704 | ||
705 | ;; This should probably go somewhere other than format.el. Then again, | |
706 | ;; indent.el has alter-text-property. NOTE: We can also use | |
707 | ;; next-single-property-change instead of text-property-not-all, but then | |
708 | ;; we have to see if we passed TO. | |
709 | (defun format-property-increment-region (from to prop delta default) | |
710 | "Increment property PROP over the region between FROM and TO by the | |
711 | amount DELTA (which may be negative). If property PROP is nil anywhere | |
712 | in the region, it is treated as though it were DEFAULT." | |
713 | (let ((cur from) val newval next) | |
714 | (while cur | |
715 | (setq val (get-text-property cur prop) | |
716 | newval (+ (or val default) delta) | |
717 | next (text-property-not-all cur to prop val)) | |
718 | (put-text-property cur (or next to) prop newval) | |
719 | (setq cur next)))) | |
720 | ||
029894b9 BG |
721 | ;;; |
722 | ;;; Encoding | |
723 | ;;; | |
724 | ||
725 | (defun format-insert-annotations (list &optional offset) | |
726 | "Apply list of annotations to buffer as `write-region' would. | |
727 | Inserts each element of the given LIST of buffer annotations at its | |
728 | appropriate place. Use second arg OFFSET if the annotations' locations are | |
729 | not relative to the beginning of the buffer: annotations will be inserted | |
730 | at their location-OFFSET+1 \(ie, the offset is treated as the character number | |
731 | of the first character in the buffer)." | |
732 | (if (not offset) | |
733 | (setq offset 0) | |
734 | (setq offset (1- offset))) | |
735 | (let ((l (reverse list))) | |
736 | (while l | |
737 | (goto-char (- (car (car l)) offset)) | |
738 | (insert (cdr (car l))) | |
739 | (setq l (cdr l))))) | |
740 | ||
741 | (defun format-annotate-value (old new) | |
742 | "Return OLD and NEW as a \(close . open) annotation pair. | |
743 | Useful as a default function for TRANSLATIONS alist when the value of the text | |
744 | property is the name of the annotation that you want to use, as it is for the | |
745 | `unknown' text property." | |
746 | (cons (if old (list old)) | |
747 | (if new (list new)))) | |
748 | ||
749 | (defun format-annotate-region (from to trans format-fn ignore) | |
750 | "Generate annotations for text properties in the region. | |
751 | Searches for changes between FROM and TO, and describes them with a list of | |
752 | annotations as defined by alist TRANSLATIONS and FORMAT-FN. IGNORE lists text | |
753 | properties not to consider; any text properties that are neither ignored nor | |
754 | listed in TRANSLATIONS are warned about. | |
755 | If you actually want to modify the region, give the return value of this | |
756 | function to `format-insert-annotations'. | |
757 | ||
758 | Format of the TRANSLATIONS argument: | |
759 | ||
760 | Each element is a list whose car is a PROPERTY, and the following | |
761 | elements are VALUES of that property followed by the names of zero or more | |
762 | ANNOTATIONS. Whenever the property takes on that value, the annotations | |
763 | \(as formatted by FORMAT-FN) are inserted into the file. | |
764 | When the property stops having that value, the matching negated annotation | |
765 | will be inserted \(it may actually be closed earlier and reopened, if | |
766 | necessary, to keep proper nesting). | |
767 | ||
768 | If the property's value is a list, then each element of the list is dealt with | |
769 | separately. | |
770 | ||
771 | If a VALUE is numeric, then it is assumed that there is a single annotation | |
772 | and each occurrence of it increments the value of the property by that number. | |
773 | Thus, given the entry \(left-margin \(4 \"indent\")), if the left margin | |
774 | changes from 4 to 12, two <indent> annotations will be generated. | |
775 | ||
776 | If the VALUE is nil, then instead of annotations, a function should be | |
777 | specified. This function is used as a default: it is called for all | |
778 | transitions not explicitly listed in the table. The function is called with | |
779 | two arguments, the OLD and NEW values of the property. It should return | |
780 | lists of annotations like `format-annotate-location' does. | |
781 | ||
782 | The same structure can be used in reverse for reading files." | |
783 | (let ((all-ans nil) ; All annotations - becomes return value | |
784 | (open-ans nil) ; Annotations not yet closed | |
785 | (loc nil) ; Current location | |
786 | (not-found nil)) ; Properties that couldn't be saved | |
787 | (while (or (null loc) | |
788 | (and (setq loc (next-property-change loc nil to)) | |
789 | (< loc to))) | |
790 | (or loc (setq loc from)) | |
791 | (let* ((ans (format-annotate-location loc (= loc from) ignore trans)) | |
792 | (neg-ans (format-reorder (aref ans 0) open-ans)) | |
793 | (pos-ans (aref ans 1)) | |
794 | (ignored (aref ans 2))) | |
795 | (setq not-found (append ignored not-found) | |
796 | ignore (append ignored ignore)) | |
797 | ;; First do the negative (closing) annotations | |
798 | (while neg-ans | |
799 | ;; Check if it's missing. This can happen (eg, a numeric property | |
800 | ;; going negative can generate closing annotations before there are | |
801 | ;; any open). Warn user & ignore. | |
802 | (if (not (member (car neg-ans) open-ans)) | |
803 | (message "Can't close %s: not open." (car neg-ans)) | |
804 | (while (not (equal (car neg-ans) (car open-ans))) | |
805 | ;; To close anno. N, need to first close ans 1 to N-1, | |
806 | ;; remembering to re-open them later. | |
807 | (setq pos-ans (cons (car open-ans) pos-ans)) | |
808 | (setq all-ans | |
809 | (cons (cons loc (funcall format-fn (car open-ans) nil)) | |
810 | all-ans)) | |
811 | (setq open-ans (cdr open-ans))) | |
812 | ;; Now remove the one we're really interested in from open list. | |
813 | (setq open-ans (cdr open-ans)) | |
814 | ;; And put the closing annotation here. | |
815 | (setq all-ans | |
816 | (cons (cons loc (funcall format-fn (car neg-ans) nil)) | |
817 | all-ans))) | |
818 | (setq neg-ans (cdr neg-ans))) | |
819 | ;; Now deal with positive (opening) annotations | |
820 | (let ((p pos-ans)) | |
821 | (while pos-ans | |
822 | (setq open-ans (cons (car pos-ans) open-ans)) | |
823 | (setq all-ans | |
824 | (cons (cons loc (funcall format-fn (car pos-ans) t)) | |
825 | all-ans)) | |
826 | (setq pos-ans (cdr pos-ans)))))) | |
827 | ||
828 | ;; Close any annotations still open | |
829 | (while open-ans | |
830 | (setq all-ans | |
831 | (cons (cons to (funcall format-fn (car open-ans) nil)) | |
832 | all-ans)) | |
833 | (setq open-ans (cdr open-ans))) | |
834 | (if not-found | |
835 | (message "These text properties could not be saved:\n %s" | |
836 | not-found)) | |
837 | (nreverse all-ans))) | |
838 | ||
839 | ;;; Internal functions for format-annotate-region. | |
840 | ||
841 | (defun format-annotate-location (loc all ignore trans) | |
842 | "Return annotation(s) needed at LOCATION. | |
843 | This includes any properties that change between LOC-1 and LOC. | |
844 | If ALL is true, don't look at previous location, but generate annotations for | |
845 | all non-nil properties. | |
846 | Third argument IGNORE is a list of text-properties not to consider. | |
847 | ||
848 | Return value is a vector of 3 elements: | |
849 | 1. List of names of the annotations to close | |
850 | 2. List of the names of annotations to open. | |
851 | 3. List of properties that were ignored or couldn't be annotated." | |
852 | (let* ((prev-loc (1- loc)) | |
853 | (before-plist (if all nil (text-properties-at prev-loc))) | |
854 | (after-plist (text-properties-at loc)) | |
855 | p negatives positives prop props not-found) | |
856 | ;; make list of all property names involved | |
857 | (setq p before-plist) | |
858 | (while p | |
859 | (if (not (memq (car p) props)) | |
860 | (setq props (cons (car p) props))) | |
861 | (setq p (cdr (cdr p)))) | |
862 | (setq p after-plist) | |
863 | (while p | |
864 | (if (not (memq (car p) props)) | |
865 | (setq props (cons (car p) props))) | |
866 | (setq p (cdr (cdr p)))) | |
867 | ||
868 | (while props | |
869 | (setq prop (car props) | |
870 | props (cdr props)) | |
871 | (if (memq prop ignore) | |
872 | nil ; If it's been ignored before, ignore it now. | |
873 | (let ((before (if all nil (car (cdr (memq prop before-plist))))) | |
874 | (after (car (cdr (memq prop after-plist))))) | |
875 | (if (equal before after) | |
876 | nil ; no change; ignore | |
877 | (let ((result (format-annotate-single-property-change | |
878 | prop before after trans))) | |
879 | (if (not result) | |
880 | (setq not-found (cons prop not-found)) | |
881 | (setq negatives (nconc negatives (car result)) | |
882 | positives (nconc positives (cdr result))))))))) | |
883 | (vector negatives positives not-found))) | |
884 | ||
885 | (defun format-annotate-single-property-change (prop old new trans) | |
886 | "Return annotations for PROPERTY changing from OLD to NEW. | |
887 | These are searched for in the TRANSLATIONS alist. | |
888 | If NEW does not appear in the list, but there is a default function, then that | |
889 | function is called. | |
890 | Annotations to open and to close are returned as a dotted pair." | |
891 | (let ((prop-alist (cdr (assoc prop trans))) | |
892 | default) | |
893 | (if (not prop-alist) | |
894 | nil | |
029894b9 BG |
895 | ;; If either old or new is a list, have to treat both that way. |
896 | (if (or (consp old) (consp new)) | |
897 | (let* ((old (if (listp old) old (list old))) | |
898 | (new (if (listp new) new (list new))) | |
899 | (tail (format-common-tail old new)) | |
900 | close open) | |
901 | (while old | |
902 | (setq close | |
903 | (append (car (format-annotate-atomic-property-change | |
904 | prop-alist (car old) nil)) | |
905 | close) | |
906 | old (cdr old))) | |
907 | (while new | |
908 | (setq open | |
909 | (append (cdr (format-annotate-atomic-property-change | |
910 | prop-alist nil (car new))) | |
911 | open) | |
912 | new (cdr new))) | |
913 | (format-make-relatively-unique close open)) | |
914 | (format-annotate-atomic-property-change prop-alist old new))))) | |
915 | ||
916 | (defun format-annotate-atomic-property-change (prop-alist old new) | |
917 | "Internal function annotate a single property change. | |
fa1f61e5 | 918 | PROP-ALIST is the relevant segment of a TRANSLATIONS list. |
029894b9 | 919 | OLD and NEW are the values." |
e5a60108 RS |
920 | (let (num-ann) |
921 | ;; If old and new values are numbers, | |
922 | ;; look for a number in PROP-ALIST. | |
5f525595 RS |
923 | (if (and (or (null old) (numberp old)) |
924 | (or (null new) (numberp new))) | |
e5a60108 RS |
925 | (progn |
926 | (setq num-ann prop-alist) | |
927 | (while (and num-ann (not (numberp (car (car num-ann))))) | |
928 | (setq num-ann (cdr num-ann))))) | |
929 | (if num-ann | |
5f525595 | 930 | ;; Numerical annotation - use difference |
34fbecde RS |
931 | (progn |
932 | ;; If property is numeric, nil means 0 | |
933 | (cond ((and (numberp old) (null new)) | |
934 | (setq new 0)) | |
935 | ((and (numberp new) (null old)) | |
936 | (setq old 0))) | |
937 | ||
938 | (let* ((entry (car num-ann)) | |
939 | (increment (car entry)) | |
940 | (n (ceiling (/ (float (- new old)) (float increment)))) | |
941 | (anno (car (cdr entry)))) | |
942 | (if (> n 0) | |
943 | (cons nil (make-list n anno)) | |
944 | (cons (make-list (- n) anno) nil)))) | |
e5a60108 RS |
945 | |
946 | ;; Standard annotation | |
947 | (let ((close (and old (cdr (assoc old prop-alist)))) | |
029894b9 BG |
948 | (open (and new (cdr (assoc new prop-alist))))) |
949 | (if (or close open) | |
950 | (format-make-relatively-unique close open) | |
951 | ;; Call "Default" function, if any | |
952 | (let ((default (assq nil prop-alist))) | |
953 | (if default | |
954 | (funcall (car (cdr default)) old new)))))))) | |
955 | ||
c84cf2bf | 956 | (provide 'format) |
029894b9 | 957 | ;; format.el ends here |