Commit | Line | Data |
---|---|---|
1cd7adc6 | 1 | ;;; xml.el --- XML parser |
47db06aa | 2 | |
a98e819b | 3 | ;; Copyright (C) 2000, 2001, 2003 Free Software Foundation, Inc. |
47db06aa GM |
4 | |
5 | ;; Author: Emmanuel Briot <briot@gnat.com> | |
720058f2 | 6 | ;; Maintainer: Mark A. Hershberger <mah@everybody.org> |
a98e819b | 7 | ;; Keywords: xml, data |
47db06aa GM |
8 | |
9 | ;; This file is part of GNU Emacs. | |
10 | ||
11 | ;; GNU Emacs is free software; you can redistribute it and/or modify | |
12 | ;; it under the terms of the GNU General Public License as published by | |
13 | ;; the Free Software Foundation; either version 2, or (at your option) | |
14 | ;; any later version. | |
15 | ||
16 | ;; GNU Emacs is distributed in the hope that it will be useful, | |
17 | ;; but WITHOUT ANY WARRANTY; without even the implied warranty of | |
18 | ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
19 | ;; GNU General Public License for more details. | |
20 | ||
21 | ;; You should have received a copy of the GNU General Public License | |
22 | ;; along with GNU Emacs; see the file COPYING. If not, write to the | |
23 | ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
24 | ;; Boston, MA 02111-1307, USA. | |
25 | ||
26 | ;;; Commentary: | |
27 | ||
a98e819b DL |
28 | ;; This file contains a somewhat incomplete non-validating XML parser. It |
29 | ;; parses a file, and returns a list that can be used internally by | |
30 | ;; any other lisp libraries. | |
47db06aa GM |
31 | |
32 | ;;; FILE FORMAT | |
33 | ||
a98e819b DL |
34 | ;; The document type declaration may either be ignored or (optionally) |
35 | ;; parsed, but currently the parsing will only accept element | |
36 | ;; declarations. The XML file is assumed to be well-formed. In case | |
37 | ;; of error, the parsing stops and the XML file is shown where the | |
38 | ;; parsing stopped. | |
47db06aa | 39 | ;; |
a98e819b | 40 | ;; It also knows how to ignore comments and processing instructions. |
47db06aa GM |
41 | ;; |
42 | ;; The XML file should have the following format: | |
653558a1 GM |
43 | ;; <node1 attr1="name1" attr2="name2" ...>value |
44 | ;; <node2 attr3="name3" attr4="name4">value2</node2> | |
45 | ;; <node3 attr5="name5" attr6="name6">value3</node3> | |
47db06aa GM |
46 | ;; </node1> |
47 | ;; Of course, the name of the nodes and attributes can be anything. There can | |
48 | ;; be any number of attributes (or none), as well as any number of children | |
49 | ;; below the nodes. | |
50 | ;; | |
51 | ;; There can be only top level node, but with any number of children below. | |
52 | ||
53 | ;;; LIST FORMAT | |
54 | ||
55 | ;; The functions `xml-parse-file' and `xml-parse-tag' return a list with | |
56 | ;; the following format: | |
57 | ;; | |
58 | ;; xml-list ::= (node node ...) | |
59 | ;; node ::= (tag_name attribute-list . child_node_list) | |
60 | ;; child_node_list ::= child_node child_node ... | |
61 | ;; child_node ::= node | string | |
62 | ;; tag_name ::= string | |
63 | ;; attribute_list ::= (("attribute" . "value") ("attribute" . "value") ...) | |
64 | ;; | nil | |
65 | ;; string ::= "..." | |
66 | ;; | |
a98e819b DL |
67 | ;; Some macros are provided to ease the parsing of this list. |
68 | ;; Whitespace is preserved. Fixme: There should be a tree-walker that | |
69 | ;; can remove it. | |
47db06aa GM |
70 | |
71 | ;;; Code: | |
72 | ||
a98e819b DL |
73 | ;; Note that {buffer-substring,match-string}-no-properties were |
74 | ;; formerly used in several places, but that removes composition info. | |
75 | ||
47db06aa GM |
76 | ;;******************************************************************* |
77 | ;;** | |
78 | ;;** Macros to parse the list | |
79 | ;;** | |
80 | ;;******************************************************************* | |
81 | ||
971489ea | 82 | (defsubst xml-node-name (node) |
47db06aa GM |
83 | "Return the tag associated with NODE. |
84 | The tag is a lower-case symbol." | |
971489ea | 85 | (car node)) |
47db06aa | 86 | |
971489ea | 87 | (defsubst xml-node-attributes (node) |
47db06aa GM |
88 | "Return the list of attributes of NODE. |
89 | The list can be nil." | |
971489ea | 90 | (nth 1 node)) |
47db06aa | 91 | |
971489ea | 92 | (defsubst xml-node-children (node) |
47db06aa GM |
93 | "Return the list of children of NODE. |
94 | This is a list of nodes, and it can be nil." | |
971489ea | 95 | (cddr node)) |
47db06aa GM |
96 | |
97 | (defun xml-get-children (node child-name) | |
98 | "Return the children of NODE whose tag is CHILD-NAME. | |
99 | CHILD-NAME should be a lower case symbol." | |
971489ea SM |
100 | (let ((match ())) |
101 | (dolist (child (xml-node-children node)) | |
102 | (if child | |
103 | (if (equal (xml-node-name child) child-name) | |
104 | (push child match)))) | |
105 | (nreverse match))) | |
47db06aa | 106 | |
9bcd6a7e | 107 | (defun xml-get-attribute-or-nil (node attribute) |
47db06aa | 108 | "Get from NODE the value of ATTRIBUTE. |
65f3f600 | 109 | Return `nil' if the attribute was not found. |
9bcd6a7e EZ |
110 | |
111 | See also `xml-get-attribute'." | |
5ebe0443 EZ |
112 | (when (xml-node-attributes node) |
113 | (let ((value (assoc attribute (xml-node-attributes node)))) | |
114 | (when value | |
115 | (cdr value))))) | |
9bcd6a7e EZ |
116 | |
117 | (defsubst xml-get-attribute (node attribute) | |
118 | "Get from NODE the value of ATTRIBUTE. | |
119 | An empty string is returned if the attribute was not found. | |
120 | ||
121 | See also `xml-get-attribute-or-nil'." | |
122 | (or (xml-get-attribute-or-nil node attribute) "")) | |
47db06aa GM |
123 | |
124 | ;;******************************************************************* | |
125 | ;;** | |
126 | ;;** Creating the list | |
127 | ;;** | |
128 | ;;******************************************************************* | |
129 | ||
a98e819b | 130 | ;;;###autoload |
2d42509a | 131 | (defun xml-parse-file (file &optional parse-dtd parse-ns) |
a98e819b DL |
132 | "Parse the well-formed XML file FILE. |
133 | If FILE is already visited, use its buffer and don't kill it. | |
47db06aa | 134 | Returns the top node with all its children. |
2d42509a JB |
135 | If PARSE-DTD is non-nil, the DTD is parsed rather than skipped. |
136 | If PARSE-NS is non-nil, then QNAMES are expanded." | |
653558a1 GM |
137 | (let ((keep)) |
138 | (if (get-file-buffer file) | |
139 | (progn | |
140 | (set-buffer (get-file-buffer file)) | |
141 | (setq keep (point))) | |
a98e819b DL |
142 | (let (auto-mode-alist) ; no need for xml-mode |
143 | (find-file file))) | |
524425ae | 144 | |
653558a1 GM |
145 | (let ((xml (xml-parse-region (point-min) |
146 | (point-max) | |
147 | (current-buffer) | |
2d42509a | 148 | parse-dtd parse-ns))) |
653558a1 GM |
149 | (if keep |
150 | (goto-char keep) | |
151 | (kill-buffer (current-buffer))) | |
152 | xml))) | |
47db06aa | 153 | |
a98e819b DL |
154 | ;; Note that this is setup so that we can do whitespace-skipping with |
155 | ;; `(skip-syntax-forward " ")', inter alia. Previously this was slow | |
156 | ;; compared with `re-search-forward', but that has been fixed. Also | |
157 | ;; note that the standard syntax table contains other characters with | |
158 | ;; whitespace syntax, like NBSP, but they are invalid in contexts in | |
159 | ;; which we might skip whitespace -- specifically, they're not | |
160 | ;; NameChars [XML 4]. | |
161 | ||
162 | (defvar xml-syntax-table | |
163 | (let ((table (make-syntax-table))) | |
164 | ;; Get space syntax correct per XML [3]. | |
165 | (dotimes (c 31) | |
166 | (modify-syntax-entry c "." table)) ; all are space in standard table | |
167 | (dolist (c '(?\t ?\n ?\r)) ; these should be space | |
168 | (modify-syntax-entry c " " table)) | |
169 | ;; For skipping attributes. | |
170 | (modify-syntax-entry ?\" "\"" table) | |
171 | (modify-syntax-entry ?' "\"" table) | |
172 | ;; Non-alnum name chars should be symbol constituents (`-' and `_' | |
173 | ;; are OK by default). | |
174 | (modify-syntax-entry ?. "_" table) | |
175 | (modify-syntax-entry ?: "_" table) | |
176 | ;; XML [89] | |
177 | (dolist (c '(#x00B7 #x02D0 #x02D1 #x0387 #x0640 #x0E46 #x0EC6 #x3005 | |
178 | #x3031 #x3032 #x3033 #x3034 #x3035 #x309D #x309E #x30FC | |
179 | #x30FD #x30FE)) | |
180 | (modify-syntax-entry (decode-char 'ucs c) "w" table)) | |
181 | ;; Fixme: rest of [4] | |
182 | table) | |
183 | "Syntax table used by `xml-parse-region'.") | |
184 | ||
185 | ;; XML [5] | |
186 | ;; Note that [:alpha:] matches all multibyte chars with word syntax. | |
ab161457 JPW |
187 | (eval-and-compile |
188 | (defconst xml-name-regexp "[[:alpha:]_:][[:alnum:]._:-]*")) | |
a98e819b DL |
189 | |
190 | ;; Fixme: This needs re-writing to deal with the XML grammar properly, i.e. | |
191 | ;; document ::= prolog element Misc* | |
192 | ;; prolog ::= XMLDecl? Misc* (doctypedecl Misc*)? | |
193 | ||
194 | ;;;###autoload | |
2d42509a | 195 | (defun xml-parse-region (beg end &optional buffer parse-dtd parse-ns) |
47db06aa GM |
196 | "Parse the region from BEG to END in BUFFER. |
197 | If BUFFER is nil, it defaults to the current buffer. | |
198 | Returns the XML list for the region, or raises an error if the region | |
2d42509a | 199 | is not well-formed XML. |
47db06aa | 200 | If PARSE-DTD is non-nil, the DTD is parsed rather than skipped, |
2d42509a JB |
201 | and returned as the first element of the list. |
202 | If PARSE-NS is non-nil, then QNAMES are expanded." | |
a98e819b DL |
203 | (save-restriction |
204 | (narrow-to-region beg end) | |
205 | ;; Use fixed syntax table to ensure regexp char classes and syntax | |
206 | ;; specs DTRT. | |
207 | (with-syntax-table (standard-syntax-table) | |
208 | (let ((case-fold-search nil) ; XML is case-sensitive. | |
209 | xml result dtd) | |
210 | (save-excursion | |
211 | (if buffer | |
212 | (set-buffer buffer)) | |
213 | (goto-char (point-min)) | |
214 | (while (not (eobp)) | |
215 | (if (search-forward "<" nil t) | |
216 | (progn | |
217 | (forward-char -1) | |
34638996 EZ |
218 | (setq result (xml-parse-tag parse-dtd parse-ns)) |
219 | (if (and xml result) | |
a98e819b DL |
220 | ;; translation of rule [1] of XML specifications |
221 | (error "XML files can have only one toplevel tag") | |
47db06aa | 222 | (cond |
971489ea | 223 | ((null result)) |
34638996 EZ |
224 | ((and (listp (car result)) |
225 | parse-dtd) | |
971489ea | 226 | (setq dtd (car result)) |
a98e819b DL |
227 | (if (cdr result) ; possible leading comment |
228 | (add-to-list 'xml (cdr result)))) | |
47db06aa | 229 | (t |
a98e819b DL |
230 | (add-to-list 'xml result))))) |
231 | (goto-char (point-max)))) | |
232 | (if parse-dtd | |
233 | (cons dtd (nreverse xml)) | |
234 | (nreverse xml))))))) | |
47db06aa | 235 | |
34638996 EZ |
236 | (defun xml-ns-parse-ns-attrs (attr-list &optional xml-ns) |
237 | "Parse the namespace attributes and return a list of cons in the form: | |
238 | \(namespace . prefix)" | |
239 | ||
240 | (mapcar | |
241 | (lambda (attr) | |
242 | (let* ((splitup (split-string (car attr) ":")) | |
243 | (prefix (nth 0 splitup)) | |
244 | (lname (nth 1 splitup))) | |
245 | (when (string= "xmlns" prefix) | |
246 | (push (cons (if lname | |
247 | lname | |
248 | "") | |
249 | (cdr attr)) | |
250 | xml-ns)))) attr-list) | |
251 | xml-ns) | |
252 | ||
253 | ;; expand element names | |
254 | (defun xml-ns-expand-el (el xml-ns) | |
255 | "Expand the XML elements from \"prefix:local-name\" to a cons in the form | |
256 | \"(namespace . local-name)\"." | |
257 | ||
258 | (let* ((splitup (split-string el ":")) | |
259 | (lname (or (nth 1 splitup) | |
260 | (nth 0 splitup))) | |
261 | (prefix (if (nth 1 splitup) | |
262 | (nth 0 splitup) | |
263 | (if (string= lname "xmlns") | |
264 | "xmlns" | |
265 | ""))) | |
266 | (ns (cdr (assoc-string prefix xml-ns)))) | |
267 | (if (string= "" ns) | |
268 | lname | |
269 | (cons (intern (concat ":" ns)) | |
270 | lname)))) | |
271 | ||
272 | ;; expand attribute names | |
273 | (defun xml-ns-expand-attr (attr-list xml-ns) | |
274 | "Expand the attribute list for a particular element from the form | |
275 | \"prefix:local-name\" to the form \"{namespace}:local-name\"." | |
276 | ||
277 | (mapcar | |
278 | (lambda (attr) | |
279 | (let* ((splitup (split-string (car attr) ":")) | |
280 | (lname (or (nth 1 splitup) | |
281 | (nth 0 splitup))) | |
282 | (prefix (if (nth 1 splitup) | |
283 | (nth 0 splitup) | |
284 | (if (string= (car attr) "xmlns") | |
285 | "xmlns" | |
286 | ""))) | |
287 | (ns (cdr (assoc-string prefix xml-ns)))) | |
288 | (setcar attr | |
289 | (if (string= "" ns) | |
290 | lname | |
291 | (cons (intern (concat ":" ns)) | |
292 | lname))))) | |
293 | attr-list) | |
294 | attr-list) | |
295 | ||
34638996 EZ |
296 | (defun xml-intern-attrlist (attr-list) |
297 | "Convert attribute names to symbols for backward compatibility." | |
298 | (mapcar (lambda (attr) | |
299 | (setcar attr (intern (car attr)))) | |
300 | attr-list) | |
301 | attr-list) | |
47db06aa | 302 | |
2d42509a | 303 | (defun xml-parse-tag (&optional parse-dtd parse-ns) |
a98e819b | 304 | "Parse the tag at point. |
47db06aa GM |
305 | If PARSE-DTD is non-nil, the DTD of the document, if any, is parsed and |
306 | returned as the first element in the list. | |
2d42509a | 307 | If PARSE-NS is non-nil, then QNAMES are expanded. |
47db06aa | 308 | Returns one of: |
a98e819b DL |
309 | - a list : the matching node |
310 | - nil : the point is not looking at a tag. | |
311 | - a pair : the first element is the DTD, the second is the node." | |
2d42509a JB |
312 | (let ((xml-ns (if (consp parse-ns) |
313 | parse-ns | |
314 | (if parse-ns | |
315 | (list | |
316 | ;; Default no namespace | |
317 | (cons "" "") | |
318 | ;; We need to seed the xmlns namespace | |
319 | (cons "xmlns" "http://www.w3.org/2000/xmlns/")))))) | |
320 | (cond | |
321 | ;; Processing instructions (like the <?xml version="1.0"?> tag at the | |
322 | ;; beginning of a document). | |
323 | ((looking-at "<\\?") | |
324 | (search-forward "?>") | |
325 | (skip-syntax-forward " ") | |
326 | (xml-parse-tag parse-dtd xml-ns)) | |
327 | ;; Character data (CDATA) sections, in which no tag should be interpreted | |
328 | ((looking-at "<!\\[CDATA\\[") | |
329 | (let ((pos (match-end 0))) | |
330 | (unless (search-forward "]]>" nil t) | |
331 | (error "CDATA section does not end anywhere in the document")) | |
332 | (buffer-substring pos (match-beginning 0)))) | |
333 | ;; DTD for the document | |
334 | ((looking-at "<!DOCTYPE") | |
335 | (let (dtd) | |
336 | (if parse-dtd | |
337 | (setq dtd (xml-parse-dtd)) | |
338 | (xml-skip-dtd)) | |
a98e819b | 339 | (skip-syntax-forward " ") |
47db06aa | 340 | (if dtd |
2d42509a JB |
341 | (cons dtd (xml-parse-tag nil xml-ns)) |
342 | (xml-parse-tag nil xml-ns)))) | |
343 | ;; skip comments | |
344 | ((looking-at "<!--") | |
345 | (search-forward "-->") | |
346 | nil) | |
347 | ;; end tag | |
348 | ((looking-at "</") | |
349 | '()) | |
350 | ;; opening tag | |
351 | ((looking-at "<\\([^/>[:space:]]+\\)") | |
352 | (goto-char (match-end 1)) | |
34638996 EZ |
353 | |
354 | ;; Parse this node | |
2d42509a | 355 | (let* ((node-name (match-string 1)) |
34638996 EZ |
356 | (attr-list (xml-parse-attlist)) |
357 | (children (if (consp xml-ns) ;; take care of namespace parsing | |
5ebe0443 | 358 | (progn |
34638996 EZ |
359 | (setq xml-ns (xml-ns-parse-ns-attrs |
360 | attr-list xml-ns)) | |
5ebe0443 | 361 | (list (xml-ns-expand-attr |
34638996 | 362 | attr-list xml-ns) |
5ebe0443 | 363 | (xml-ns-expand-el |
34638996 EZ |
364 | node-name xml-ns))) |
365 | (list (xml-intern-attrlist attr-list) | |
366 | (intern node-name)))) | |
2d42509a JB |
367 | pos) |
368 | ||
2d42509a JB |
369 | ;; is this an empty element ? |
370 | (if (looking-at "/>") | |
371 | (progn | |
372 | (forward-char 2) | |
373 | (nreverse children)) | |
47db06aa GM |
374 | |
375 | ;; is this a valid start tag ? | |
e54030af | 376 | (if (eq (char-after) ?>) |
47db06aa GM |
377 | (progn |
378 | (forward-char 1) | |
971489ea SM |
379 | ;; Now check that we have the right end-tag. Note that this |
380 | ;; one might contain spaces after the tag name | |
a98e819b DL |
381 | (let ((end (concat "</" node-name "\\s-*>"))) |
382 | (while (not (looking-at end)) | |
383 | (cond | |
384 | ((looking-at "</") | |
385 | (error "XML: Invalid end tag (expecting %s) at pos %d" | |
386 | node-name (point))) | |
387 | ((= (char-after) ?<) | |
2d42509a | 388 | (let ((tag (xml-parse-tag nil xml-ns))) |
a98e819b DL |
389 | (when tag |
390 | (push tag children)))) | |
391 | (t | |
392 | (setq pos (point)) | |
393 | (search-forward "<") | |
394 | (forward-char -1) | |
395 | (let ((string (buffer-substring pos (point))) | |
396 | (pos 0)) | |
397 | ||
398 | ;; Clean up the string. As per XML | |
399 | ;; specifications, the XML processor should | |
400 | ;; always pass the whole string to the | |
401 | ;; application. But \r's should be replaced: | |
402 | ;; http://www.w3.org/TR/2000/REC-xml-20001006#sec-line-ends | |
403 | (while (string-match "\r\n?" string pos) | |
404 | (setq string (replace-match "\n" t t string)) | |
405 | (setq pos (1+ (match-beginning 0)))) | |
406 | ||
407 | (setq string (xml-substitute-special string)) | |
408 | (setq children | |
409 | (if (stringp (car children)) | |
410 | ;; The two strings were separated by a comment. | |
411 | (cons (concat (car children) string) | |
412 | (cdr children)) | |
413 | (cons string children)))))))) | |
414 | ||
47db06aa | 415 | (goto-char (match-end 0)) |
971489ea | 416 | (nreverse children)) |
47db06aa | 417 | ;; This was an invalid start tag |
a98e819b | 418 | (error "XML: Invalid attribute list"))))) |
2d42509a JB |
419 | (t ;; This is not a tag. |
420 | (error "XML: Invalid character"))))) | |
47db06aa | 421 | |
a98e819b | 422 | (defun xml-parse-attlist () |
34638996 EZ |
423 | "Return the attribute-list after point. Leave point at the |
424 | first non-blank character after the tag." | |
971489ea | 425 | (let ((attlist ()) |
34638996 | 426 | end-pos name) |
a98e819b DL |
427 | (skip-syntax-forward " ") |
428 | (while (looking-at (eval-when-compile | |
429 | (concat "\\(" xml-name-regexp "\\)\\s-*=\\s-*"))) | |
34638996 | 430 | (setq name (match-string 1)) |
47db06aa GM |
431 | (goto-char (match-end 0)) |
432 | ||
a158ff81 JB |
433 | ;; See also: http://www.w3.org/TR/2000/REC-xml-20001006#AVNormalize |
434 | ||
47db06aa GM |
435 | ;; Do we have a string between quotes (or double-quotes), |
436 | ;; or a simple word ? | |
a158ff81 | 437 | (if (looking-at "\"\\([^\"]*\\)\"") |
34638996 | 438 | (setq end-pos (match-end 0)) |
f0ec1711 | 439 | (if (looking-at "'\\([^']*\\)'") |
34638996 | 440 | (setq end-pos (match-end 0)) |
1cd7adc6 | 441 | (error "XML: Attribute values must be given between quotes"))) |
47db06aa GM |
442 | |
443 | ;; Each attribute must be unique within a given element | |
444 | (if (assoc name attlist) | |
1cd7adc6 | 445 | (error "XML: each attribute must be unique within an element")) |
524425ae | 446 | |
a158ff81 JB |
447 | ;; Multiple whitespace characters should be replaced with a single one |
448 | ;; in the attributes | |
a98e819b | 449 | (let ((string (match-string 1)) |
a158ff81 | 450 | (pos 0)) |
a98e819b | 451 | (replace-regexp-in-string "\\s-\\{2,\\}" " " string) |
a158ff81 JB |
452 | (push (cons name (xml-substitute-special string)) attlist)) |
453 | ||
34638996 | 454 | (goto-char end-pos) |
a98e819b | 455 | (skip-syntax-forward " ")) |
971489ea | 456 | (nreverse attlist))) |
47db06aa GM |
457 | |
458 | ;;******************************************************************* | |
459 | ;;** | |
460 | ;;** The DTD (document type declaration) | |
461 | ;;** The following functions know how to skip or parse the DTD of | |
462 | ;;** a document | |
463 | ;;** | |
464 | ;;******************************************************************* | |
465 | ||
a98e819b DL |
466 | ;; Fixme: This fails at least if the DTD contains conditional sections. |
467 | ||
468 | (defun xml-skip-dtd () | |
469 | "Skip the DTD at point. | |
47db06aa GM |
470 | This follows the rule [28] in the XML specifications." |
471 | (forward-char (length "<!DOCTYPE")) | |
a98e819b | 472 | (if (looking-at "\\s-*>") |
47db06aa GM |
473 | (error "XML: invalid DTD (excepting name of the document)")) |
474 | (condition-case nil | |
475 | (progn | |
a98e819b DL |
476 | (forward-sexp) |
477 | (skip-syntax-forward " ") | |
47db06aa | 478 | (if (looking-at "\\[") |
a98e819b DL |
479 | (re-search-forward "]\\s-*>") |
480 | (search-forward ">"))) | |
47db06aa GM |
481 | (error (error "XML: No end to the DTD")))) |
482 | ||
a98e819b DL |
483 | (defun xml-parse-dtd () |
484 | "Parse the DTD at point." | |
485 | (forward-char (eval-when-compile (length "<!DOCTYPE"))) | |
486 | (skip-syntax-forward " ") | |
971489ea SM |
487 | (if (looking-at ">") |
488 | (error "XML: invalid DTD (excepting name of the document)")) | |
524425ae | 489 | |
971489ea | 490 | ;; Get the name of the document |
a98e819b DL |
491 | (looking-at xml-name-regexp) |
492 | (let ((dtd (list (match-string 0) 'dtd)) | |
971489ea | 493 | type element end-pos) |
47db06aa GM |
494 | (goto-char (match-end 0)) |
495 | ||
a98e819b DL |
496 | (skip-syntax-forward " ") |
497 | ;; XML [75] | |
498 | (cond ((looking-at "PUBLIC\\s-+") | |
499 | (goto-char (match-end 0)) | |
500 | (unless (or (re-search-forward | |
501 | "\\=\"\\([[:space:][:alnum:]-'()+,./:=?;!*#@$_%]*\\)\"" | |
502 | nil t) | |
503 | (re-search-forward | |
504 | "\\='\\([[:space:][:alnum:]-()+,./:=?;!*#@$_%]*\\)'" | |
505 | nil t)) | |
506 | (error "XML: missing public id")) | |
507 | (let ((pubid (match-string 1))) | |
508 | (unless (or (re-search-forward "\\='\\([^']*\\)'" nil t) | |
509 | (re-search-forward "\\=\"\\([^\"]*\\)\"" nil t)) | |
510 | (error "XML: missing system id")) | |
511 | (push (list pubid (match-string 1) 'public) dtd))) | |
512 | ((looking-at "SYSTEM\\s-+") | |
513 | (goto-char (match-end 0)) | |
514 | (unless (or (re-search-forward "\\='\\([^']*\\)'" nil t) | |
515 | (re-search-forward "\\=\"\\([^\"]*\\)\"" nil t)) | |
516 | (error "XML: missing system id")) | |
517 | (push (list (match-string 1) 'system) dtd))) | |
518 | (skip-syntax-forward " ") | |
519 | (if (eq ?> (char-after)) | |
520 | (forward-char) | |
521 | (skip-syntax-forward " ") | |
522 | (if (not (eq (char-after) ?\[)) | |
523 | (error "XML: bad DTD") | |
524 | (forward-char) | |
525 | ;; Parse the rest of the DTD | |
526 | ;; Fixme: Deal with ENTITY, ATTLIST, NOTATION, PIs. | |
527 | (while (not (looking-at "\\s-*\\]")) | |
528 | (skip-syntax-forward " ") | |
529 | (cond | |
530 | ||
531 | ;; Translation of rule [45] of XML specifications | |
532 | ((looking-at | |
533 | "<!ELEMENT\\s-+\\([[:alnum:].%;]+\\)\\s-+\\([^>]+\\)>") | |
534 | ||
34638996 | 535 | (setq element (match-string 1) |
a98e819b DL |
536 | type (match-string-no-properties 2)) |
537 | (setq end-pos (match-end 0)) | |
538 | ||
539 | ;; Translation of rule [46] of XML specifications | |
540 | (cond | |
541 | ((string-match "^EMPTY[ \t\n\r]*$" type) ;; empty declaration | |
542 | (setq type 'empty)) | |
543 | ((string-match "^ANY[ \t\n\r]*$" type) ;; any type of contents | |
544 | (setq type 'any)) | |
545 | ((string-match "^(\\(.*\\))[ \t\n\r]*$" type) ;; children ([47]) | |
546 | (setq type (xml-parse-elem-type (match-string 1 type)))) | |
547 | ((string-match "^%[^;]+;[ \t\n\r]*$" type) ;; substitution | |
548 | nil) | |
549 | (t | |
550 | (error "XML: Invalid element type in the DTD"))) | |
551 | ||
552 | ;; rule [45]: the element declaration must be unique | |
553 | (if (assoc element dtd) | |
554 | (error "XML: element declarations must be unique in a DTD (<%s>)" | |
461f3ad0 | 555 | element)) |
a98e819b DL |
556 | |
557 | ;; Store the element in the DTD | |
558 | (push (list element type) dtd) | |
559 | (goto-char end-pos)) | |
560 | ((looking-at "<!--") | |
561 | (search-forward "-->")) | |
562 | ||
563 | (t | |
564 | (error "XML: Invalid DTD item"))) | |
565 | ||
566 | ;; Skip the end of the DTD | |
567 | (search-forward ">")))) | |
461f3ad0 | 568 | (nreverse dtd))) |
47db06aa GM |
569 | |
570 | (defun xml-parse-elem-type (string) | |
a98e819b | 571 | "Convert element type STRING into a Lisp structure." |
47db06aa GM |
572 | |
573 | (let (elem modifier) | |
574 | (if (string-match "(\\([^)]+\\))\\([+*?]?\\)" string) | |
575 | (progn | |
576 | (setq elem (match-string 1 string) | |
577 | modifier (match-string 2 string)) | |
578 | (if (string-match "|" elem) | |
971489ea | 579 | (setq elem (cons 'choice |
47db06aa GM |
580 | (mapcar 'xml-parse-elem-type |
581 | (split-string elem "|")))) | |
582 | (if (string-match "," elem) | |
971489ea | 583 | (setq elem (cons 'seq |
47db06aa | 584 | (mapcar 'xml-parse-elem-type |
a98e819b | 585 | (split-string elem ","))))))) |
a158ff81 JB |
586 | (if (string-match "[ \t\n\r]*\\([^+*?]+\\)\\([+*?]?\\)" string) |
587 | (setq elem (match-string 1 string) | |
47db06aa GM |
588 | modifier (match-string 2 string)))) |
589 | ||
971489ea SM |
590 | (if (and (stringp elem) (string= elem "#PCDATA")) |
591 | (setq elem 'pcdata)) | |
524425ae | 592 | |
971489ea SM |
593 | (cond |
594 | ((string= modifier "+") | |
595 | (list '+ elem)) | |
596 | ((string= modifier "*") | |
597 | (list '* elem)) | |
598 | ((string= modifier "?") | |
0fa6f70c | 599 | (list '\? elem)) |
971489ea SM |
600 | (t |
601 | elem)))) | |
47db06aa | 602 | |
47db06aa GM |
603 | ;;******************************************************************* |
604 | ;;** | |
605 | ;;** Substituting special XML sequences | |
606 | ;;** | |
607 | ;;******************************************************************* | |
608 | ||
a98e819b DL |
609 | (eval-when-compile |
610 | (defvar str)) ; dynamic from replace-regexp-in-string | |
611 | ||
612 | ;; Fixme: Take declared entities from the DTD when they're available. | |
613 | (defun xml-substitute-entity (match) | |
614 | "Subroutine of xml-substitute-special." | |
615 | (save-match-data | |
616 | (let ((match1 (match-string 1 str))) | |
617 | (cond ((string= match1 "lt") "<") | |
618 | ((string= match1 "gt") ">") | |
619 | ((string= match1 "apos") "'") | |
620 | ((string= match1 "quot") "\"") | |
621 | ((string= match1 "amp") "&") | |
622 | ((and (string-match "#\\([0-9]+\\)" match1) | |
623 | (let ((c (decode-char | |
624 | 'ucs | |
625 | (string-to-number (match-string 1 match1))))) | |
626 | (if c (string c))))) ; else unrepresentable | |
627 | ((and (string-match "#x\\([[:xdigit:]]+\\)" match1) | |
628 | (let ((c (decode-char | |
629 | 'ucs | |
630 | (string-to-number (match-string 1 match1) 16)))) | |
631 | (if c (string c))))) | |
632 | ;; Default to asis. Arguably, unrepresentable code points | |
633 | ;; might be best replaced with U+FFFD. | |
634 | (t match))))) | |
635 | ||
47db06aa | 636 | (defun xml-substitute-special (string) |
a98e819b DL |
637 | "Return STRING, after subsituting entity references." |
638 | ;; This originally made repeated passes through the string from the | |
639 | ;; beginning, which isn't correct, since then either "&amp;" or | |
640 | ;; "&amp;" won't DTRT. | |
641 | (replace-regexp-in-string "&\\([^;]+\\);" | |
642 | #'xml-substitute-entity string t t)) | |
47db06aa GM |
643 | |
644 | ;;******************************************************************* | |
645 | ;;** | |
646 | ;;** Printing a tree. | |
647 | ;;** This function is intended mainly for debugging purposes. | |
648 | ;;** | |
649 | ;;******************************************************************* | |
650 | ||
651 | (defun xml-debug-print (xml) | |
971489ea SM |
652 | (dolist (node xml) |
653 | (xml-debug-print-internal node ""))) | |
47db06aa | 654 | |
971489ea | 655 | (defun xml-debug-print-internal (xml indent-string) |
47db06aa | 656 | "Outputs the XML tree in the current buffer. |
a98e819b | 657 | The first line is indented with INDENT-STRING." |
47db06aa GM |
658 | (let ((tree xml) |
659 | attlist) | |
a98e819b | 660 | (insert indent-string ?< (symbol-name (xml-node-name tree))) |
524425ae | 661 | |
47db06aa | 662 | ;; output the attribute list |
971489ea | 663 | (setq attlist (xml-node-attributes tree)) |
47db06aa | 664 | (while attlist |
a98e819b | 665 | (insert ?\ (symbol-name (caar attlist)) "=\"" (cdar attlist) ?\") |
971489ea | 666 | (setq attlist (cdr attlist))) |
524425ae | 667 | |
a98e819b | 668 | (insert ?>) |
524425ae | 669 | |
971489ea | 670 | (setq tree (xml-node-children tree)) |
47db06aa GM |
671 | |
672 | ;; output the children | |
971489ea | 673 | (dolist (node tree) |
47db06aa | 674 | (cond |
971489ea | 675 | ((listp node) |
a98e819b | 676 | (insert ?\n) |
971489ea SM |
677 | (xml-debug-print-internal node (concat indent-string " "))) |
678 | ((stringp node) (insert node)) | |
47db06aa | 679 | (t |
971489ea | 680 | (error "Invalid XML tree")))) |
47db06aa | 681 | |
a98e819b DL |
682 | (insert ?\n indent-string |
683 | ?< ?/ (symbol-name (xml-node-name xml)) ?>))) | |
47db06aa GM |
684 | |
685 | (provide 'xml) | |
686 | ||
ab5796a9 | 687 | ;;; arch-tag: 5864b283-5a68-4b59-a20d-36a72b353b9b |
47db06aa | 688 | ;;; xml.el ends here |