Commit | Line | Data |
---|---|---|
1cd7adc6 | 1 | ;;; xml.el --- XML parser |
47db06aa | 2 | |
2e9bdf15 | 3 | ;; Copyright (C) 2000, 01, 03, 2004 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 | |
a1dfa9a3 | 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 | |
a1dfa9a3 | 36 | ;; declarations. The XML file is assumed to be well-formed. In case |
a98e819b DL |
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 | 46 | ;; </node1> |
a1dfa9a3 | 47 | ;; Of course, the name of the nodes and attributes can be anything. There can |
47db06aa GM |
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 | ||
c7f8d055 SM |
55 | ;; The functions `xml-parse-file', `xml-parse-region' and |
56 | ;; `xml-parse-tag' return a list with the following format: | |
47db06aa GM |
57 | ;; |
58 | ;; xml-list ::= (node node ...) | |
c7f8d055 | 59 | ;; node ::= (qname attribute-list . child_node_list) |
47db06aa GM |
60 | ;; child_node_list ::= child_node child_node ... |
61 | ;; child_node ::= node | string | |
c7f8d055 SM |
62 | ;; qname ::= (:namespace-uri . "name") | "name" |
63 | ;; attribute_list ::= ((qname . "value") (qname . "value") ...) | |
47db06aa GM |
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 | 70 | |
c7f8d055 SM |
71 | ;; TODO: |
72 | ;; * xml:base, xml:space support | |
73 | ;; * more complete DOCTYPE parsing | |
74 | ;; * pi support | |
75 | ||
47db06aa GM |
76 | ;;; Code: |
77 | ||
a98e819b DL |
78 | ;; Note that {buffer-substring,match-string}-no-properties were |
79 | ;; formerly used in several places, but that removes composition info. | |
80 | ||
47db06aa GM |
81 | ;;******************************************************************* |
82 | ;;** | |
83 | ;;** Macros to parse the list | |
84 | ;;** | |
85 | ;;******************************************************************* | |
86 | ||
6d12a4df MH |
87 | (defvar xml-entity-alist |
88 | '(("lt" . "<") | |
89 | ("gt" . ">") | |
90 | ("apos" . "'") | |
91 | ("quot" . "\"") | |
92 | ("amp" . "&")) | |
93 | "The defined entities. Entities are added to this when the DTD is parsed.") | |
94 | ||
95 | (defvar xml-sub-parser nil | |
96 | "Dynamically set this to a non-nil value if you want to parse an XML fragment.") | |
97 | ||
98 | (defvar xml-validating-parser nil | |
99 | "Set to non-nil to get validity checking.") | |
100 | ||
971489ea | 101 | (defsubst xml-node-name (node) |
47db06aa | 102 | "Return the tag associated with NODE. |
a1dfa9a3 SM |
103 | Without namespace-aware parsing, the tag is a symbol. |
104 | ||
105 | With namespace-aware parsing, the tag is a cons of a string | |
106 | representing the uri of the namespace with the local name of the | |
107 | tag. For example, | |
108 | ||
109 | <foo> | |
110 | ||
111 | would be represented by | |
112 | ||
113 | '(\"\" . \"foo\")." | |
114 | ||
971489ea | 115 | (car node)) |
47db06aa | 116 | |
971489ea | 117 | (defsubst xml-node-attributes (node) |
47db06aa GM |
118 | "Return the list of attributes of NODE. |
119 | The list can be nil." | |
971489ea | 120 | (nth 1 node)) |
47db06aa | 121 | |
971489ea | 122 | (defsubst xml-node-children (node) |
47db06aa GM |
123 | "Return the list of children of NODE. |
124 | This is a list of nodes, and it can be nil." | |
971489ea | 125 | (cddr node)) |
47db06aa GM |
126 | |
127 | (defun xml-get-children (node child-name) | |
128 | "Return the children of NODE whose tag is CHILD-NAME. | |
a1dfa9a3 | 129 | CHILD-NAME should match the value returned by `xml-node-name'." |
971489ea SM |
130 | (let ((match ())) |
131 | (dolist (child (xml-node-children node)) | |
a1dfa9a3 SM |
132 | (if (and (listp child) |
133 | (equal (xml-node-name child) child-name)) | |
134 | (push child match))) | |
971489ea | 135 | (nreverse match))) |
47db06aa | 136 | |
9bcd6a7e | 137 | (defun xml-get-attribute-or-nil (node attribute) |
47db06aa | 138 | "Get from NODE the value of ATTRIBUTE. |
a1dfa9a3 | 139 | Return nil if the attribute was not found. |
9bcd6a7e EZ |
140 | |
141 | See also `xml-get-attribute'." | |
2e9bdf15 | 142 | (cdr (assoc attribute (xml-node-attributes node)))) |
9bcd6a7e EZ |
143 | |
144 | (defsubst xml-get-attribute (node attribute) | |
145 | "Get from NODE the value of ATTRIBUTE. | |
146 | An empty string is returned if the attribute was not found. | |
147 | ||
148 | See also `xml-get-attribute-or-nil'." | |
149 | (or (xml-get-attribute-or-nil node attribute) "")) | |
47db06aa GM |
150 | |
151 | ;;******************************************************************* | |
152 | ;;** | |
153 | ;;** Creating the list | |
154 | ;;** | |
155 | ;;******************************************************************* | |
156 | ||
a98e819b | 157 | ;;;###autoload |
2d42509a | 158 | (defun xml-parse-file (file &optional parse-dtd parse-ns) |
a98e819b DL |
159 | "Parse the well-formed XML file FILE. |
160 | If FILE is already visited, use its buffer and don't kill it. | |
47db06aa | 161 | Returns the top node with all its children. |
2d42509a JB |
162 | If PARSE-DTD is non-nil, the DTD is parsed rather than skipped. |
163 | If PARSE-NS is non-nil, then QNAMES are expanded." | |
653558a1 GM |
164 | (let ((keep)) |
165 | (if (get-file-buffer file) | |
166 | (progn | |
167 | (set-buffer (get-file-buffer file)) | |
168 | (setq keep (point))) | |
a98e819b DL |
169 | (let (auto-mode-alist) ; no need for xml-mode |
170 | (find-file file))) | |
524425ae | 171 | |
653558a1 GM |
172 | (let ((xml (xml-parse-region (point-min) |
173 | (point-max) | |
174 | (current-buffer) | |
2d42509a | 175 | parse-dtd parse-ns))) |
653558a1 GM |
176 | (if keep |
177 | (goto-char keep) | |
178 | (kill-buffer (current-buffer))) | |
179 | xml))) | |
47db06aa | 180 | |
6d12a4df MH |
181 | |
182 | (let* ((start-chars (concat ":[:alpha:]_")) | |
183 | (name-chars (concat "-[:digit:]." start-chars)) | |
184 | ;;[3] S ::= (#x20 | #x9 | #xD | #xA)+ | |
185 | (whitespace "[ \t\n\r]")) | |
186 | ;;[4] NameStartChar ::= ":" | [A-Z] | "_" | [a-z] | [#xC0-#xD6] | |
187 | ;; | [#xD8-#xF6] | [#xF8-#x2FF] | [#x370-#x37D] | [#x37F-#x1FFF] | |
188 | ;; | [#x200C-#x200D] | [#x2070-#x218F] | [#x2C00-#x2FEF] | [#x3001-#xD7FF] | |
189 | ;; | [#xF900-#xFDCF] | [#xFDF0-#xFFFD] | [#x10000-#xEFFFF] | |
190 | (defvar xml-name-start-char-re (concat "[" start-chars "]")) | |
191 | ;;[4a] NameChar ::= NameStartChar | "-" | "." | [0-9] | #xB7 | [#x0300-#x036F] | [#x203F-#x2040] | |
192 | (defvar xml-name-char-re (concat "[" name-chars "]")) | |
193 | ;;[5] Name ::= NameStartChar (NameChar)* | |
194 | (defvar xml-name-re (concat xml-name-start-char-re xml-name-char-re "*")) | |
195 | ;;[6] Names ::= Name (#x20 Name)* | |
196 | (defvar xml-names-re (concat xml-name-re "\\(?: " xml-name-re "\\)*")) | |
197 | ;;[7] Nmtoken ::= (NameChar)+ | |
198 | (defvar xml-nmtoken-re (concat xml-name-char-re "+")) | |
199 | ;;[8] Nmtokens ::= Nmtoken (#x20 Nmtoken)* | |
200 | (defvar xml-nmtokens-re (concat xml-nmtoken-re "\\(?: " xml-name-re "\\)*")) | |
201 | ;;[66] CharRef ::= '&#' [0-9]+ ';' | '&#x' [0-9a-fA-F]+ ';' | |
202 | (defvar xml-char-ref-re "\\(?:&#[0-9]+;\\|&#x[0-9a-fA-F]+;\\)") | |
203 | ;;[68] EntityRef ::= '&' Name ';' | |
204 | (defvar xml-entity-ref (concat "&" xml-name-re ";")) | |
205 | ;;[69] PEReference ::= '%' Name ';' | |
206 | (defvar xml-pe-reference-re (concat "%" xml-name-re ";")) | |
207 | ;;[67] Reference ::= EntityRef | CharRef | |
208 | (defvar xml-reference-re (concat "\\(?:" xml-entity-ref "\\|" xml-char-ref-re "\\)")) | |
209 | ;;[9] EntityValue ::= '"' ([^%&"] | PEReference | Reference)* '"' | |
210 | ;; | "'" ([^%&'] | PEReference | Reference)* "'" | |
211 | (defvar xml-entity-value-re (concat "\\(?:\"\\(?:[^%&\"]\\|" xml-pe-reference-re | |
212 | "\\|" xml-reference-re "\\)*\"\\|'\\(?:[^%&']\\|" | |
213 | xml-pe-reference-re "\\|" xml-reference-re "\\)*'\\)"))) | |
214 | ;;[75] ExternalID ::= 'SYSTEM' S SystemLiteral | |
215 | ;; | 'PUBLIC' S PubidLiteral S SystemLiteral | |
216 | ;;[76] NDataDecl ::= S 'NDATA' S | |
217 | ;;[73] EntityDef ::= EntityValue| (ExternalID NDataDecl?) | |
218 | ;;[71] GEDecl ::= '<!ENTITY' S Name S EntityDef S? '>' | |
219 | ;;[74] PEDef ::= EntityValue | ExternalID | |
220 | ;;[72] PEDecl ::= '<!ENTITY' S '%' S Name S PEDef S? '>' | |
221 | ;;[70] EntityDecl ::= GEDecl | PEDecl | |
222 | ||
a98e819b DL |
223 | ;; Note that this is setup so that we can do whitespace-skipping with |
224 | ;; `(skip-syntax-forward " ")', inter alia. Previously this was slow | |
225 | ;; compared with `re-search-forward', but that has been fixed. Also | |
226 | ;; note that the standard syntax table contains other characters with | |
227 | ;; whitespace syntax, like NBSP, but they are invalid in contexts in | |
228 | ;; which we might skip whitespace -- specifically, they're not | |
229 | ;; NameChars [XML 4]. | |
230 | ||
231 | (defvar xml-syntax-table | |
232 | (let ((table (make-syntax-table))) | |
233 | ;; Get space syntax correct per XML [3]. | |
234 | (dotimes (c 31) | |
235 | (modify-syntax-entry c "." table)) ; all are space in standard table | |
236 | (dolist (c '(?\t ?\n ?\r)) ; these should be space | |
237 | (modify-syntax-entry c " " table)) | |
238 | ;; For skipping attributes. | |
239 | (modify-syntax-entry ?\" "\"" table) | |
240 | (modify-syntax-entry ?' "\"" table) | |
241 | ;; Non-alnum name chars should be symbol constituents (`-' and `_' | |
242 | ;; are OK by default). | |
243 | (modify-syntax-entry ?. "_" table) | |
244 | (modify-syntax-entry ?: "_" table) | |
245 | ;; XML [89] | |
246 | (dolist (c '(#x00B7 #x02D0 #x02D1 #x0387 #x0640 #x0E46 #x0EC6 #x3005 | |
247 | #x3031 #x3032 #x3033 #x3034 #x3035 #x309D #x309E #x30FC | |
248 | #x30FD #x30FE)) | |
249 | (modify-syntax-entry (decode-char 'ucs c) "w" table)) | |
250 | ;; Fixme: rest of [4] | |
251 | table) | |
252 | "Syntax table used by `xml-parse-region'.") | |
253 | ||
254 | ;; XML [5] | |
255 | ;; Note that [:alpha:] matches all multibyte chars with word syntax. | |
ab161457 JPW |
256 | (eval-and-compile |
257 | (defconst xml-name-regexp "[[:alpha:]_:][[:alnum:]._:-]*")) | |
a98e819b DL |
258 | |
259 | ;; Fixme: This needs re-writing to deal with the XML grammar properly, i.e. | |
260 | ;; document ::= prolog element Misc* | |
261 | ;; prolog ::= XMLDecl? Misc* (doctypedecl Misc*)? | |
262 | ||
263 | ;;;###autoload | |
2d42509a | 264 | (defun xml-parse-region (beg end &optional buffer parse-dtd parse-ns) |
47db06aa GM |
265 | "Parse the region from BEG to END in BUFFER. |
266 | If BUFFER is nil, it defaults to the current buffer. | |
267 | Returns the XML list for the region, or raises an error if the region | |
2d42509a | 268 | is not well-formed XML. |
47db06aa | 269 | If PARSE-DTD is non-nil, the DTD is parsed rather than skipped, |
2d42509a JB |
270 | and returned as the first element of the list. |
271 | If PARSE-NS is non-nil, then QNAMES are expanded." | |
a98e819b DL |
272 | (save-restriction |
273 | (narrow-to-region beg end) | |
274 | ;; Use fixed syntax table to ensure regexp char classes and syntax | |
275 | ;; specs DTRT. | |
276 | (with-syntax-table (standard-syntax-table) | |
277 | (let ((case-fold-search nil) ; XML is case-sensitive. | |
278 | xml result dtd) | |
279 | (save-excursion | |
280 | (if buffer | |
281 | (set-buffer buffer)) | |
282 | (goto-char (point-min)) | |
283 | (while (not (eobp)) | |
284 | (if (search-forward "<" nil t) | |
285 | (progn | |
286 | (forward-char -1) | |
34638996 | 287 | (setq result (xml-parse-tag parse-dtd parse-ns)) |
6d12a4df | 288 | (if (and xml result (not xml-sub-parser)) |
a98e819b | 289 | ;; translation of rule [1] of XML specifications |
6d12a4df | 290 | (error "XML: (Not Well-Formed) Only one root tag allowed") |
47db06aa | 291 | (cond |
971489ea | 292 | ((null result)) |
34638996 EZ |
293 | ((and (listp (car result)) |
294 | parse-dtd) | |
971489ea | 295 | (setq dtd (car result)) |
a98e819b DL |
296 | (if (cdr result) ; possible leading comment |
297 | (add-to-list 'xml (cdr result)))) | |
47db06aa | 298 | (t |
a98e819b DL |
299 | (add-to-list 'xml result))))) |
300 | (goto-char (point-max)))) | |
301 | (if parse-dtd | |
302 | (cons dtd (nreverse xml)) | |
303 | (nreverse xml))))))) | |
47db06aa | 304 | |
c7f8d055 | 305 | (defun xml-maybe-do-ns (name default xml-ns) |
a1dfa9a3 SM |
306 | "Perform any namespace expansion. |
307 | NAME is the name to perform the expansion on. | |
c7f8d055 SM |
308 | DEFAULT is the default namespace. XML-NS is a cons of namespace |
309 | names to uris. When namespace-aware parsing is off, then XML-NS | |
310 | is nil. | |
311 | ||
312 | During namespace-aware parsing, any name without a namespace is | |
313 | put into the namespace identified by DEFAULT. nil is used to | |
314 | specify that the name shouldn't be given a namespace." | |
315 | (if (consp xml-ns) | |
316 | (let* ((nsp (string-match ":" name)) | |
317 | (lname (if nsp (substring name (match-end 0)) name)) | |
318 | (prefix (if nsp (substring name 0 (match-beginning 0)) default)) | |
319 | (special (and (string-equal lname "xmlns") (not prefix))) | |
320 | ;; Setting default to nil will insure that there is not | |
321 | ;; matching cons in xml-ns. In which case we | |
322 | (ns (or (cdr (assoc (if special "xmlns" prefix) | |
323 | xml-ns)) | |
6d12a4df | 324 | ""))) |
c7f8d055 SM |
325 | (cons ns (if special "" lname))) |
326 | (intern name))) | |
47db06aa | 327 | |
6d12a4df MH |
328 | (defun xml-parse-fragment (&optional parse-dtd parse-ns) |
329 | "Parse xml-like fragments." | |
330 | (let ((xml-sub-parser t) | |
331 | children) | |
332 | (while (not (eobp)) | |
333 | (let ((bit (xml-parse-tag | |
334 | parse-dtd parse-ns))) | |
335 | (if children | |
336 | (setq children (append (list bit) children)) | |
337 | (if (stringp bit) | |
338 | (setq children (list bit)) | |
339 | (setq children bit))))) | |
340 | (reverse children))) | |
341 | ||
2d42509a | 342 | (defun xml-parse-tag (&optional parse-dtd parse-ns) |
a98e819b | 343 | "Parse the tag at point. |
47db06aa GM |
344 | If PARSE-DTD is non-nil, the DTD of the document, if any, is parsed and |
345 | returned as the first element in the list. | |
2d42509a | 346 | If PARSE-NS is non-nil, then QNAMES are expanded. |
47db06aa | 347 | Returns one of: |
a98e819b DL |
348 | - a list : the matching node |
349 | - nil : the point is not looking at a tag. | |
350 | - a pair : the first element is the DTD, the second is the node." | |
6d12a4df MH |
351 | (let ((xml-validating-parser (or parse-dtd xml-validating-parser)) |
352 | (xml-ns (if (consp parse-ns) | |
2d42509a JB |
353 | parse-ns |
354 | (if parse-ns | |
355 | (list | |
c7f8d055 | 356 | ;; Default for empty prefix is no namespace |
6d12a4df | 357 | (cons "" "") |
c7f8d055 | 358 | ;; "xml" namespace |
6d12a4df | 359 | (cons "xml" "http://www.w3.org/XML/1998/namespace") |
2d42509a | 360 | ;; We need to seed the xmlns namespace |
6d12a4df | 361 | (cons "xmlns" "http://www.w3.org/2000/xmlns/")))))) |
2d42509a JB |
362 | (cond |
363 | ;; Processing instructions (like the <?xml version="1.0"?> tag at the | |
364 | ;; beginning of a document). | |
365 | ((looking-at "<\\?") | |
366 | (search-forward "?>") | |
367 | (skip-syntax-forward " ") | |
368 | (xml-parse-tag parse-dtd xml-ns)) | |
369 | ;; Character data (CDATA) sections, in which no tag should be interpreted | |
370 | ((looking-at "<!\\[CDATA\\[") | |
371 | (let ((pos (match-end 0))) | |
372 | (unless (search-forward "]]>" nil t) | |
6d12a4df | 373 | (error "XML: (Not Well Formed) CDATA section does not end anywhere in the document")) |
2d42509a JB |
374 | (buffer-substring pos (match-beginning 0)))) |
375 | ;; DTD for the document | |
376 | ((looking-at "<!DOCTYPE") | |
6d12a4df MH |
377 | (let ((dtd (xml-parse-dtd parse-ns))) |
378 | (skip-syntax-forward " ") | |
379 | (if xml-validating-parser | |
380 | (cons dtd (xml-parse-tag nil xml-ns)) | |
381 | (xml-parse-tag nil xml-ns)))) | |
2d42509a JB |
382 | ;; skip comments |
383 | ((looking-at "<!--") | |
384 | (search-forward "-->") | |
385 | nil) | |
386 | ;; end tag | |
387 | ((looking-at "</") | |
388 | '()) | |
389 | ;; opening tag | |
390 | ((looking-at "<\\([^/>[:space:]]+\\)") | |
391 | (goto-char (match-end 1)) | |
34638996 EZ |
392 | |
393 | ;; Parse this node | |
2d42509a | 394 | (let* ((node-name (match-string 1)) |
c7f8d055 SM |
395 | ;; Parse the attribute list. |
396 | (attrs (xml-parse-attlist xml-ns)) | |
397 | children pos) | |
398 | ||
399 | ;; add the xmlns:* attrs to our cache | |
400 | (when (consp xml-ns) | |
401 | (dolist (attr attrs) | |
402 | (when (and (consp (car attr)) | |
6d12a4df MH |
403 | (equal "http://www.w3.org/2000/xmlns/" |
404 | (caar attr))) | |
405 | (push (cons (cdar attr) (cdr attr)) | |
c7f8d055 SM |
406 | xml-ns)))) |
407 | ||
43b5fd81 | 408 | (setq children (list attrs (xml-maybe-do-ns node-name "" xml-ns))) |
c7f8d055 | 409 | |
2d42509a JB |
410 | ;; is this an empty element ? |
411 | (if (looking-at "/>") | |
47db06aa | 412 | (progn |
6d12a4df | 413 | (forward-char 2) |
971489ea | 414 | (nreverse children)) |
6d12a4df MH |
415 | |
416 | ;; is this a valid start tag ? | |
417 | (if (eq (char-after) ?>) | |
418 | (progn | |
419 | (forward-char 1) | |
420 | ;; Now check that we have the right end-tag. Note that this | |
421 | ;; one might contain spaces after the tag name | |
422 | (let ((end (concat "</" node-name "\\s-*>"))) | |
423 | (while (not (looking-at end)) | |
424 | (cond | |
425 | ((looking-at "</") | |
426 | (error "XML: (Not Well-Formed) Invalid end tag (expecting %s) at pos %d" | |
427 | node-name (point))) | |
428 | ((= (char-after) ?<) | |
429 | (let ((tag (xml-parse-tag nil xml-ns))) | |
430 | (when tag | |
431 | (push tag children)))) | |
432 | (t | |
433 | (let ((expansion (xml-parse-string))) | |
434 | (setq children | |
435 | (if (stringp expansion) | |
436 | (if (stringp (car children)) | |
437 | ;; The two strings were separated by a comment. | |
438 | (setq children (append (concat (car children) expansion) | |
439 | (cdr children))) | |
440 | (setq children (append (list expansion) children))) | |
441 | (setq children (append expansion children)))))))) | |
442 | ||
443 | (goto-char (match-end 0)) | |
444 | (nreverse children))) | |
445 | ;; This was an invalid start tag (Expected ">", but didn't see it.) | |
446 | (error "XML: (Well-Formed) Couldn't parse tag: %s" | |
447 | (buffer-substring (- (point) 10) (+ (point) 1))))))) | |
448 | (t ;; (Not one of PI, CDATA, Comment, End tag, or Start tag) | |
449 | (unless xml-sub-parser ; Usually, we error out. | |
450 | (error "XML: (Well-Formed) Invalid character")) | |
451 | ||
452 | ;; However, if we're parsing incrementally, then we need to deal | |
453 | ;; with stray CDATA. | |
454 | (xml-parse-string))))) | |
455 | ||
456 | (defun xml-parse-string () | |
457 | "Parse the next whatever. Could be a string, or an element." | |
458 | (let* ((pos (point)) | |
459 | (string (progn (if (search-forward "<" nil t) | |
460 | (forward-char -1) | |
461 | (goto-char (point-max))) | |
462 | (buffer-substring pos (point))))) | |
463 | ;; Clean up the string. As per XML specifications, the XML | |
464 | ;; processor should always pass the whole string to the | |
465 | ;; application. But \r's should be replaced: | |
466 | ;; http://www.w3.org/TR/2000/REC-xml-20001006#sec-line-ends | |
467 | (setq pos 0) | |
468 | (while (string-match "\r\n?" string pos) | |
469 | (setq string (replace-match "\n" t t string)) | |
470 | (setq pos (1+ (match-beginning 0)))) | |
471 | ||
472 | (xml-substitute-special string))) | |
47db06aa | 473 | |
c7f8d055 | 474 | (defun xml-parse-attlist (&optional xml-ns) |
a1dfa9a3 SM |
475 | "Return the attribute-list after point. |
476 | Leave point at the first non-blank character after the tag." | |
971489ea | 477 | (let ((attlist ()) |
34638996 | 478 | end-pos name) |
a98e819b DL |
479 | (skip-syntax-forward " ") |
480 | (while (looking-at (eval-when-compile | |
481 | (concat "\\(" xml-name-regexp "\\)\\s-*=\\s-*"))) | |
c7f8d055 SM |
482 | (setq end-pos (match-end 0)) |
483 | (setq name (xml-maybe-do-ns (match-string 1) nil xml-ns)) | |
484 | (goto-char end-pos) | |
47db06aa | 485 | |
a158ff81 JB |
486 | ;; See also: http://www.w3.org/TR/2000/REC-xml-20001006#AVNormalize |
487 | ||
47db06aa GM |
488 | ;; Do we have a string between quotes (or double-quotes), |
489 | ;; or a simple word ? | |
a158ff81 | 490 | (if (looking-at "\"\\([^\"]*\\)\"") |
34638996 | 491 | (setq end-pos (match-end 0)) |
f0ec1711 | 492 | (if (looking-at "'\\([^']*\\)'") |
34638996 | 493 | (setq end-pos (match-end 0)) |
6d12a4df | 494 | (error "XML: (Not Well-Formed) Attribute values must be given between quotes"))) |
47db06aa GM |
495 | |
496 | ;; Each attribute must be unique within a given element | |
497 | (if (assoc name attlist) | |
6d12a4df | 498 | (error "XML: (Not Well-Formed) Each attribute must be unique within an element")) |
524425ae | 499 | |
a158ff81 JB |
500 | ;; Multiple whitespace characters should be replaced with a single one |
501 | ;; in the attributes | |
a98e819b | 502 | (let ((string (match-string 1)) |
a158ff81 | 503 | (pos 0)) |
a98e819b | 504 | (replace-regexp-in-string "\\s-\\{2,\\}" " " string) |
6d12a4df MH |
505 | (let ((expansion (xml-substitute-special string))) |
506 | (unless (stringp expansion) | |
507 | ; We say this is the constraint. It is acctually that | |
508 | ; external entities nor "<" can be in an attribute value. | |
509 | (error "XML: (Not Well-Formed) Entities in attributes cannot expand into elements")) | |
510 | (push (cons name expansion) attlist))) | |
a158ff81 | 511 | |
34638996 | 512 | (goto-char end-pos) |
a98e819b | 513 | (skip-syntax-forward " ")) |
971489ea | 514 | (nreverse attlist))) |
47db06aa GM |
515 | |
516 | ;;******************************************************************* | |
517 | ;;** | |
518 | ;;** The DTD (document type declaration) | |
519 | ;;** The following functions know how to skip or parse the DTD of | |
520 | ;;** a document | |
521 | ;;** | |
522 | ;;******************************************************************* | |
523 | ||
a98e819b DL |
524 | ;; Fixme: This fails at least if the DTD contains conditional sections. |
525 | ||
526 | (defun xml-skip-dtd () | |
527 | "Skip the DTD at point. | |
47db06aa | 528 | This follows the rule [28] in the XML specifications." |
6d12a4df MH |
529 | (let ((xml-validating-parser nil)) |
530 | (xml-parse-dtd))) | |
47db06aa | 531 | |
6d12a4df | 532 | (defun xml-parse-dtd (&optional parse-ns) |
a98e819b DL |
533 | "Parse the DTD at point." |
534 | (forward-char (eval-when-compile (length "<!DOCTYPE"))) | |
535 | (skip-syntax-forward " ") | |
6d12a4df MH |
536 | (if (and (looking-at ">") |
537 | xml-validating-parser) | |
538 | (error "XML: (Validity) Invalid DTD (expecting name of the document)")) | |
524425ae | 539 | |
971489ea | 540 | ;; Get the name of the document |
a98e819b DL |
541 | (looking-at xml-name-regexp) |
542 | (let ((dtd (list (match-string 0) 'dtd)) | |
971489ea | 543 | type element end-pos) |
47db06aa GM |
544 | (goto-char (match-end 0)) |
545 | ||
a98e819b DL |
546 | (skip-syntax-forward " ") |
547 | ;; XML [75] | |
548 | (cond ((looking-at "PUBLIC\\s-+") | |
549 | (goto-char (match-end 0)) | |
550 | (unless (or (re-search-forward | |
551 | "\\=\"\\([[:space:][:alnum:]-'()+,./:=?;!*#@$_%]*\\)\"" | |
552 | nil t) | |
553 | (re-search-forward | |
554 | "\\='\\([[:space:][:alnum:]-()+,./:=?;!*#@$_%]*\\)'" | |
555 | nil t)) | |
6d12a4df | 556 | (error "XML: Missing Public ID")) |
a98e819b | 557 | (let ((pubid (match-string 1))) |
6d12a4df | 558 | (skip-syntax-forward " ") |
a98e819b DL |
559 | (unless (or (re-search-forward "\\='\\([^']*\\)'" nil t) |
560 | (re-search-forward "\\=\"\\([^\"]*\\)\"" nil t)) | |
6d12a4df | 561 | (error "XML: Missing System ID")) |
a98e819b DL |
562 | (push (list pubid (match-string 1) 'public) dtd))) |
563 | ((looking-at "SYSTEM\\s-+") | |
564 | (goto-char (match-end 0)) | |
565 | (unless (or (re-search-forward "\\='\\([^']*\\)'" nil t) | |
566 | (re-search-forward "\\=\"\\([^\"]*\\)\"" nil t)) | |
6d12a4df | 567 | (error "XML: Missing System ID")) |
a98e819b DL |
568 | (push (list (match-string 1) 'system) dtd))) |
569 | (skip-syntax-forward " ") | |
570 | (if (eq ?> (char-after)) | |
571 | (forward-char) | |
a98e819b | 572 | (if (not (eq (char-after) ?\[)) |
6d12a4df | 573 | (error "XML: Bad DTD") |
a98e819b DL |
574 | (forward-char) |
575 | ;; Parse the rest of the DTD | |
6d12a4df | 576 | ;; Fixme: Deal with ATTLIST, NOTATION, PIs. |
a98e819b DL |
577 | (while (not (looking-at "\\s-*\\]")) |
578 | (skip-syntax-forward " ") | |
579 | (cond | |
580 | ||
581 | ;; Translation of rule [45] of XML specifications | |
582 | ((looking-at | |
583 | "<!ELEMENT\\s-+\\([[:alnum:].%;]+\\)\\s-+\\([^>]+\\)>") | |
584 | ||
34638996 | 585 | (setq element (match-string 1) |
a98e819b DL |
586 | type (match-string-no-properties 2)) |
587 | (setq end-pos (match-end 0)) | |
588 | ||
589 | ;; Translation of rule [46] of XML specifications | |
590 | (cond | |
591 | ((string-match "^EMPTY[ \t\n\r]*$" type) ;; empty declaration | |
592 | (setq type 'empty)) | |
593 | ((string-match "^ANY[ \t\n\r]*$" type) ;; any type of contents | |
594 | (setq type 'any)) | |
595 | ((string-match "^(\\(.*\\))[ \t\n\r]*$" type) ;; children ([47]) | |
596 | (setq type (xml-parse-elem-type (match-string 1 type)))) | |
597 | ((string-match "^%[^;]+;[ \t\n\r]*$" type) ;; substitution | |
598 | nil) | |
599 | (t | |
6d12a4df MH |
600 | (if xml-validating-parser |
601 | error "XML: (Validity) Invalid element type in the DTD"))) | |
a98e819b DL |
602 | |
603 | ;; rule [45]: the element declaration must be unique | |
6d12a4df MH |
604 | (if (and (assoc element dtd) |
605 | xml-validating-parser) | |
606 | (error "XML: (Validity) Element declarations must be unique in a DTD (<%s>)" | |
461f3ad0 | 607 | element)) |
a98e819b DL |
608 | |
609 | ;; Store the element in the DTD | |
610 | (push (list element type) dtd) | |
611 | (goto-char end-pos)) | |
612 | ((looking-at "<!--") | |
613 | (search-forward "-->")) | |
6d12a4df MH |
614 | ((looking-at (concat "<!ENTITY[ \t\n\r]*\\(" xml-name-re |
615 | "\\)[ \t\n\r]*\\(" xml-entity-value-re | |
616 | "\\)[ \t\n\r]*>")) | |
617 | (let ((name (buffer-substring (nth 2 (match-data)) | |
618 | (nth 3 (match-data)))) | |
619 | (value (buffer-substring (+ (nth 4 (match-data)) 1) | |
620 | (- (nth 5 (match-data)) 1)))) | |
621 | (goto-char (nth 1 (match-data))) | |
622 | (setq xml-entity-alist | |
623 | (append xml-entity-alist | |
624 | (list (cons name | |
625 | (with-temp-buffer | |
626 | (insert value) | |
627 | (goto-char (point-min)) | |
628 | (xml-parse-fragment | |
629 | xml-validating-parser | |
630 | parse-ns)))))))) | |
631 | ((or (looking-at (concat "<!ENTITY[ \t\n\r]+\\(" xml-name-re | |
632 | "\\)[ \t\n\r]+SYSTEM[ \t\n\r]+" | |
633 | "\\(\"[^\"]*\"\\|'[^']*'\\)[ \t\n\r]*>")) | |
634 | (looking-at (concat "<!ENTITY[ \t\n\r]+\\(" xml-name-re | |
635 | "\\)[ \t\n\r]+PUBLIC[ \t\n\r]+" | |
636 | "\"[- \r\na-zA-Z0-9'()+,./:=?;!*#@$_%]*\"" | |
637 | "\\|'[- \r\na-zA-Z0-9()+,./:=?;!*#@$_%]*'" | |
638 | "[ \t\n\r]+\\(\"[^\"]*\"\\|'[^']*'\\)" | |
639 | "[ \t\n\r]*>"))) | |
640 | (let ((name (buffer-substring (nth 2 (match-data)) | |
641 | (nth 3 (match-data)))) | |
642 | (file (buffer-substring (+ (nth 4 (match-data)) 1) | |
643 | (- (nth 5 (match-data)) 1)))) | |
644 | (goto-char (nth 1 (match-data))) | |
645 | (setq xml-entity-alist | |
646 | (append xml-entity-alist | |
647 | (list (cons name (with-temp-buffer | |
648 | (insert-file-contents file) | |
649 | (goto-char (point-min)) | |
650 | (xml-parse-fragment | |
651 | xml-validating-parser | |
652 | parse-ns)))))))) | |
a98e819b | 653 | (t |
6d12a4df MH |
654 | (error "XML: (Validity) Invalid DTD item"))))) |
655 | (if (looking-at "\\s-*]>") | |
656 | (goto-char (nth 1 (match-data))))) | |
461f3ad0 | 657 | (nreverse dtd))) |
47db06aa GM |
658 | |
659 | (defun xml-parse-elem-type (string) | |
a98e819b | 660 | "Convert element type STRING into a Lisp structure." |
47db06aa GM |
661 | |
662 | (let (elem modifier) | |
663 | (if (string-match "(\\([^)]+\\))\\([+*?]?\\)" string) | |
664 | (progn | |
665 | (setq elem (match-string 1 string) | |
666 | modifier (match-string 2 string)) | |
667 | (if (string-match "|" elem) | |
971489ea | 668 | (setq elem (cons 'choice |
47db06aa GM |
669 | (mapcar 'xml-parse-elem-type |
670 | (split-string elem "|")))) | |
671 | (if (string-match "," elem) | |
971489ea | 672 | (setq elem (cons 'seq |
47db06aa | 673 | (mapcar 'xml-parse-elem-type |
a98e819b | 674 | (split-string elem ","))))))) |
a158ff81 JB |
675 | (if (string-match "[ \t\n\r]*\\([^+*?]+\\)\\([+*?]?\\)" string) |
676 | (setq elem (match-string 1 string) | |
47db06aa GM |
677 | modifier (match-string 2 string)))) |
678 | ||
971489ea SM |
679 | (if (and (stringp elem) (string= elem "#PCDATA")) |
680 | (setq elem 'pcdata)) | |
524425ae | 681 | |
971489ea SM |
682 | (cond |
683 | ((string= modifier "+") | |
684 | (list '+ elem)) | |
685 | ((string= modifier "*") | |
686 | (list '* elem)) | |
687 | ((string= modifier "?") | |
0fa6f70c | 688 | (list '\? elem)) |
971489ea SM |
689 | (t |
690 | elem)))) | |
47db06aa | 691 | |
47db06aa GM |
692 | ;;******************************************************************* |
693 | ;;** | |
694 | ;;** Substituting special XML sequences | |
695 | ;;** | |
696 | ;;******************************************************************* | |
697 | ||
698 | (defun xml-substitute-special (string) | |
a98e819b DL |
699 | "Return STRING, after subsituting entity references." |
700 | ;; This originally made repeated passes through the string from the | |
701 | ;; beginning, which isn't correct, since then either "&amp;" or | |
702 | ;; "&amp;" won't DTRT. | |
47db06aa | 703 | |
6d12a4df MH |
704 | (let ((point 0) |
705 | children end-point) | |
706 | (while (string-match "&\\([^;]+\\);" string point) | |
707 | (setq end-point (match-end 0)) | |
708 | (let* ((this-part (match-string 1 string)) | |
709 | (prev-part (substring string point (match-beginning 0))) | |
710 | (entity (assoc this-part xml-entity-alist)) | |
711 | (expansion | |
712 | (cond ((string-match "#\\([0-9]+\\)" this-part) | |
713 | (let ((c (decode-char | |
714 | 'ucs | |
715 | (string-to-number (match-string 1 this-part))))) | |
716 | (if c (string c)))) | |
717 | ((string-match "#x\\([[:xdigit:]]+\\)" this-part) | |
718 | (let ((c (decode-char | |
719 | 'ucs | |
720 | (string-to-number (match-string 1 this-part) 16)))) | |
721 | (if c (string c)))) | |
722 | (entity | |
723 | (cdr entity)) | |
724 | (t | |
725 | (if xml-validating-parser | |
726 | (error "XML: (Validity) Undefined entity `%s'" | |
727 | (match-string 1 this-part))))))) | |
728 | ||
729 | (cond ((null children) | |
730 | (if (stringp expansion) | |
731 | (setq children (concat prev-part expansion)) | |
732 | (if (stringp (car (last expansion))) | |
733 | (progn | |
734 | (setq children | |
735 | (list (concat prev-part (car expansion)) | |
736 | (cdr expansion)))) | |
737 | (setq children (append expansion prev-part))))) | |
738 | ((stringp children) | |
739 | (if (stringp expansion) | |
740 | (setq children (concat children prev-part expansion)) | |
741 | (setq children (list expansion (concat prev-part children))))) | |
742 | ((and (stringp expansion) | |
743 | (stringp (car children))) | |
744 | (setcar children (concat prev-part expansion (car children)))) | |
745 | ((stringp expansion) | |
746 | (setq children (append (concat prev-part expansion) | |
747 | children))) | |
748 | ((stringp (car children)) | |
749 | (setcar children (concat (car children) prev-part)) | |
750 | (setq children (append expansion children))) | |
751 | (t | |
752 | (setq children (list expansion | |
753 | prev-part | |
754 | children)))) | |
755 | (setq point end-point))) | |
756 | (cond ((stringp children) | |
757 | (concat children (substring string point))) | |
758 | ((stringp (car (last children))) | |
759 | (concat (car children) (substring string point))) | |
760 | ((null children) | |
761 | string) | |
762 | (t | |
763 | (nreverse children))))) | |
47db06aa GM |
764 | ;;******************************************************************* |
765 | ;;** | |
766 | ;;** Printing a tree. | |
767 | ;;** This function is intended mainly for debugging purposes. | |
768 | ;;** | |
769 | ;;******************************************************************* | |
770 | ||
27240aa4 AS |
771 | (defun xml-debug-print (xml &optional indent-string) |
772 | "Outputs the XML in the current buffer. | |
773 | XML can be a tree or a list of nodes. | |
774 | The first line is indented with the optional INDENT-STRING." | |
775 | (setq indent-string (or indent-string "")) | |
971489ea | 776 | (dolist (node xml) |
27240aa4 AS |
777 | (xml-debug-print-internal node indent-string))) |
778 | ||
779 | (defalias 'xml-print 'xml-debug-print) | |
47db06aa | 780 | |
971489ea | 781 | (defun xml-debug-print-internal (xml indent-string) |
47db06aa | 782 | "Outputs the XML tree in the current buffer. |
a98e819b | 783 | The first line is indented with INDENT-STRING." |
47db06aa GM |
784 | (let ((tree xml) |
785 | attlist) | |
a98e819b | 786 | (insert indent-string ?< (symbol-name (xml-node-name tree))) |
524425ae | 787 | |
47db06aa | 788 | ;; output the attribute list |
971489ea | 789 | (setq attlist (xml-node-attributes tree)) |
47db06aa | 790 | (while attlist |
a98e819b | 791 | (insert ?\ (symbol-name (caar attlist)) "=\"" (cdar attlist) ?\") |
971489ea | 792 | (setq attlist (cdr attlist))) |
524425ae | 793 | |
971489ea | 794 | (setq tree (xml-node-children tree)) |
47db06aa | 795 | |
27240aa4 AS |
796 | (if (null tree) |
797 | (insert ?/ ?>) | |
798 | (insert ?>) | |
799 | ||
800 | ;; output the children | |
801 | (dolist (node tree) | |
802 | (cond | |
803 | ((listp node) | |
804 | (insert ?\n) | |
805 | (xml-debug-print-internal node (concat indent-string " "))) | |
806 | ((stringp node) (insert node)) | |
807 | (t | |
808 | (error "Invalid XML tree")))) | |
809 | ||
810 | (when (not (and (null (cdr tree)) | |
811 | (stringp (car tree)))) | |
812 | (insert ?\n indent-string)) | |
813 | (insert ?< ?/ (symbol-name (xml-node-name xml)) ?>)))) | |
47db06aa GM |
814 | |
815 | (provide 'xml) | |
816 | ||
8a02e193 | 817 | ;; arch-tag: 5864b283-5a68-4b59-a20d-36a72b353b9b |
47db06aa | 818 | ;;; xml.el ends here |