1 /* Interface to libxml2.
2 Copyright (C) 2010-2015 Free Software Foundation, Inc.
4 This file is part of GNU Emacs.
6 GNU Emacs is free software: you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation, either version 3 of the License, or
9 (at your option) any later version.
11 GNU Emacs is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
16 You should have received a copy of the GNU General Public License
17 along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
23 #include <libxml/tree.h>
24 #include <libxml/parser.h>
25 #include <libxml/HTMLparser.h>
28 #include "character.h"
37 DEF_DLL_FN (htmlDocPtr
, htmlReadMemory
,
38 (const char *, int, const char *, const char *, int));
39 DEF_DLL_FN (xmlDocPtr
, xmlReadMemory
,
40 (const char *, int, const char *, const char *, int));
41 DEF_DLL_FN (xmlNodePtr
, xmlDocGetRootElement
, (xmlDocPtr
));
42 DEF_DLL_FN (void, xmlFreeDoc
, (xmlDocPtr
));
43 DEF_DLL_FN (void, xmlCleanupParser
, (void));
44 DEF_DLL_FN (void, xmlCheckVersion
, (int));
47 libxml2_loaded_p (void)
49 Lisp_Object found
= Fassq (Qlibxml2_dll
, Vlibrary_cache
);
52 return EQ (XCDR (found
), Qt
) ? 1 : 0;
56 # undef htmlReadMemory
57 # undef xmlCheckVersion
58 # undef xmlCleanupParser
59 # undef xmlDocGetRootElement
63 # define htmlReadMemory fn_htmlReadMemory
64 # define xmlCheckVersion fn_xmlCheckVersion
65 # define xmlCleanupParser fn_xmlCleanupParser
66 # define xmlDocGetRootElement fn_xmlDocGetRootElement
67 # define xmlFreeDoc fn_xmlFreeDoc
68 # define xmlReadMemory fn_xmlReadMemory
71 load_dll_functions (HMODULE library
)
73 LOAD_DLL_FN (library
, htmlReadMemory
);
74 LOAD_DLL_FN (library
, xmlReadMemory
);
75 LOAD_DLL_FN (library
, xmlDocGetRootElement
);
76 LOAD_DLL_FN (library
, xmlFreeDoc
);
77 LOAD_DLL_FN (library
, xmlCleanupParser
);
78 LOAD_DLL_FN (library
, xmlCheckVersion
);
82 #else /* !WINDOWSNT */
85 libxml2_loaded_p (void)
90 #endif /* !WINDOWSNT */
93 init_libxml2_functions (void)
96 if (libxml2_loaded_p ())
102 if (!(library
= w32_delayed_load (Qlibxml2_dll
)))
104 message1 ("libxml2 library not found");
108 if (! load_dll_functions (library
))
111 Vlibrary_cache
= Fcons (Fcons (Qlibxml2_dll
, Qt
), Vlibrary_cache
);
116 Vlibrary_cache
= Fcons (Fcons (Qlibxml2_dll
, Qnil
), Vlibrary_cache
);
119 #else /* !WINDOWSNT */
121 #endif /* !WINDOWSNT */
125 make_dom (xmlNode
*node
)
127 if (node
->type
== XML_ELEMENT_NODE
)
129 Lisp_Object result
= list1 (intern ((char *) node
->name
));
132 Lisp_Object plist
= Qnil
;
134 /* First add the attributes. */
135 property
= node
->properties
;
136 while (property
!= NULL
)
138 if (property
->children
&&
139 property
->children
->content
)
141 char *content
= (char *) property
->children
->content
;
142 plist
= Fcons (Fcons (intern ((char *) property
->name
),
143 build_string (content
)),
146 property
= property
->next
;
148 result
= Fcons (Fnreverse (plist
), result
);
150 /* Then add the children of the node. */
151 child
= node
->children
;
152 while (child
!= NULL
)
154 result
= Fcons (make_dom (child
), result
);
158 return Fnreverse (result
);
160 else if (node
->type
== XML_TEXT_NODE
|| node
->type
== XML_CDATA_SECTION_NODE
)
163 return build_string ((char *) node
->content
);
167 else if (node
->type
== XML_COMMENT_NODE
)
170 return list3 (intern ("comment"), Qnil
,
171 build_string ((char *) node
->content
));
180 parse_region (Lisp_Object start
, Lisp_Object end
, Lisp_Object base_url
, Lisp_Object discard_comments
, int htmlp
)
183 Lisp_Object result
= Qnil
;
184 const char *burl
= "";
185 ptrdiff_t istart
, iend
, istart_byte
, iend_byte
;
187 xmlCheckVersion (LIBXML_VERSION
);
189 validate_region (&start
, &end
);
191 istart
= XINT (start
);
193 istart_byte
= CHAR_TO_BYTE (istart
);
194 iend_byte
= CHAR_TO_BYTE (iend
);
196 if (istart
< GPT
&& GPT
< iend
)
197 move_gap_both (iend
, iend_byte
);
199 if (! NILP (base_url
))
201 CHECK_STRING (base_url
);
202 burl
= SSDATA (base_url
);
206 doc
= htmlReadMemory ((char *) BYTE_POS_ADDR (istart_byte
),
207 iend_byte
- istart_byte
, burl
, "utf-8",
208 HTML_PARSE_RECOVER
|HTML_PARSE_NONET
|
209 HTML_PARSE_NOWARNING
|HTML_PARSE_NOERROR
|
210 HTML_PARSE_NOBLANKS
);
212 doc
= xmlReadMemory ((char *) BYTE_POS_ADDR (istart_byte
),
213 iend_byte
- istart_byte
, burl
, "utf-8",
214 XML_PARSE_NONET
|XML_PARSE_NOWARNING
|
215 XML_PARSE_NOBLANKS
|XML_PARSE_NOERROR
);
219 Lisp_Object r
= Qnil
;
220 if (NILP(discard_comments
))
222 /* If the document has toplevel comments, then this should
223 get us the nodes and the comments. */
224 xmlNode
*n
= doc
->children
;
228 result
= Fcons (r
, result
);
235 /* The document doesn't have toplevel comments or we discarded
236 them. Get the tree the proper way. */
237 xmlNode
*node
= xmlDocGetRootElement (doc
);
239 result
= make_dom (node
);
241 result
= Fcons (intern ("top"),
242 Fcons (Qnil
, Fnreverse (Fcons (r
, result
))));
251 xml_cleanup_parser (void)
253 if (libxml2_loaded_p ())
257 DEFUN ("libxml-parse-html-region", Flibxml_parse_html_region
,
258 Slibxml_parse_html_region
,
260 doc
: /* Parse the region as an HTML document and return the parse tree.
261 If BASE-URL is non-nil, it is used to expand relative URLs.
262 If DISCARD-COMMENTS is non-nil, all HTML comments are discarded. */)
263 (Lisp_Object start
, Lisp_Object end
, Lisp_Object base_url
, Lisp_Object discard_comments
)
265 if (init_libxml2_functions ())
266 return parse_region (start
, end
, base_url
, discard_comments
, 1);
270 DEFUN ("libxml-parse-xml-region", Flibxml_parse_xml_region
,
271 Slibxml_parse_xml_region
,
273 doc
: /* Parse the region as an XML document and return the parse tree.
274 If BASE-URL is non-nil, it is used to expand relative URLs.
275 If DISCARD-COMMENTS is non-nil, all HTML comments are discarded. */)
276 (Lisp_Object start
, Lisp_Object end
, Lisp_Object base_url
, Lisp_Object discard_comments
)
278 if (init_libxml2_functions ())
279 return parse_region (start
, end
, base_url
, discard_comments
, 0);
284 /***********************************************************************
286 ***********************************************************************/
290 defsubr (&Slibxml_parse_html_region
);
291 defsubr (&Slibxml_parse_xml_region
);
293 DEFSYM (Qlibxml2_dll
, "libxml2");
296 #endif /* HAVE_LIBXML2 */