src/xml.c

   1 /* Interface to libxml2.
   2    Copyright (C) 2010-2012 Free Software Foundation, Inc.
   3
   4 This file is part of GNU Emacs.
   5
   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.
  10
  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.
  15
  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/>.  */
  18
  19 #include <config.h>
  20
  21 #ifdef HAVE_LIBXML2
  22
  23 #include <setjmp.h>
  24 #include <libxml/tree.h>
  25 #include <libxml/parser.h>
  26 #include <libxml/HTMLparser.h>
  27
  28 #include "lisp.h"
  29 #include "character.h"
  30 #include "buffer.h"
  31
  32 \f
  33 static Lisp_Object Qlibxml2_dll;
  34
  35 #ifdef WINDOWSNT
  36
  37 #include <windows.h>
  38 #include "w32.h"
  39
  40 /* Macro for defining functions that will be loaded from the libxml2 DLL.  */
  41 #define DEF_XML2_FN(rettype,func,args) static rettype (FAR CDECL *fn_##func)args
  42
  43 /* Macro for loading libxml2 functions from the library.  */
  44 #define LOAD_XML2_FN(lib,func) {                                        \
  45     fn_##func = (void *) GetProcAddress (lib, #func);                   \
  46     if (!fn_##func) goto bad_library;                                   \
  47   }
  48
  49 DEF_XML2_FN (htmlDocPtr, htmlReadMemory,
  50              (const char *, int, const char *, const char *, int));
  51 DEF_XML2_FN (xmlDocPtr, xmlReadMemory,
  52              (const char *, int, const char *, const char *, int));
  53 DEF_XML2_FN (xmlNodePtr, xmlDocGetRootElement, (xmlDocPtr));
  54 DEF_XML2_FN (void, xmlFreeDoc, (xmlDocPtr));
  55 DEF_XML2_FN (void, xmlCleanupParser, (void));
  56 DEF_XML2_FN (void, xmlCheckVersion, (int));
  57
  58 static int
  59 libxml2_loaded_p (void)
  60 {
  61   Lisp_Object found = Fassq (Qlibxml2_dll, Vlibrary_cache);
  62
  63   if (CONSP (found))
  64     return EQ (XCDR (found), Qt) ? 1 : 0;
  65   return 0;
  66 }
  67
  68 #else  /* !WINDOWSNT */
  69
  70 #define fn_htmlReadMemory       htmlReadMemory
  71 #define fn_xmlReadMemory        xmlReadMemory
  72 #define fn_xmlDocGetRootElement xmlDocGetRootElement
  73 #define fn_xmlFreeDoc           xmlFreeDoc
  74 #define fn_xmlCleanupParser     xmlCleanupParser
  75 #define fn_xmlCheckVersion      xmlCheckVersion
  76
  77 static inline int
  78 libxml2_loaded_p (void)
  79 {
  80   return 1;
  81 }
  82
  83 #endif  /* !WINDOWSNT */
  84
  85 static int
  86 init_libxml2_functions (Lisp_Object libraries)
  87 {
  88 #ifdef WINDOWSNT
  89   if (libxml2_loaded_p ())
  90     return 1;
  91   else
  92     {
  93       HMODULE library;
  94
  95       if (!(library = w32_delayed_load (libraries, Qlibxml2_dll)))
  96         {
  97           message ("%s", "libxml2 library not found");
  98           return 0;
  99         }
 100
 101       /* LOAD_XML2_FN jumps to bad_library if it fails to find the
 102          named function.  */
 103       LOAD_XML2_FN (library, htmlReadMemory);
 104       LOAD_XML2_FN (library, xmlReadMemory);
 105       LOAD_XML2_FN (library, xmlDocGetRootElement);
 106       LOAD_XML2_FN (library, xmlFreeDoc);
 107       LOAD_XML2_FN (library, xmlCleanupParser);
 108       LOAD_XML2_FN (library, xmlCheckVersion);
 109
 110       Vlibrary_cache = Fcons (Fcons (Qlibxml2_dll, Qt), Vlibrary_cache);
 111       return 1;
 112     }
 113
 114  bad_library:
 115   Vlibrary_cache = Fcons (Fcons (Qlibxml2_dll, Qnil), Vlibrary_cache);
 116
 117   return 0;
 118 #else  /* !WINDOWSNT */
 119   return 1;
 120 #endif  /* !WINDOWSNT */
 121 }
 122
 123 static Lisp_Object
 124 make_dom (xmlNode *node)
 125 {
 126   if (node->type == XML_ELEMENT_NODE)
 127     {
 128       Lisp_Object result = Fcons (intern ((char *) node->name), Qnil);
 129       xmlNode *child;
 130       xmlAttr *property;
 131       Lisp_Object plist = Qnil;
 132
 133       /* First add the attributes. */
 134       property = node->properties;
 135       while (property != NULL)
 136         {
 137           if (property->children &&
 138               property->children->content)
 139             {
 140               char *content = (char *) property->children->content;
 141               plist = Fcons (Fcons (intern ((char *) property->name),
 142                                     build_string (content)),
 143                              plist);
 144             }
 145           property = property->next;
 146         }
 147       result = Fcons (Fnreverse (plist), result);
 148
 149       /* Then add the children of the node. */
 150       child = node->children;
 151       while (child != NULL)
 152         {
 153           result = Fcons (make_dom (child), result);
 154           child = child->next;
 155         }
 156
 157       return Fnreverse (result);
 158     }
 159   else if (node->type == XML_TEXT_NODE || node->type == XML_CDATA_SECTION_NODE)
 160     {
 161       if (node->content)
 162         return build_string ((char *) node->content);
 163       else
 164         return Qnil;
 165     }
 166   else if (node->type == XML_COMMENT_NODE)
 167     {
 168       if (node->content)
 169         return list3 (intern ("comment"), Qnil,
 170                       build_string ((char *) node->content));
 171       else
 172         return Qnil;
 173     }
 174   else
 175     return Qnil;
 176 }
 177
 178 static Lisp_Object
 179 parse_region (Lisp_Object start, Lisp_Object end, Lisp_Object base_url, int htmlp)
 180 {
 181   xmlDoc *doc;
 182   Lisp_Object result = Qnil;
 183   const char *burl = "";
 184   ptrdiff_t bytes;
 185   ptrdiff_t istart, iend;
 186
 187   fn_xmlCheckVersion (LIBXML_VERSION);
 188
 189   validate_region (&start, &end);
 190
 191   istart = XINT (start);
 192   iend = XINT (end);
 193
 194   if (istart < GPT && GPT < iend)
 195     move_gap (iend);
 196
 197   if (! NILP (base_url))
 198     {
 199       CHECK_STRING (base_url);
 200       burl = SSDATA (base_url);
 201     }
 202
 203   bytes = CHAR_TO_BYTE (iend) - CHAR_TO_BYTE (istart);
 204
 205   if (htmlp)
 206     doc = fn_htmlReadMemory ((char *) BYTE_POS_ADDR (CHAR_TO_BYTE (istart)),
 207                              bytes, burl, "utf-8",
 208                              HTML_PARSE_RECOVER|HTML_PARSE_NONET|
 209                              HTML_PARSE_NOWARNING|HTML_PARSE_NOERROR|
 210                              HTML_PARSE_NOBLANKS);
 211   else
 212     doc = fn_xmlReadMemory ((char *) BYTE_POS_ADDR (CHAR_TO_BYTE (istart)),
 213                             bytes, burl, "utf-8",
 214                             XML_PARSE_NONET|XML_PARSE_NOWARNING|
 215                             XML_PARSE_NOBLANKS |XML_PARSE_NOERROR);
 216
 217   if (doc != NULL)
 218     {
 219       /* If the document is just comments, then this should get us the
 220          nodes anyway. */
 221       xmlNode *n = doc->children->next;
 222       Lisp_Object r = Qnil;
 223
 224       while (n) {
 225         if (!NILP (r))
 226           result = Fcons (r, result);
 227         r = make_dom (n);
 228         n = n->next;
 229       }
 230
 231       if (NILP (result)) {
 232         /* The document isn't just comments, so get the tree the
 233            proper way. */
 234         xmlNode *node = fn_xmlDocGetRootElement (doc);
 235         if (node != NULL)
 236           result = make_dom (node);
 237       } else
 238         result = Fcons (intern ("top"),
 239                         Fcons (Qnil, Fnreverse (Fcons (r, result))));
 240
 241       fn_xmlFreeDoc (doc);
 242     }
 243
 244   return result;
 245 }
 246
 247 void
 248 xml_cleanup_parser (void)
 249 {
 250   if (libxml2_loaded_p ())
 251     fn_xmlCleanupParser ();
 252 }
 253
 254 DEFUN ("libxml-parse-html-region", Flibxml_parse_html_region,
 255        Slibxml_parse_html_region,
 256        2, 3, 0,
 257        doc: /* Parse the region as an HTML document and return the parse tree.
 258 If BASE-URL is non-nil, it is used to expand relative URLs.  */)
 259   (Lisp_Object start, Lisp_Object end, Lisp_Object base_url)
 260 {
 261   if (init_libxml2_functions (Vdynamic_library_alist))
 262     return parse_region (start, end, base_url, 1);
 263   return Qnil;
 264 }
 265
 266 DEFUN ("libxml-parse-xml-region", Flibxml_parse_xml_region,
 267        Slibxml_parse_xml_region,
 268        2, 3, 0,
 269        doc: /* Parse the region as an XML document and return the parse tree.
 270 If BASE-URL is non-nil, it is used to expand relative URLs.  */)
 271   (Lisp_Object start, Lisp_Object end, Lisp_Object base_url)
 272 {
 273   if (init_libxml2_functions (Vdynamic_library_alist))
 274     return parse_region (start, end, base_url, 0);
 275   return Qnil;
 276 }
 277
 278 \f
 279 /***********************************************************************
 280                             Initialization
 281  ***********************************************************************/
 282 void
 283 syms_of_xml (void)
 284 {
 285   defsubr (&Slibxml_parse_html_region);
 286   defsubr (&Slibxml_parse_xml_region);
 287
 288   DEFSYM (Qlibxml2_dll, "libxml2");
 289 }
 290
 291 #endif /* HAVE_LIBXML2 */