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