src/xml.c

   1 /* Interface to libxml2.
   2    Copyright (C) 2010-2014 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 <libxml/tree.h>
  24 #include <libxml/parser.h>
  25 #include <libxml/HTMLparser.h>
  26
  27 #include "lisp.h"
  28 #include "character.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 int
  77 libxml2_loaded_p (void)
  78 {
  79   return 1;
  80 }
  81
  82 #endif  /* !WINDOWSNT */
  83
  84 static int
  85 init_libxml2_functions (void)
  86 {
  87 #ifdef WINDOWSNT
  88   if (libxml2_loaded_p ())
  89     return 1;
  90   else
  91     {
  92       HMODULE library;
  93
  94       if (!(library = w32_delayed_load (Qlibxml2_dll)))
  95         {
  96           message1 ("libxml2 library not found");
  97           return 0;
  98         }
  99
 100       /* LOAD_XML2_FN jumps to bad_library if it fails to find the
 101          named function.  */
 102       LOAD_XML2_FN (library, htmlReadMemory);
 103       LOAD_XML2_FN (library, xmlReadMemory);
 104       LOAD_XML2_FN (library, xmlDocGetRootElement);
 105       LOAD_XML2_FN (library, xmlFreeDoc);
 106       LOAD_XML2_FN (library, xmlCleanupParser);
 107       LOAD_XML2_FN (library, xmlCheckVersion);
 108
 109       Vlibrary_cache = Fcons (Fcons (Qlibxml2_dll, Qt), Vlibrary_cache);
 110       return 1;
 111     }
 112
 113  bad_library:
 114   Vlibrary_cache = Fcons (Fcons (Qlibxml2_dll, Qnil), Vlibrary_cache);
 115
 116   return 0;
 117 #else  /* !WINDOWSNT */
 118   return 1;
 119 #endif  /* !WINDOWSNT */
 120 }
 121
 122 static Lisp_Object
 123 make_dom (xmlNode *node)
 124 {
 125   if (node->type == XML_ELEMENT_NODE)
 126     {
 127       Lisp_Object result = list1 (intern ((char *) node->name));
 128       xmlNode *child;
 129       xmlAttr *property;
 130       Lisp_Object plist = Qnil;
 131
 132       /* First add the attributes. */
 133       property = node->properties;
 134       while (property != NULL)
 135         {
 136           if (property->children &&
 137               property->children->content)
 138             {
 139               char *content = (char *) property->children->content;
 140               plist = Fcons (Fcons (intern ((char *) property->name),
 141                                     build_string (content)),
 142                              plist);
 143             }
 144           property = property->next;
 145         }
 146       result = Fcons (Fnreverse (plist), result);
 147
 148       /* Then add the children of the node. */
 149       child = node->children;
 150       while (child != NULL)
 151         {
 152           result = Fcons (make_dom (child), result);
 153           child = child->next;
 154         }
 155
 156       return Fnreverse (result);
 157     }
 158   else if (node->type == XML_TEXT_NODE || node->type == XML_CDATA_SECTION_NODE)
 159     {
 160       if (node->content)
 161         return build_string ((char *) node->content);
 162       else
 163         return Qnil;
 164     }
 165   else if (node->type == XML_COMMENT_NODE)
 166     {
 167       if (node->content)
 168         return list3 (intern ("comment"), Qnil,
 169                       build_string ((char *) node->content));
 170       else
 171         return Qnil;
 172     }
 173   else
 174     return Qnil;
 175 }
 176
 177 static Lisp_Object
 178 parse_region (Lisp_Object start, Lisp_Object end, Lisp_Object base_url, Lisp_Object discard_comments, int htmlp)
 179 {
 180   xmlDoc *doc;
 181   Lisp_Object result = Qnil;
 182   const char *burl = "";
 183   ptrdiff_t istart, iend, istart_byte, iend_byte;
 184
 185   fn_xmlCheckVersion (LIBXML_VERSION);
 186
 187   validate_region (&start, &end);
 188
 189   istart = XINT (start);
 190   iend = XINT (end);
 191   istart_byte = CHAR_TO_BYTE (istart);
 192   iend_byte = CHAR_TO_BYTE (iend);
 193
 194   if (istart < GPT && GPT < iend)
 195     move_gap_both (iend, iend_byte);
 196
 197   if (! NILP (base_url))
 198     {
 199       CHECK_STRING (base_url);
 200       burl = SSDATA (base_url);
 201     }
 202
 203   if (htmlp)
 204     doc = fn_htmlReadMemory ((char *) BYTE_POS_ADDR (istart_byte),
 205                              iend_byte - istart_byte, burl, "utf-8",
 206                              HTML_PARSE_RECOVER|HTML_PARSE_NONET|
 207                              HTML_PARSE_NOWARNING|HTML_PARSE_NOERROR|
 208                              HTML_PARSE_NOBLANKS);
 209   else
 210     doc = fn_xmlReadMemory ((char *) BYTE_POS_ADDR (istart_byte),
 211                             iend_byte - istart_byte, burl, "utf-8",
 212                             XML_PARSE_NONET|XML_PARSE_NOWARNING|
 213                             XML_PARSE_NOBLANKS |XML_PARSE_NOERROR);
 214
 215   if (doc != NULL)
 216     {
 217       Lisp_Object r = Qnil;
 218       if (NILP(discard_comments))
 219         {
 220           /* If the document has toplevel comments, then this should
 221              get us the nodes and the comments. */
 222           xmlNode *n = doc->children;
 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
 232       if (NILP (result)) {
 233         /* The document doesn't have toplevel comments or we discarded
 234            them.  Get the tree the 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, 4, 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 If DISCARD-COMMENTS is non-nil, all HTML comments are discarded. */)
 261   (Lisp_Object start, Lisp_Object end, Lisp_Object base_url, Lisp_Object discard_comments)
 262 {
 263   if (init_libxml2_functions ())
 264     return parse_region (start, end, base_url, discard_comments, 1);
 265   return Qnil;
 266 }
 267
 268 DEFUN ("libxml-parse-xml-region", Flibxml_parse_xml_region,
 269        Slibxml_parse_xml_region,
 270        2, 4, 0,
 271        doc: /* Parse the region as an XML document and return the parse tree.
 272 If BASE-URL is non-nil, it is used to expand relative URLs.
 273 If DISCARD-COMMENTS is non-nil, all HTML comments are discarded. */)
 274   (Lisp_Object start, Lisp_Object end, Lisp_Object base_url, Lisp_Object discard_comments)
 275 {
 276   if (init_libxml2_functions ())
 277     return parse_region (start, end, base_url, discard_comments, 0);
 278   return Qnil;
 279 }
 280
 281 \f
 282 /***********************************************************************
 283                             Initialization
 284  ***********************************************************************/
 285 void
 286 syms_of_xml (void)
 287 {
 288   defsubr (&Slibxml_parse_html_region);
 289   defsubr (&Slibxml_parse_xml_region);
 290
 291   DEFSYM (Qlibxml2_dll, "libxml2");
 292 }
 293
 294 #endif /* HAVE_LIBXML2 */