libxml-ruby 2.7.0 → 2.9.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (85) hide show
  1. checksums.yaml +6 -14
  2. data/HISTORY +33 -11
  3. data/README.rdoc +7 -7
  4. data/Rakefile +80 -77
  5. data/ext/libxml/extconf.h +4 -5
  6. data/ext/libxml/extconf.rb +57 -118
  7. data/ext/libxml/libxml.c +4 -0
  8. data/ext/libxml/ruby_xml.c +977 -893
  9. data/ext/libxml/ruby_xml.h +20 -10
  10. data/ext/libxml/ruby_xml_attr.c +333 -333
  11. data/ext/libxml/ruby_xml_attr_decl.c +2 -2
  12. data/ext/libxml/ruby_xml_cbg.c +85 -85
  13. data/ext/libxml/ruby_xml_document.c +1133 -1147
  14. data/ext/libxml/ruby_xml_dtd.c +261 -268
  15. data/ext/libxml/ruby_xml_encoding.c +262 -260
  16. data/ext/libxml/ruby_xml_encoding.h +19 -19
  17. data/ext/libxml/ruby_xml_html_parser_context.c +337 -338
  18. data/ext/libxml/ruby_xml_input_cbg.c +191 -191
  19. data/ext/libxml/ruby_xml_io.c +52 -50
  20. data/ext/libxml/ruby_xml_namespace.c +2 -2
  21. data/ext/libxml/ruby_xml_node.c +1446 -1452
  22. data/ext/libxml/ruby_xml_parser_context.c +999 -1001
  23. data/ext/libxml/ruby_xml_reader.c +1226 -1228
  24. data/ext/libxml/ruby_xml_relaxng.c +110 -111
  25. data/ext/libxml/ruby_xml_sax2_handler.c +326 -328
  26. data/ext/libxml/ruby_xml_schema.c +300 -301
  27. data/ext/libxml/ruby_xml_version.h +3 -3
  28. data/ext/libxml/ruby_xml_writer.c +34 -16
  29. data/ext/libxml/ruby_xml_xpath.c +188 -187
  30. data/ext/libxml/ruby_xml_xpath_context.c +360 -361
  31. data/ext/libxml/ruby_xml_xpath_object.c +335 -335
  32. data/libxml-ruby.gemspec +47 -44
  33. data/test/tc_attr.rb +5 -7
  34. data/test/tc_attr_decl.rb +5 -6
  35. data/test/tc_attributes.rb +1 -2
  36. data/test/tc_canonicalize.rb +1 -2
  37. data/test/tc_deprecated_require.rb +1 -2
  38. data/test/tc_document.rb +4 -5
  39. data/test/tc_document_write.rb +2 -3
  40. data/test/tc_dtd.rb +4 -5
  41. data/test/tc_encoding.rb +126 -126
  42. data/test/tc_encoding_sax.rb +4 -3
  43. data/test/tc_error.rb +14 -15
  44. data/test/tc_html_parser.rb +15 -7
  45. data/test/tc_html_parser_context.rb +1 -2
  46. data/test/tc_namespace.rb +2 -3
  47. data/test/tc_namespaces.rb +5 -6
  48. data/test/tc_node.rb +2 -3
  49. data/test/tc_node_cdata.rb +2 -3
  50. data/test/tc_node_comment.rb +1 -2
  51. data/test/tc_node_copy.rb +1 -2
  52. data/test/tc_node_edit.rb +5 -7
  53. data/test/tc_node_pi.rb +1 -2
  54. data/test/tc_node_text.rb +2 -3
  55. data/test/tc_node_write.rb +2 -3
  56. data/test/tc_node_xlink.rb +1 -2
  57. data/test/tc_parser.rb +18 -24
  58. data/test/tc_parser_context.rb +6 -7
  59. data/test/tc_properties.rb +1 -2
  60. data/test/tc_reader.rb +9 -10
  61. data/test/tc_relaxng.rb +4 -5
  62. data/test/tc_sax_parser.rb +9 -10
  63. data/test/tc_schema.rb +4 -5
  64. data/test/tc_traversal.rb +1 -2
  65. data/test/tc_writer.rb +1 -2
  66. data/test/tc_xinclude.rb +1 -2
  67. data/test/tc_xml.rb +1 -2
  68. data/test/tc_xpath.rb +8 -9
  69. data/test/tc_xpath_context.rb +3 -4
  70. data/test/tc_xpath_expression.rb +3 -4
  71. data/test/tc_xpointer.rb +1 -3
  72. data/test/test_helper.rb +3 -1
  73. data/test/test_suite.rb +0 -1
  74. metadata +90 -72
  75. data/test/etc_doc_to_s.rb +0 -21
  76. data/test/ets_doc_file.rb +0 -17
  77. data/test/ets_doc_to_s.rb +0 -23
  78. data/test/ets_gpx.rb +0 -28
  79. data/test/ets_node_gc.rb +0 -23
  80. data/test/ets_test.xml +0 -2
  81. data/test/ets_tsr.rb +0 -11
  82. data/test/model/kml_sample.xml +0 -915
  83. data/test/remove_test.rb +0 -9
  84. data/test/tc_gc.rb +0 -86
  85. data/test/tc_parser.rb.orig +0 -384
@@ -1,338 +1,337 @@
1
- /* Please see the LICENSE file for copyright and distribution information */
2
-
3
- #include "ruby_libxml.h"
4
- #include "ruby_xml_html_parser_context.h"
5
-
6
-
7
- /*
8
- * Document-class: LibXML::XML::HTMLParser::Context
9
- *
10
- * The XML::HTMLParser::Context class provides in-depth control over how
11
- * a document is parsed.
12
- */
13
-
14
- VALUE cXMLHtmlParserContext;
15
- static ID IO_ATTR;
16
-
17
- /* OS X 10.5 ships with libxml2 version 2.6.16 which does not expose the
18
- htmlNewParserCtxt (or htmlInitParserCtxt which it uses) method. htmlNewParserCtxt
19
- wasn't added to the libxml2 header files until 2.6.27. So the next two
20
- methods are simply copied from a newer version of libxml2 (2.7.2). */
21
- #if LIBXML_VERSION < 20627
22
- #define XML_CTXT_FINISH_DTD_0 0xabcd1234
23
- static int htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
24
- {
25
- htmlSAXHandler *sax;
26
- if (ctxt == NULL) return(-1);
27
-
28
- memset(ctxt, 0, sizeof(htmlParserCtxt));
29
- ctxt->dict = xmlDictCreate();
30
- if (ctxt->dict == NULL) {
31
- rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
32
- return(-1);
33
- }
34
- sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
35
- if (sax == NULL) {
36
- rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
37
- return(-1);
38
- }
39
- else
40
- memset(sax, 0, sizeof(htmlSAXHandler));
41
-
42
- ctxt->inputTab = (htmlParserInputPtr *) xmlMalloc(5 * sizeof(htmlParserInputPtr));
43
- if (ctxt->inputTab == NULL) {
44
- rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
45
- ctxt->inputNr = 0;
46
- ctxt->inputMax = 0;
47
- ctxt->input = NULL;
48
- return(-1);
49
- }
50
- ctxt->inputNr = 0;
51
- ctxt->inputMax = 5;
52
- ctxt->input = NULL;
53
- ctxt->version = NULL;
54
- ctxt->encoding = NULL;
55
- ctxt->standalone = -1;
56
- ctxt->instate = XML_PARSER_START;
57
-
58
- ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
59
- if (ctxt->nodeTab == NULL) {
60
- rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
61
- ctxt->nodeNr = 0;
62
- ctxt->nodeMax = 0;
63
- ctxt->node = NULL;
64
- ctxt->inputNr = 0;
65
- ctxt->inputMax = 0;
66
- ctxt->input = NULL;
67
- return(-1);
68
- }
69
- ctxt->nodeNr = 0;
70
- ctxt->nodeMax = 10;
71
- ctxt->node = NULL;
72
-
73
- ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
74
- if (ctxt->nameTab == NULL) {
75
- rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
76
- ctxt->nameNr = 0;
77
- ctxt->nameMax = 10;
78
- ctxt->name = NULL;
79
- ctxt->nodeNr = 0;
80
- ctxt->nodeMax = 0;
81
- ctxt->node = NULL;
82
- ctxt->inputNr = 0;
83
- ctxt->inputMax = 0;
84
- ctxt->input = NULL;
85
- return(-1);
86
- }
87
- ctxt->nameNr = 0;
88
- ctxt->nameMax = 10;
89
- ctxt->name = NULL;
90
-
91
- if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
92
- else {
93
- ctxt->sax = sax;
94
- memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
95
- }
96
- ctxt->userData = ctxt;
97
- ctxt->myDoc = NULL;
98
- ctxt->wellFormed = 1;
99
- ctxt->replaceEntities = 0;
100
- ctxt->linenumbers = xmlLineNumbersDefaultValue;
101
- ctxt->html = 1;
102
- ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
103
- ctxt->vctxt.userData = ctxt;
104
- ctxt->vctxt.error = xmlParserValidityError;
105
- ctxt->vctxt.warning = xmlParserValidityWarning;
106
- ctxt->record_info = 0;
107
- ctxt->validate = 0;
108
- ctxt->nbChars = 0;
109
- ctxt->checkIndex = 0;
110
- ctxt->catalogs = NULL;
111
- xmlInitNodeInfoSeq(&ctxt->node_seq);
112
- return(0);
113
- }
114
-
115
- static htmlParserCtxtPtr htmlNewParserCtxt(void)
116
- {
117
- xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
118
- if (ctxt == NULL) {
119
- rb_raise(rb_eNoMemError, "NewParserCtxt: out of memory\n");
120
- return(NULL);
121
- }
122
- memset(ctxt, 0, sizeof(xmlParserCtxt));
123
- if (htmlInitParserCtxt(ctxt) < 0) {
124
- htmlFreeParserCtxt(ctxt);
125
- return(NULL);
126
- }
127
- return(ctxt);
128
- }
129
- #endif
130
-
131
- static void rxml_html_parser_context_free(htmlParserCtxtPtr ctxt)
132
- {
133
- htmlFreeParserCtxt(ctxt);
134
- }
135
-
136
- static VALUE rxml_html_parser_context_wrap(htmlParserCtxtPtr ctxt)
137
- {
138
- return Data_Wrap_Struct(cXMLHtmlParserContext, NULL, rxml_html_parser_context_free, ctxt);
139
- }
140
-
141
- /* call-seq:
142
- * XML::HTMLParser::Context.file(file) -> XML::HTMLParser::Context
143
- *
144
- * Creates a new parser context based on the specified file or uri.
145
- *
146
- * Parameters:
147
- *
148
- * file - A filename or uri.
149
- */
150
- static VALUE rxml_html_parser_context_file(VALUE klass, VALUE file)
151
- {
152
- htmlParserCtxtPtr ctxt = htmlCreateFileParserCtxt(StringValuePtr(file), NULL);
153
- if (!ctxt)
154
- rxml_raise(&xmlLastError);
155
-
156
- /* This is annoying, but xmlInitParserCtxt (called indirectly above) and
157
- xmlCtxtUseOptionsInternal (called below) initialize slightly different
158
- context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
159
- sets to 0 and xmlCtxtUseOptionsInternal sets to 1. So we have to call both. */
160
- htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
161
-
162
- return rxml_html_parser_context_wrap(ctxt);
163
- }
164
-
165
- /* call-seq:
166
- * XML::HTMLParser::Context.io(io) -> XML::HTMLParser::Context
167
- *
168
- * Creates a new parser context based on the specified io object.
169
- *
170
- * Parameters:
171
- *
172
- * io - A ruby IO object.
173
- */
174
- static VALUE rxml_html_parser_context_io(VALUE klass, VALUE io)
175
- {
176
- VALUE result;
177
- htmlParserCtxtPtr ctxt;
178
- xmlParserInputBufferPtr input;
179
- xmlParserInputPtr stream;
180
-
181
- if (NIL_P(io))
182
- rb_raise(rb_eTypeError, "Must pass in an IO object");
183
-
184
- input = xmlParserInputBufferCreateIO((xmlInputReadCallback) rxml_read_callback, NULL,
185
- (void*)io, XML_CHAR_ENCODING_NONE);
186
-
187
- ctxt = htmlNewParserCtxt();
188
- if (!ctxt)
189
- {
190
- xmlFreeParserInputBuffer(input);
191
- rxml_raise(&xmlLastError);
192
- }
193
-
194
- /* This is annoying, but xmlInitParserCtxt (called indirectly above) and
195
- xmlCtxtUseOptionsInternal (called below) initialize slightly different
196
- context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
197
- sets to 0 and xmlCtxtUseOptionsInternal sets to 1. So we have to call both. */
198
- htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
199
-
200
- stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
201
-
202
- if (!stream)
203
- {
204
- xmlFreeParserInputBuffer(input);
205
- xmlFreeParserCtxt(ctxt);
206
- rxml_raise(&xmlLastError);
207
- }
208
- inputPush(ctxt, stream);
209
- result = rxml_html_parser_context_wrap(ctxt);
210
-
211
- /* Attach io object to parser so it won't get freed.*/
212
- rb_ivar_set(result, IO_ATTR, io);
213
-
214
- return result;
215
- }
216
-
217
- /* call-seq:
218
- * XML::HTMLParser::Context.string(string) -> XML::HTMLParser::Context
219
- *
220
- * Creates a new parser context based on the specified string.
221
- *
222
- * Parameters:
223
- *
224
- * string - A string that contains the data to parse.
225
- */
226
- static VALUE rxml_html_parser_context_string(VALUE klass, VALUE string)
227
- {
228
- htmlParserCtxtPtr ctxt;
229
- Check_Type(string, T_STRING);
230
-
231
- if (RSTRING_LEN(string) == 0)
232
- rb_raise(rb_eArgError, "Must specify a string with one or more characters");
233
-
234
- ctxt = xmlCreateMemoryParserCtxt(StringValuePtr(string),
235
- RSTRING_LEN(string));
236
- if (!ctxt)
237
- rxml_raise(&xmlLastError);
238
-
239
- /* This is annoying, but xmlInitParserCtxt (called indirectly above) and
240
- xmlCtxtUseOptionsInternal (called below) initialize slightly different
241
- context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
242
- sets to 0 and xmlCtxtUseOptionsInternal sets to 1. So we have to call both. */
243
- htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
244
-
245
- htmlDefaultSAXHandlerInit();
246
- if (ctxt->sax != NULL)
247
- memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
248
-
249
- return rxml_html_parser_context_wrap(ctxt);
250
- }
251
-
252
- /*
253
- * call-seq:
254
- * context.close -> nil
255
- *
256
- * Closes the underlying input streams. This is useful when parsing a large amount of
257
- * files and you want to close the files without relying on Ruby's garbage collector
258
- * to run.
259
- */
260
- static VALUE rxml_html_parser_context_close(VALUE self)
261
- {
262
- htmlParserCtxtPtr ctxt;
263
- xmlParserInputPtr xinput;
264
- Data_Get_Struct(self, htmlParserCtxt, ctxt);
265
-
266
- while ((xinput = inputPop(ctxt)) != NULL)
267
- {
268
- xmlFreeInputStream(xinput);
269
- }
270
- return Qnil;
271
- }
272
-
273
- /*
274
- * call-seq:
275
- * context.disable_cdata = (true|false)
276
- *
277
- * Control whether the CDATA nodes will be created in this context.
278
- */
279
- static VALUE rxml_html_parser_context_disable_cdata_set(VALUE self, VALUE bool)
280
- {
281
- htmlParserCtxtPtr ctxt;
282
- Data_Get_Struct(self, htmlParserCtxt, ctxt);
283
-
284
- if (ctxt->sax == NULL)
285
- rb_raise(rb_eRuntimeError, "Sax handler is not yet set");
286
-
287
- /* LibXML controls this internally with the default SAX handler. */
288
- if (bool)
289
- ctxt->sax->cdataBlock = NULL;
290
- else
291
- ctxt->sax->cdataBlock = htmlDefaultSAXHandler.cdataBlock;
292
-
293
- return bool;
294
- }
295
-
296
- /*
297
- * call-seq:
298
- * context.options = XML::Parser::Options::NOENT |
299
- XML::Parser::Options::NOCDATA
300
- *
301
- * Provides control over the execution of a parser. Valid values
302
- * are the constants defined on XML::Parser::Options. Multiple
303
- * options can be combined by using Bitwise OR (|).
304
- */
305
- static VALUE rxml_html_parser_context_options_set(VALUE self, VALUE options)
306
- {
307
- int result;
308
- int xml_options = NUM2INT(options);
309
- htmlParserCtxtPtr ctxt;
310
- Check_Type(options, T_FIXNUM);
311
-
312
- Data_Get_Struct(self, htmlParserCtxt, ctxt);
313
- result = htmlCtxtUseOptions(ctxt, xml_options);
314
-
315
- #if LIBXML_VERSION >= 20707
316
- /* Big hack here, but htmlCtxtUseOptions doens't support HTML_PARSE_NOIMPLIED.
317
- So do it ourselves. There must be a better way??? */
318
- if (xml_options & HTML_PARSE_NOIMPLIED)
319
- {
320
- ctxt->options |= HTML_PARSE_NOIMPLIED;
321
- }
322
- #endif
323
-
324
- return self;
325
- }
326
-
327
- void rxml_init_html_parser_context(void)
328
- {
329
- IO_ATTR = ID2SYM(rb_intern("@io"));
330
- cXMLHtmlParserContext = rb_define_class_under(cXMLHtmlParser, "Context", cXMLParserContext);
331
-
332
- rb_define_singleton_method(cXMLHtmlParserContext, "file", rxml_html_parser_context_file, 1);
333
- rb_define_singleton_method(cXMLHtmlParserContext, "io", rxml_html_parser_context_io, 1);
334
- rb_define_singleton_method(cXMLHtmlParserContext, "string", rxml_html_parser_context_string, 1);
335
- rb_define_method(cXMLHtmlParserContext, "close", rxml_html_parser_context_close, 0);
336
- rb_define_method(cXMLHtmlParserContext, "disable_cdata=", rxml_html_parser_context_disable_cdata_set, 1);
337
- rb_define_method(cXMLHtmlParserContext, "options=", rxml_html_parser_context_options_set, 1);
338
- }
1
+ /* Please see the LICENSE file for copyright and distribution information */
2
+
3
+ #include "ruby_libxml.h"
4
+ #include "ruby_xml_html_parser_context.h"
5
+
6
+
7
+ /*
8
+ * Document-class: LibXML::XML::HTMLParser::Context
9
+ *
10
+ * The XML::HTMLParser::Context class provides in-depth control over how
11
+ * a document is parsed.
12
+ */
13
+
14
+ VALUE cXMLHtmlParserContext;
15
+ static ID IO_ATTR;
16
+
17
+ /* OS X 10.5 ships with libxml2 version 2.6.16 which does not expose the
18
+ htmlNewParserCtxt (or htmlInitParserCtxt which it uses) method. htmlNewParserCtxt
19
+ wasn't added to the libxml2 header files until 2.6.27. So the next two
20
+ methods are simply copied from a newer version of libxml2 (2.7.2). */
21
+ #if LIBXML_VERSION < 20627
22
+ #define XML_CTXT_FINISH_DTD_0 0xabcd1234
23
+ static int htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
24
+ {
25
+ htmlSAXHandler *sax;
26
+ if (ctxt == NULL) return(-1);
27
+
28
+ memset(ctxt, 0, sizeof(htmlParserCtxt));
29
+ ctxt->dict = xmlDictCreate();
30
+ if (ctxt->dict == NULL) {
31
+ rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
32
+ return(-1);
33
+ }
34
+ sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
35
+ if (sax == NULL) {
36
+ rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
37
+ return(-1);
38
+ }
39
+ else
40
+ memset(sax, 0, sizeof(htmlSAXHandler));
41
+
42
+ ctxt->inputTab = (htmlParserInputPtr *) xmlMalloc(5 * sizeof(htmlParserInputPtr));
43
+ if (ctxt->inputTab == NULL) {
44
+ rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
45
+ ctxt->inputNr = 0;
46
+ ctxt->inputMax = 0;
47
+ ctxt->input = NULL;
48
+ return(-1);
49
+ }
50
+ ctxt->inputNr = 0;
51
+ ctxt->inputMax = 5;
52
+ ctxt->input = NULL;
53
+ ctxt->version = NULL;
54
+ ctxt->encoding = NULL;
55
+ ctxt->standalone = -1;
56
+ ctxt->instate = XML_PARSER_START;
57
+
58
+ ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
59
+ if (ctxt->nodeTab == NULL) {
60
+ rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
61
+ ctxt->nodeNr = 0;
62
+ ctxt->nodeMax = 0;
63
+ ctxt->node = NULL;
64
+ ctxt->inputNr = 0;
65
+ ctxt->inputMax = 0;
66
+ ctxt->input = NULL;
67
+ return(-1);
68
+ }
69
+ ctxt->nodeNr = 0;
70
+ ctxt->nodeMax = 10;
71
+ ctxt->node = NULL;
72
+
73
+ ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
74
+ if (ctxt->nameTab == NULL) {
75
+ rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
76
+ ctxt->nameNr = 0;
77
+ ctxt->nameMax = 10;
78
+ ctxt->name = NULL;
79
+ ctxt->nodeNr = 0;
80
+ ctxt->nodeMax = 0;
81
+ ctxt->node = NULL;
82
+ ctxt->inputNr = 0;
83
+ ctxt->inputMax = 0;
84
+ ctxt->input = NULL;
85
+ return(-1);
86
+ }
87
+ ctxt->nameNr = 0;
88
+ ctxt->nameMax = 10;
89
+ ctxt->name = NULL;
90
+
91
+ if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
92
+ else {
93
+ ctxt->sax = sax;
94
+ memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
95
+ }
96
+ ctxt->userData = ctxt;
97
+ ctxt->myDoc = NULL;
98
+ ctxt->wellFormed = 1;
99
+ ctxt->replaceEntities = 0;
100
+ ctxt->linenumbers = xmlLineNumbersDefaultValue;
101
+ ctxt->html = 1;
102
+ ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
103
+ ctxt->vctxt.userData = ctxt;
104
+ ctxt->vctxt.error = xmlParserValidityError;
105
+ ctxt->vctxt.warning = xmlParserValidityWarning;
106
+ ctxt->record_info = 0;
107
+ ctxt->validate = 0;
108
+ ctxt->nbChars = 0;
109
+ ctxt->checkIndex = 0;
110
+ ctxt->catalogs = NULL;
111
+ xmlInitNodeInfoSeq(&ctxt->node_seq);
112
+ return(0);
113
+ }
114
+
115
+ static htmlParserCtxtPtr htmlNewParserCtxt(void)
116
+ {
117
+ xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
118
+ if (ctxt == NULL) {
119
+ rb_raise(rb_eNoMemError, "NewParserCtxt: out of memory\n");
120
+ return(NULL);
121
+ }
122
+ memset(ctxt, 0, sizeof(xmlParserCtxt));
123
+ if (htmlInitParserCtxt(ctxt) < 0) {
124
+ htmlFreeParserCtxt(ctxt);
125
+ return(NULL);
126
+ }
127
+ return(ctxt);
128
+ }
129
+ #endif
130
+
131
+ static void rxml_html_parser_context_free(htmlParserCtxtPtr ctxt)
132
+ {
133
+ htmlFreeParserCtxt(ctxt);
134
+ }
135
+
136
+ static VALUE rxml_html_parser_context_wrap(htmlParserCtxtPtr ctxt)
137
+ {
138
+ return Data_Wrap_Struct(cXMLHtmlParserContext, NULL, rxml_html_parser_context_free, ctxt);
139
+ }
140
+
141
+ /* call-seq:
142
+ * XML::HTMLParser::Context.file(file) -> XML::HTMLParser::Context
143
+ *
144
+ * Creates a new parser context based on the specified file or uri.
145
+ *
146
+ * Parameters:
147
+ *
148
+ * file - A filename or uri.
149
+ */
150
+ static VALUE rxml_html_parser_context_file(VALUE klass, VALUE file)
151
+ {
152
+ htmlParserCtxtPtr ctxt = htmlCreateFileParserCtxt(StringValuePtr(file), NULL);
153
+ if (!ctxt)
154
+ rxml_raise(&xmlLastError);
155
+
156
+ /* This is annoying, but xmlInitParserCtxt (called indirectly above) and
157
+ xmlCtxtUseOptionsInternal (called below) initialize slightly different
158
+ context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
159
+ sets to 0 and xmlCtxtUseOptionsInternal sets to 1. So we have to call both. */
160
+ htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
161
+
162
+ return rxml_html_parser_context_wrap(ctxt);
163
+ }
164
+
165
+ /* call-seq:
166
+ * XML::HTMLParser::Context.io(io) -> XML::HTMLParser::Context
167
+ *
168
+ * Creates a new parser context based on the specified io object.
169
+ *
170
+ * Parameters:
171
+ *
172
+ * io - A ruby IO object.
173
+ */
174
+ static VALUE rxml_html_parser_context_io(VALUE klass, VALUE io)
175
+ {
176
+ VALUE result;
177
+ htmlParserCtxtPtr ctxt;
178
+ xmlParserInputBufferPtr input;
179
+ xmlParserInputPtr stream;
180
+
181
+ if (NIL_P(io))
182
+ rb_raise(rb_eTypeError, "Must pass in an IO object");
183
+
184
+ input = xmlParserInputBufferCreateIO((xmlInputReadCallback) rxml_read_callback, NULL,
185
+ (void*)io, XML_CHAR_ENCODING_NONE);
186
+
187
+ ctxt = htmlNewParserCtxt();
188
+ if (!ctxt)
189
+ {
190
+ xmlFreeParserInputBuffer(input);
191
+ rxml_raise(&xmlLastError);
192
+ }
193
+
194
+ /* This is annoying, but xmlInitParserCtxt (called indirectly above) and
195
+ xmlCtxtUseOptionsInternal (called below) initialize slightly different
196
+ context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
197
+ sets to 0 and xmlCtxtUseOptionsInternal sets to 1. So we have to call both. */
198
+ htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
199
+
200
+ stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
201
+
202
+ if (!stream)
203
+ {
204
+ xmlFreeParserInputBuffer(input);
205
+ xmlFreeParserCtxt(ctxt);
206
+ rxml_raise(&xmlLastError);
207
+ }
208
+ inputPush(ctxt, stream);
209
+ result = rxml_html_parser_context_wrap(ctxt);
210
+
211
+ /* Attach io object to parser so it won't get freed.*/
212
+ rb_ivar_set(result, IO_ATTR, io);
213
+
214
+ return result;
215
+ }
216
+
217
+ /* call-seq:
218
+ * XML::HTMLParser::Context.string(string) -> XML::HTMLParser::Context
219
+ *
220
+ * Creates a new parser context based on the specified string.
221
+ *
222
+ * Parameters:
223
+ *
224
+ * string - A string that contains the data to parse.
225
+ */
226
+ static VALUE rxml_html_parser_context_string(VALUE klass, VALUE string)
227
+ {
228
+ htmlParserCtxtPtr ctxt;
229
+ Check_Type(string, T_STRING);
230
+
231
+ if (RSTRING_LEN(string) == 0)
232
+ rb_raise(rb_eArgError, "Must specify a string with one or more characters");
233
+
234
+ ctxt = xmlCreateMemoryParserCtxt(StringValuePtr(string),
235
+ (int)RSTRING_LEN(string));
236
+ if (!ctxt)
237
+ rxml_raise(&xmlLastError);
238
+
239
+ /* This is annoying, but xmlInitParserCtxt (called indirectly above) and
240
+ xmlCtxtUseOptionsInternal (called below) initialize slightly different
241
+ context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
242
+ sets to 0 and xmlCtxtUseOptionsInternal sets to 1. So we have to call both. */
243
+ htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
244
+
245
+ htmlDefaultSAXHandlerInit();
246
+ if (ctxt->sax != NULL)
247
+ memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
248
+
249
+ return rxml_html_parser_context_wrap(ctxt);
250
+ }
251
+
252
+ /*
253
+ * call-seq:
254
+ * context.close -> nil
255
+ *
256
+ * Closes the underlying input streams. This is useful when parsing a large amount of
257
+ * files and you want to close the files without relying on Ruby's garbage collector
258
+ * to run.
259
+ */
260
+ static VALUE rxml_html_parser_context_close(VALUE self)
261
+ {
262
+ htmlParserCtxtPtr ctxt;
263
+ xmlParserInputPtr xinput;
264
+ Data_Get_Struct(self, htmlParserCtxt, ctxt);
265
+
266
+ while ((xinput = inputPop(ctxt)) != NULL)
267
+ {
268
+ xmlFreeInputStream(xinput);
269
+ }
270
+ return Qnil;
271
+ }
272
+
273
+ /*
274
+ * call-seq:
275
+ * context.disable_cdata = (true|false)
276
+ *
277
+ * Control whether the CDATA nodes will be created in this context.
278
+ */
279
+ static VALUE rxml_html_parser_context_disable_cdata_set(VALUE self, VALUE value)
280
+ {
281
+ htmlParserCtxtPtr ctxt;
282
+ Data_Get_Struct(self, htmlParserCtxt, ctxt);
283
+
284
+ if (ctxt->sax == NULL)
285
+ rb_raise(rb_eRuntimeError, "Sax handler is not yet set");
286
+
287
+ /* LibXML controls this internally with the default SAX handler. */
288
+ if (value)
289
+ ctxt->sax->cdataBlock = NULL;
290
+ else
291
+ ctxt->sax->cdataBlock = htmlDefaultSAXHandler.cdataBlock;
292
+
293
+ return value;
294
+ }
295
+
296
+ /*
297
+ * call-seq:
298
+ * context.options = XML::Parser::Options::NOENT |
299
+ XML::Parser::Options::NOCDATA
300
+ *
301
+ * Provides control over the execution of a parser. Valid values
302
+ * are the constants defined on XML::Parser::Options. Multiple
303
+ * options can be combined by using Bitwise OR (|).
304
+ */
305
+ static VALUE rxml_html_parser_context_options_set(VALUE self, VALUE options)
306
+ {
307
+ int xml_options = NUM2INT(options);
308
+ htmlParserCtxtPtr ctxt;
309
+ Check_Type(options, T_FIXNUM);
310
+
311
+ Data_Get_Struct(self, htmlParserCtxt, ctxt);
312
+ htmlCtxtUseOptions(ctxt, xml_options);
313
+
314
+ #if LIBXML_VERSION >= 20707
315
+ /* Big hack here, but htmlCtxtUseOptions doens't support HTML_PARSE_NOIMPLIED.
316
+ So do it ourselves. There must be a better way??? */
317
+ if (xml_options & HTML_PARSE_NOIMPLIED)
318
+ {
319
+ ctxt->options |= HTML_PARSE_NOIMPLIED;
320
+ }
321
+ #endif
322
+
323
+ return self;
324
+ }
325
+
326
+ void rxml_init_html_parser_context(void)
327
+ {
328
+ IO_ATTR = ID2SYM(rb_intern("@io"));
329
+ cXMLHtmlParserContext = rb_define_class_under(cXMLHtmlParser, "Context", cXMLParserContext);
330
+
331
+ rb_define_singleton_method(cXMLHtmlParserContext, "file", rxml_html_parser_context_file, 1);
332
+ rb_define_singleton_method(cXMLHtmlParserContext, "io", rxml_html_parser_context_io, 1);
333
+ rb_define_singleton_method(cXMLHtmlParserContext, "string", rxml_html_parser_context_string, 1);
334
+ rb_define_method(cXMLHtmlParserContext, "close", rxml_html_parser_context_close, 0);
335
+ rb_define_method(cXMLHtmlParserContext, "disable_cdata=", rxml_html_parser_context_disable_cdata_set, 1);
336
+ rb_define_method(cXMLHtmlParserContext, "options=", rxml_html_parser_context_options_set, 1);
337
+ }