libxml-ruby 0.9.3-x86-mswin32-60 → 0.9.4-x86-mswin32-60
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.
- data/CHANGES +9 -0
- data/README +61 -128
- data/doc/css/normal.css +182 -0
- data/doc/img/raze-tiny.png +0 -0
- data/doc/img/red-cube.jpg +0 -0
- data/doc/img/xml-ruby.png +0 -0
- data/doc/index.xml +43 -0
- data/doc/install.xml +77 -0
- data/doc/layout.rhtml +38 -0
- data/doc/layout.xsl +67 -0
- data/doc/license.xml +32 -0
- data/doc/log/changelog.xml +1324 -0
- data/doc/log/changelog.xsl +42 -0
- data/ext/libxml/ruby_xml_document.c +1084 -1057
- data/ext/libxml/ruby_xml_html_parser.c +37 -40
- data/ext/libxml/ruby_xml_input.c +17 -40
- data/ext/libxml/ruby_xml_input.h +2 -2
- data/ext/libxml/ruby_xml_parser.c +151 -151
- data/ext/libxml/ruby_xml_reader.c +910 -893
- data/ext/libxml/ruby_xml_sax_parser.c +174 -174
- data/ext/libxml/ruby_xml_sax_parser.h +12 -12
- data/ext/libxml/ruby_xml_xpointer.h +13 -25
- data/ext/libxml/version.h +2 -2
- data/ext/mingw/libxml_ruby.dll.a +0 -0
- data/ext/mingw/libxml_ruby.so +0 -0
- data/ext/vc/libxml_ruby.vcproj +1 -1
- data/test/model/ruby-lang.html +238 -0
- data/test/tc_html_parser.rb +2 -12
- data/test/tc_reader.rb +87 -87
- metadata +17 -3
- data/test/test.rb +0 -8
@@ -1,893 +1,910 @@
|
|
1
|
-
/* Copyright (c) 2006-2007 Apple Inc.
|
2
|
-
* Please see the LICENSE file for copyright and distribution information. */
|
3
|
-
|
4
|
-
#include "ruby_libxml.h"
|
5
|
-
#include "ruby_xml_reader.h"
|
6
|
-
|
7
|
-
VALUE cXMLReader;
|
8
|
-
|
9
|
-
/*
|
10
|
-
* Document-class: LibXML::XML::Reader
|
11
|
-
*
|
12
|
-
* The XML::Reader class provides a simpler, alternative way of parsing an XML
|
13
|
-
* document in contrast to XML::Parser or XML::SaxParser. A XML::Reader instance
|
14
|
-
* acts like a cursor going forward in a document stream, stopping at each node
|
15
|
-
* it encounters. To advance to the next node, simply cadd XML::Reader#read.
|
16
|
-
*
|
17
|
-
* The XML::Reader API closely matches the DOM Core specification and supports
|
18
|
-
* namespaces, xml:base, entity handling and DTDs.
|
19
|
-
*
|
20
|
-
* To summarize, XML::Reader provides a far simpler API to use versus XML::SaxParser
|
21
|
-
* and is more memory efficient than using XML::Parser to create a DOM tree.
|
22
|
-
*
|
23
|
-
* Example:
|
24
|
-
*
|
25
|
-
* parser = XML::Reader.new("<foo><bar>1</bar><bar>2</bar><bar>3</bar></foo>")
|
26
|
-
* parser.read
|
27
|
-
* assert_equal('foo', parser.name)
|
28
|
-
* assert_equal(nil, parser.value)
|
29
|
-
*
|
30
|
-
* 3.times do |i|
|
31
|
-
* parser.read
|
32
|
-
* assert_equal(XML::Reader::TYPE_ELEMENT, parser.node_type)
|
33
|
-
* assert_equal('bar', parser.name)
|
34
|
-
* parser.read
|
35
|
-
* assert_equal(XML::Reader::TYPE_TEXT, parser.node_type)
|
36
|
-
* assert_equal((i + 1).to_s, parser.value)
|
37
|
-
* parser.read
|
38
|
-
* assert_equal(XML::Reader::TYPE_END_ELEMENT, parser.node_type)
|
39
|
-
* end
|
40
|
-
*
|
41
|
-
* For a more in depth tutorial, albeit in C, see http://xmlsoft.org/xmlreader.html.*/
|
42
|
-
|
43
|
-
|
44
|
-
|
45
|
-
|
46
|
-
|
47
|
-
|
48
|
-
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
53
|
-
|
54
|
-
|
55
|
-
|
56
|
-
|
57
|
-
|
58
|
-
|
59
|
-
|
60
|
-
|
61
|
-
|
62
|
-
|
63
|
-
|
64
|
-
|
65
|
-
|
66
|
-
|
67
|
-
|
68
|
-
|
69
|
-
|
70
|
-
|
71
|
-
|
72
|
-
|
73
|
-
|
74
|
-
|
75
|
-
|
76
|
-
|
77
|
-
|
78
|
-
|
79
|
-
|
80
|
-
|
81
|
-
|
82
|
-
|
83
|
-
|
84
|
-
|
85
|
-
|
86
|
-
|
87
|
-
|
88
|
-
|
89
|
-
|
90
|
-
|
91
|
-
|
92
|
-
|
93
|
-
|
94
|
-
|
95
|
-
|
96
|
-
|
97
|
-
|
98
|
-
|
99
|
-
|
100
|
-
|
101
|
-
*
|
102
|
-
|
103
|
-
|
104
|
-
|
105
|
-
|
106
|
-
|
107
|
-
|
108
|
-
|
109
|
-
|
110
|
-
|
111
|
-
|
112
|
-
|
113
|
-
|
114
|
-
|
115
|
-
|
116
|
-
|
117
|
-
|
118
|
-
|
119
|
-
|
120
|
-
|
121
|
-
|
122
|
-
|
123
|
-
|
124
|
-
|
125
|
-
|
126
|
-
|
127
|
-
|
128
|
-
|
129
|
-
|
130
|
-
|
131
|
-
|
132
|
-
|
133
|
-
|
134
|
-
|
135
|
-
|
136
|
-
|
137
|
-
|
138
|
-
|
139
|
-
|
140
|
-
|
141
|
-
|
142
|
-
|
143
|
-
|
144
|
-
|
145
|
-
|
146
|
-
|
147
|
-
|
148
|
-
*
|
149
|
-
*
|
150
|
-
|
151
|
-
|
152
|
-
|
153
|
-
|
154
|
-
|
155
|
-
|
156
|
-
|
157
|
-
|
158
|
-
|
159
|
-
|
160
|
-
|
161
|
-
|
162
|
-
|
163
|
-
|
164
|
-
|
165
|
-
|
166
|
-
|
167
|
-
|
168
|
-
|
169
|
-
|
170
|
-
|
171
|
-
|
172
|
-
|
173
|
-
|
174
|
-
|
175
|
-
|
176
|
-
|
177
|
-
|
178
|
-
*
|
179
|
-
*
|
180
|
-
*
|
181
|
-
*
|
182
|
-
|
183
|
-
|
184
|
-
|
185
|
-
|
186
|
-
|
187
|
-
|
188
|
-
|
189
|
-
|
190
|
-
|
191
|
-
*
|
192
|
-
*
|
193
|
-
*
|
194
|
-
*
|
195
|
-
*
|
196
|
-
|
197
|
-
|
198
|
-
|
199
|
-
|
200
|
-
|
201
|
-
|
202
|
-
|
203
|
-
|
204
|
-
|
205
|
-
|
206
|
-
|
207
|
-
|
208
|
-
|
209
|
-
|
210
|
-
|
211
|
-
|
212
|
-
|
213
|
-
|
214
|
-
|
215
|
-
|
216
|
-
|
217
|
-
*
|
218
|
-
*
|
219
|
-
*
|
220
|
-
*
|
221
|
-
|
222
|
-
|
223
|
-
|
224
|
-
|
225
|
-
|
226
|
-
|
227
|
-
|
228
|
-
|
229
|
-
|
230
|
-
*
|
231
|
-
*
|
232
|
-
*
|
233
|
-
*
|
234
|
-
|
235
|
-
|
236
|
-
|
237
|
-
|
238
|
-
|
239
|
-
|
240
|
-
|
241
|
-
|
242
|
-
|
243
|
-
*
|
244
|
-
*
|
245
|
-
*
|
246
|
-
* current
|
247
|
-
|
248
|
-
|
249
|
-
|
250
|
-
|
251
|
-
|
252
|
-
|
253
|
-
|
254
|
-
|
255
|
-
|
256
|
-
*
|
257
|
-
*
|
258
|
-
*
|
259
|
-
* the
|
260
|
-
|
261
|
-
|
262
|
-
|
263
|
-
|
264
|
-
|
265
|
-
|
266
|
-
|
267
|
-
|
268
|
-
|
269
|
-
*
|
270
|
-
*
|
271
|
-
*
|
272
|
-
* the
|
273
|
-
*
|
274
|
-
|
275
|
-
|
276
|
-
|
277
|
-
|
278
|
-
|
279
|
-
|
280
|
-
|
281
|
-
|
282
|
-
|
283
|
-
*
|
284
|
-
*
|
285
|
-
*
|
286
|
-
*
|
287
|
-
|
288
|
-
|
289
|
-
|
290
|
-
|
291
|
-
|
292
|
-
|
293
|
-
|
294
|
-
|
295
|
-
|
296
|
-
*
|
297
|
-
*
|
298
|
-
*
|
299
|
-
*
|
300
|
-
*
|
301
|
-
*
|
302
|
-
*
|
303
|
-
*
|
304
|
-
*
|
305
|
-
|
306
|
-
|
307
|
-
|
308
|
-
|
309
|
-
|
310
|
-
|
311
|
-
|
312
|
-
|
313
|
-
|
314
|
-
*
|
315
|
-
*
|
316
|
-
*
|
317
|
-
*
|
318
|
-
*
|
319
|
-
*
|
320
|
-
* read,
|
321
|
-
|
322
|
-
|
323
|
-
|
324
|
-
|
325
|
-
|
326
|
-
|
327
|
-
|
328
|
-
|
329
|
-
|
330
|
-
*
|
331
|
-
*
|
332
|
-
*
|
333
|
-
*
|
334
|
-
*
|
335
|
-
*
|
336
|
-
*
|
337
|
-
|
338
|
-
|
339
|
-
|
340
|
-
|
341
|
-
|
342
|
-
|
343
|
-
|
344
|
-
|
345
|
-
|
346
|
-
*
|
347
|
-
*
|
348
|
-
*
|
349
|
-
*
|
350
|
-
*
|
351
|
-
*
|
352
|
-
|
353
|
-
|
354
|
-
|
355
|
-
|
356
|
-
|
357
|
-
|
358
|
-
|
359
|
-
|
360
|
-
|
361
|
-
|
362
|
-
*
|
363
|
-
*
|
364
|
-
*
|
365
|
-
*
|
366
|
-
*
|
367
|
-
|
368
|
-
|
369
|
-
|
370
|
-
|
371
|
-
|
372
|
-
|
373
|
-
|
374
|
-
|
375
|
-
|
376
|
-
|
377
|
-
|
378
|
-
*
|
379
|
-
|
380
|
-
|
381
|
-
|
382
|
-
|
383
|
-
|
384
|
-
|
385
|
-
|
386
|
-
|
387
|
-
|
388
|
-
|
389
|
-
|
390
|
-
*
|
391
|
-
*
|
392
|
-
*
|
393
|
-
*
|
394
|
-
|
395
|
-
|
396
|
-
|
397
|
-
|
398
|
-
|
399
|
-
|
400
|
-
|
401
|
-
|
402
|
-
|
403
|
-
|
404
|
-
|
405
|
-
|
406
|
-
*
|
407
|
-
*
|
408
|
-
*
|
409
|
-
*
|
410
|
-
*
|
411
|
-
|
412
|
-
|
413
|
-
|
414
|
-
|
415
|
-
|
416
|
-
|
417
|
-
|
418
|
-
|
419
|
-
|
420
|
-
|
421
|
-
|
422
|
-
|
423
|
-
|
424
|
-
|
425
|
-
*
|
426
|
-
*
|
427
|
-
*
|
428
|
-
*
|
429
|
-
|
430
|
-
|
431
|
-
|
432
|
-
|
433
|
-
|
434
|
-
|
435
|
-
|
436
|
-
|
437
|
-
|
438
|
-
*
|
439
|
-
|
440
|
-
|
441
|
-
|
442
|
-
|
443
|
-
|
444
|
-
|
445
|
-
|
446
|
-
|
447
|
-
|
448
|
-
|
449
|
-
|
450
|
-
|
451
|
-
|
452
|
-
|
453
|
-
*
|
454
|
-
|
455
|
-
|
456
|
-
|
457
|
-
|
458
|
-
|
459
|
-
|
460
|
-
|
461
|
-
|
462
|
-
|
463
|
-
|
464
|
-
|
465
|
-
|
466
|
-
|
467
|
-
|
468
|
-
|
469
|
-
|
470
|
-
|
471
|
-
|
472
|
-
|
473
|
-
|
474
|
-
*
|
475
|
-
|
476
|
-
|
477
|
-
|
478
|
-
|
479
|
-
|
480
|
-
|
481
|
-
|
482
|
-
|
483
|
-
|
484
|
-
|
485
|
-
|
486
|
-
*
|
487
|
-
|
488
|
-
|
489
|
-
|
490
|
-
|
491
|
-
|
492
|
-
|
493
|
-
|
494
|
-
|
495
|
-
|
496
|
-
|
497
|
-
|
498
|
-
*
|
499
|
-
*
|
500
|
-
|
501
|
-
|
502
|
-
|
503
|
-
|
504
|
-
|
505
|
-
|
506
|
-
|
507
|
-
|
508
|
-
|
509
|
-
|
510
|
-
*
|
511
|
-
*
|
512
|
-
*
|
513
|
-
|
514
|
-
|
515
|
-
|
516
|
-
|
517
|
-
|
518
|
-
return
|
519
|
-
}
|
520
|
-
|
521
|
-
/*
|
522
|
-
* call-seq:
|
523
|
-
* reader.
|
524
|
-
*
|
525
|
-
*
|
526
|
-
*/
|
527
|
-
static VALUE
|
528
|
-
|
529
|
-
{
|
530
|
-
|
531
|
-
|
532
|
-
|
533
|
-
|
534
|
-
|
535
|
-
*
|
536
|
-
*
|
537
|
-
*
|
538
|
-
|
539
|
-
|
540
|
-
|
541
|
-
|
542
|
-
|
543
|
-
|
544
|
-
|
545
|
-
|
546
|
-
|
547
|
-
|
548
|
-
*
|
549
|
-
*
|
550
|
-
*
|
551
|
-
|
552
|
-
|
553
|
-
|
554
|
-
|
555
|
-
|
556
|
-
|
557
|
-
|
558
|
-
|
559
|
-
|
560
|
-
*
|
561
|
-
*
|
562
|
-
*
|
563
|
-
*
|
564
|
-
*
|
565
|
-
|
566
|
-
|
567
|
-
|
568
|
-
|
569
|
-
|
570
|
-
|
571
|
-
|
572
|
-
|
573
|
-
|
574
|
-
|
575
|
-
*
|
576
|
-
*
|
577
|
-
*
|
578
|
-
*
|
579
|
-
|
580
|
-
|
581
|
-
|
582
|
-
|
583
|
-
|
584
|
-
|
585
|
-
|
586
|
-
|
587
|
-
|
588
|
-
|
589
|
-
*
|
590
|
-
*
|
591
|
-
|
592
|
-
|
593
|
-
|
594
|
-
|
595
|
-
|
596
|
-
|
597
|
-
|
598
|
-
|
599
|
-
|
600
|
-
|
601
|
-
|
602
|
-
*
|
603
|
-
|
604
|
-
|
605
|
-
|
606
|
-
|
607
|
-
|
608
|
-
|
609
|
-
|
610
|
-
|
611
|
-
|
612
|
-
|
613
|
-
|
614
|
-
|
615
|
-
|
616
|
-
|
617
|
-
|
618
|
-
|
619
|
-
|
620
|
-
|
621
|
-
|
622
|
-
|
623
|
-
|
624
|
-
|
625
|
-
|
626
|
-
|
627
|
-
*
|
628
|
-
*
|
629
|
-
|
630
|
-
|
631
|
-
|
632
|
-
|
633
|
-
|
634
|
-
|
635
|
-
|
636
|
-
|
637
|
-
|
638
|
-
|
639
|
-
|
640
|
-
|
641
|
-
|
642
|
-
|
643
|
-
|
644
|
-
|
645
|
-
|
646
|
-
|
647
|
-
|
648
|
-
|
649
|
-
|
650
|
-
*
|
651
|
-
|
652
|
-
|
653
|
-
|
654
|
-
|
655
|
-
|
656
|
-
|
657
|
-
|
658
|
-
|
659
|
-
|
660
|
-
|
661
|
-
|
662
|
-
|
663
|
-
|
664
|
-
*
|
665
|
-
*
|
666
|
-
*
|
667
|
-
*
|
668
|
-
|
669
|
-
|
670
|
-
|
671
|
-
|
672
|
-
|
673
|
-
|
674
|
-
|
675
|
-
|
676
|
-
|
677
|
-
|
678
|
-
|
679
|
-
|
680
|
-
|
681
|
-
|
682
|
-
|
683
|
-
|
684
|
-
|
685
|
-
|
686
|
-
|
687
|
-
|
688
|
-
|
689
|
-
|
690
|
-
|
691
|
-
|
692
|
-
|
693
|
-
|
694
|
-
|
695
|
-
|
696
|
-
|
697
|
-
|
698
|
-
|
699
|
-
|
700
|
-
|
701
|
-
|
702
|
-
|
703
|
-
|
704
|
-
|
705
|
-
|
706
|
-
|
707
|
-
|
708
|
-
|
709
|
-
|
710
|
-
|
711
|
-
|
712
|
-
*
|
713
|
-
*
|
714
|
-
|
715
|
-
|
716
|
-
|
717
|
-
|
718
|
-
|
719
|
-
|
720
|
-
|
721
|
-
|
722
|
-
|
723
|
-
|
724
|
-
|
725
|
-
|
726
|
-
|
727
|
-
|
728
|
-
|
729
|
-
|
730
|
-
|
731
|
-
|
732
|
-
|
733
|
-
|
734
|
-
|
735
|
-
|
736
|
-
|
737
|
-
|
738
|
-
|
739
|
-
*
|
740
|
-
|
741
|
-
|
742
|
-
|
743
|
-
|
744
|
-
|
745
|
-
|
746
|
-
|
747
|
-
|
748
|
-
|
749
|
-
|
750
|
-
|
751
|
-
|
752
|
-
*
|
753
|
-
|
754
|
-
|
755
|
-
|
756
|
-
|
757
|
-
|
758
|
-
|
759
|
-
|
760
|
-
|
761
|
-
|
762
|
-
|
763
|
-
|
764
|
-
|
765
|
-
|
766
|
-
|
767
|
-
|
768
|
-
|
769
|
-
|
770
|
-
|
771
|
-
|
772
|
-
|
773
|
-
|
774
|
-
|
775
|
-
|
776
|
-
|
777
|
-
|
778
|
-
|
779
|
-
|
780
|
-
|
781
|
-
|
782
|
-
|
783
|
-
|
784
|
-
|
785
|
-
|
786
|
-
|
787
|
-
|
788
|
-
|
789
|
-
|
790
|
-
|
791
|
-
|
792
|
-
|
793
|
-
|
794
|
-
|
795
|
-
|
796
|
-
|
797
|
-
|
798
|
-
|
799
|
-
|
800
|
-
|
801
|
-
|
802
|
-
|
803
|
-
|
804
|
-
|
805
|
-
|
806
|
-
|
807
|
-
|
808
|
-
|
809
|
-
|
810
|
-
|
811
|
-
|
812
|
-
|
813
|
-
|
814
|
-
|
815
|
-
|
816
|
-
|
817
|
-
|
818
|
-
|
819
|
-
rb_define_method(cXMLReader, "
|
820
|
-
|
821
|
-
|
822
|
-
rb_define_method(cXMLReader, "
|
823
|
-
rb_define_method(cXMLReader, "
|
824
|
-
rb_define_method(cXMLReader, "
|
825
|
-
rb_define_method(cXMLReader, "
|
826
|
-
rb_define_method(cXMLReader, "
|
827
|
-
rb_define_method(cXMLReader, "
|
828
|
-
rb_define_method(cXMLReader, "
|
829
|
-
rb_define_method(cXMLReader, "
|
830
|
-
rb_define_method(cXMLReader, "
|
831
|
-
rb_define_method(cXMLReader, "
|
832
|
-
rb_define_method(cXMLReader, "
|
833
|
-
|
834
|
-
rb_define_method(cXMLReader, "
|
835
|
-
|
836
|
-
|
837
|
-
|
838
|
-
|
839
|
-
rb_define_method(cXMLReader, "
|
840
|
-
rb_define_method(cXMLReader, "
|
841
|
-
|
842
|
-
rb_define_method(cXMLReader, "
|
843
|
-
rb_define_method(cXMLReader, "
|
844
|
-
|
845
|
-
|
846
|
-
rb_define_method(cXMLReader, "
|
847
|
-
|
848
|
-
|
849
|
-
rb_define_method(cXMLReader, "
|
850
|
-
rb_define_method(cXMLReader, "
|
851
|
-
|
852
|
-
rb_define_method(cXMLReader, "
|
853
|
-
|
854
|
-
rb_define_method(cXMLReader, "
|
855
|
-
rb_define_method(cXMLReader, "
|
856
|
-
|
857
|
-
|
858
|
-
|
859
|
-
|
860
|
-
|
861
|
-
|
862
|
-
|
863
|
-
|
864
|
-
|
865
|
-
|
866
|
-
|
867
|
-
|
868
|
-
|
869
|
-
|
870
|
-
|
871
|
-
|
872
|
-
|
873
|
-
|
874
|
-
rb_define_const(cXMLReader, "
|
875
|
-
rb_define_const(cXMLReader, "
|
876
|
-
rb_define_const(cXMLReader, "
|
877
|
-
rb_define_const(cXMLReader, "
|
878
|
-
|
879
|
-
rb_define_const(cXMLReader, "
|
880
|
-
rb_define_const(cXMLReader, "
|
881
|
-
rb_define_const(cXMLReader, "
|
882
|
-
rb_define_const(cXMLReader, "
|
883
|
-
|
884
|
-
rb_define_const(cXMLReader, "
|
885
|
-
|
886
|
-
|
887
|
-
rb_define_const(cXMLReader, "
|
888
|
-
rb_define_const(cXMLReader, "
|
889
|
-
rb_define_const(cXMLReader, "
|
890
|
-
rb_define_const(cXMLReader, "
|
891
|
-
rb_define_const(cXMLReader, "
|
892
|
-
rb_define_const(cXMLReader, "
|
893
|
-
|
1
|
+
/* Copyright (c) 2006-2007 Apple Inc.
|
2
|
+
* Please see the LICENSE file for copyright and distribution information. */
|
3
|
+
|
4
|
+
#include "ruby_libxml.h"
|
5
|
+
#include "ruby_xml_reader.h"
|
6
|
+
|
7
|
+
VALUE cXMLReader;
|
8
|
+
|
9
|
+
/*
|
10
|
+
* Document-class: LibXML::XML::Reader
|
11
|
+
*
|
12
|
+
* The XML::Reader class provides a simpler, alternative way of parsing an XML
|
13
|
+
* document in contrast to XML::Parser or XML::SaxParser. A XML::Reader instance
|
14
|
+
* acts like a cursor going forward in a document stream, stopping at each node
|
15
|
+
* it encounters. To advance to the next node, simply cadd XML::Reader#read.
|
16
|
+
*
|
17
|
+
* The XML::Reader API closely matches the DOM Core specification and supports
|
18
|
+
* namespaces, xml:base, entity handling and DTDs.
|
19
|
+
*
|
20
|
+
* To summarize, XML::Reader provides a far simpler API to use versus XML::SaxParser
|
21
|
+
* and is more memory efficient than using XML::Parser to create a DOM tree.
|
22
|
+
*
|
23
|
+
* Example:
|
24
|
+
*
|
25
|
+
* parser = XML::Reader.new("<foo><bar>1</bar><bar>2</bar><bar>3</bar></foo>")
|
26
|
+
* parser.read
|
27
|
+
* assert_equal('foo', parser.name)
|
28
|
+
* assert_equal(nil, parser.value)
|
29
|
+
*
|
30
|
+
* 3.times do |i|
|
31
|
+
* parser.read
|
32
|
+
* assert_equal(XML::Reader::TYPE_ELEMENT, parser.node_type)
|
33
|
+
* assert_equal('bar', parser.name)
|
34
|
+
* parser.read
|
35
|
+
* assert_equal(XML::Reader::TYPE_TEXT, parser.node_type)
|
36
|
+
* assert_equal((i + 1).to_s, parser.value)
|
37
|
+
* parser.read
|
38
|
+
* assert_equal(XML::Reader::TYPE_END_ELEMENT, parser.node_type)
|
39
|
+
* end
|
40
|
+
*
|
41
|
+
* For a more in depth tutorial, albeit in C, see http://xmlsoft.org/xmlreader.html.*/
|
42
|
+
|
43
|
+
|
44
|
+
static VALUE
|
45
|
+
rxml_reader_new(VALUE class, xmlTextReaderPtr reader)
|
46
|
+
{
|
47
|
+
return Data_Wrap_Struct(class, NULL, xmlFreeTextReader, reader);
|
48
|
+
}
|
49
|
+
|
50
|
+
static xmlTextReaderPtr
|
51
|
+
rxml_text_reader_get(VALUE obj)
|
52
|
+
{
|
53
|
+
xmlTextReaderPtr xreader;
|
54
|
+
Data_Get_Struct(obj, xmlTextReader, xreader);
|
55
|
+
return xreader;
|
56
|
+
}
|
57
|
+
|
58
|
+
/*
|
59
|
+
* call-seq:
|
60
|
+
* XML::Reader.file(path, encoding=nil, options=0) -> reader
|
61
|
+
*
|
62
|
+
* Parse an XML file from the filesystem or the network. The parsing flags
|
63
|
+
* options are a combination of xmlParserOption.
|
64
|
+
*/
|
65
|
+
static VALUE
|
66
|
+
rxml_reader_new_file(int argc, VALUE *argv, VALUE self)
|
67
|
+
{
|
68
|
+
xmlTextReaderPtr xreader;
|
69
|
+
VALUE rpath, rencoding, roptions;
|
70
|
+
char *xpath;
|
71
|
+
char *xencoding;
|
72
|
+
int options;
|
73
|
+
|
74
|
+
rb_scan_args(argc, argv, "12", &rpath, &rencoding, &roptions);
|
75
|
+
|
76
|
+
xpath = NIL_P(rpath) ? NULL : StringValueCStr(rpath);
|
77
|
+
xencoding = NIL_P(rencoding) ? NULL : StringValueCStr(rencoding);
|
78
|
+
options = NIL_P(roptions) ? 0 : FIX2INT(roptions);
|
79
|
+
|
80
|
+
xreader = xmlReaderForFile(xpath, xencoding, options);
|
81
|
+
|
82
|
+
if (xreader == NULL)
|
83
|
+
rxml_raise(&xmlLastError);
|
84
|
+
|
85
|
+
return rxml_reader_new(self, xreader);
|
86
|
+
}
|
87
|
+
|
88
|
+
/*
|
89
|
+
* call-seq:
|
90
|
+
* XML::Reader.io(io, url=nil, encoding=nil, options=0) -> reader
|
91
|
+
*
|
92
|
+
* Parse an XML file from a file handle. The parsing flags options are
|
93
|
+
* a combination of xmlParserOption.
|
94
|
+
*/
|
95
|
+
static VALUE
|
96
|
+
rxml_reader_new_io(int argc, VALUE *argv, VALUE self)
|
97
|
+
{
|
98
|
+
xmlTextReaderPtr xreader;
|
99
|
+
VALUE rio, rurl, rencoding, roptions;
|
100
|
+
char *xurl;
|
101
|
+
char *xencoding;
|
102
|
+
int options;
|
103
|
+
|
104
|
+
rb_scan_args(argc, argv, "13", &rio, &rurl, &rencoding, &roptions);
|
105
|
+
|
106
|
+
xurl = NIL_P(rurl) ? NULL : StringValueCStr(rurl);
|
107
|
+
xencoding = NIL_P(rencoding) ? NULL : StringValueCStr(rencoding);
|
108
|
+
options = NIL_P(roptions) ? 0 : FIX2INT(roptions);
|
109
|
+
|
110
|
+
xreader = xmlReaderForIO((xmlInputReadCallback) rxml_read_callback, NULL,
|
111
|
+
(void *)rio,
|
112
|
+
xurl, xencoding, options);
|
113
|
+
|
114
|
+
if (xreader == NULL)
|
115
|
+
rxml_raise(&xmlLastError);
|
116
|
+
|
117
|
+
return rxml_reader_new(self, xreader);
|
118
|
+
}
|
119
|
+
|
120
|
+
/*
|
121
|
+
* call-seq:
|
122
|
+
* XML::Reader.walker(doc) -> reader
|
123
|
+
* XML::Reader.document(doc) -> reader
|
124
|
+
*
|
125
|
+
* Create an XML text reader for a preparsed document.
|
126
|
+
*/
|
127
|
+
VALUE
|
128
|
+
rxml_reader_new_walker(VALUE self, VALUE doc)
|
129
|
+
{
|
130
|
+
xmlDocPtr xdoc;
|
131
|
+
xmlTextReaderPtr xreader;
|
132
|
+
|
133
|
+
Data_Get_Struct(doc, xmlDoc, xdoc);
|
134
|
+
|
135
|
+
xreader = xmlReaderWalker(xdoc);
|
136
|
+
|
137
|
+
if (xreader == NULL)
|
138
|
+
rxml_raise(&xmlLastError);
|
139
|
+
|
140
|
+
return rxml_reader_new(self, xreader);
|
141
|
+
}
|
142
|
+
|
143
|
+
/*
|
144
|
+
* call-seq:
|
145
|
+
* XML::Reader.new(data, url=nil, encoding=nil, options=0) -> reader
|
146
|
+
* XML::Reader.string(data, url=nil, encoding=nil, options=0) -> reader
|
147
|
+
*
|
148
|
+
* Create an XML text reader for an XML in-memory document. The parsing flags
|
149
|
+
* options are a combination of xmlParserOption.
|
150
|
+
*/
|
151
|
+
static VALUE
|
152
|
+
rxml_reader_new_data(int argc, VALUE *argv, VALUE self)
|
153
|
+
{
|
154
|
+
xmlTextReaderPtr xreader;
|
155
|
+
VALUE rdata, rurl, rencoding, roptions;
|
156
|
+
char *xdata;
|
157
|
+
char *xurl;
|
158
|
+
char *xencoding;
|
159
|
+
int options;
|
160
|
+
|
161
|
+
rb_scan_args(argc, argv, "13", &rdata, &rurl, &rencoding, &roptions);
|
162
|
+
|
163
|
+
xdata = NIL_P(rdata) ? NULL : StringValueCStr(rdata);
|
164
|
+
xurl = NIL_P(rurl) ? NULL : StringValueCStr(rurl);
|
165
|
+
xencoding = NIL_P(rencoding) ? NULL : StringValueCStr(rencoding);
|
166
|
+
options = NIL_P(roptions) ? 0 : FIX2INT(roptions);
|
167
|
+
|
168
|
+
xreader = xmlReaderForMemory(xdata, strlen(xdata),
|
169
|
+
xurl, xencoding, options);
|
170
|
+
|
171
|
+
if (xreader == NULL)
|
172
|
+
rxml_raise(&xmlLastError);
|
173
|
+
|
174
|
+
return rxml_reader_new(self, xreader);
|
175
|
+
}
|
176
|
+
|
177
|
+
/*
|
178
|
+
* call-seq:
|
179
|
+
* parser.close -> code
|
180
|
+
*
|
181
|
+
* This method releases any resources allocated by the current instance
|
182
|
+
* changes the state to Closed and close any underlying input.
|
183
|
+
*/
|
184
|
+
static VALUE
|
185
|
+
rxml_reader_close(VALUE self)
|
186
|
+
{
|
187
|
+
return INT2FIX(xmlTextReaderClose(rxml_text_reader_get(self)));
|
188
|
+
}
|
189
|
+
|
190
|
+
/*
|
191
|
+
* call-seq:
|
192
|
+
* parser.move_to_attribute(val) -> code
|
193
|
+
*
|
194
|
+
* Move the position of the current instance to the attribute with the
|
195
|
+
* specified index (if +val+ is an integer) or name (if +val+ is a string)
|
196
|
+
* relative to the containing element.
|
197
|
+
*/
|
198
|
+
static VALUE
|
199
|
+
rxml_reader_move_to_attr(VALUE self, VALUE val)
|
200
|
+
{
|
201
|
+
xmlTextReaderPtr xreader;
|
202
|
+
int ret;
|
203
|
+
|
204
|
+
xreader = rxml_text_reader_get(self);
|
205
|
+
|
206
|
+
if (TYPE(val) == T_FIXNUM) {
|
207
|
+
ret = xmlTextReaderMoveToAttributeNo(xreader, FIX2INT(val));
|
208
|
+
}
|
209
|
+
else {
|
210
|
+
ret = xmlTextReaderMoveToAttribute(xreader, (const xmlChar *)StringValueCStr(val));
|
211
|
+
}
|
212
|
+
|
213
|
+
return INT2FIX(ret);
|
214
|
+
}
|
215
|
+
|
216
|
+
/*
|
217
|
+
* call-seq:
|
218
|
+
* reader.move_to_first_attribute -> code
|
219
|
+
*
|
220
|
+
* Move the position of the current instance to the first attribute associated
|
221
|
+
* with the current node.
|
222
|
+
*/
|
223
|
+
static VALUE
|
224
|
+
rxml_reader_move_to_first_attr(VALUE self)
|
225
|
+
{
|
226
|
+
return INT2FIX(xmlTextReaderMoveToFirstAttribute(rxml_text_reader_get(self)));
|
227
|
+
}
|
228
|
+
|
229
|
+
/*
|
230
|
+
* call-seq:
|
231
|
+
* reader.move_to_next_attribute -> code
|
232
|
+
*
|
233
|
+
* Move the position of the current instance to the next attribute associated
|
234
|
+
* with the current node.
|
235
|
+
*/
|
236
|
+
static VALUE
|
237
|
+
rxml_reader_move_to_next_attr(VALUE self)
|
238
|
+
{
|
239
|
+
return INT2FIX(xmlTextReaderMoveToNextAttribute(rxml_text_reader_get(self)));
|
240
|
+
}
|
241
|
+
|
242
|
+
/*
|
243
|
+
* call-seq:
|
244
|
+
* reader.move_to_element -> code
|
245
|
+
*
|
246
|
+
* Move the position of the current instance to the node that contains the
|
247
|
+
* current attribute node.
|
248
|
+
*/
|
249
|
+
static VALUE
|
250
|
+
rxml_reader_move_to_element(VALUE self)
|
251
|
+
{
|
252
|
+
return INT2FIX(xmlTextReaderMoveToElement(rxml_text_reader_get(self)));
|
253
|
+
}
|
254
|
+
|
255
|
+
/*
|
256
|
+
* call-seq:
|
257
|
+
* reader.next -> code
|
258
|
+
*
|
259
|
+
* Skip to the node following the current one in document order while avoiding
|
260
|
+
* the subtree if any.
|
261
|
+
*/
|
262
|
+
static VALUE
|
263
|
+
rxml_reader_next(VALUE self)
|
264
|
+
{
|
265
|
+
return INT2FIX(xmlTextReaderNext(rxml_text_reader_get(self)));
|
266
|
+
}
|
267
|
+
|
268
|
+
/*
|
269
|
+
* call-seq:
|
270
|
+
* reader.next_sibling -> code
|
271
|
+
*
|
272
|
+
* Skip to the node following the current one in document order while avoiding
|
273
|
+
* the subtree if any. Currently implemented only for Readers built on a
|
274
|
+
* document.
|
275
|
+
*/
|
276
|
+
static VALUE
|
277
|
+
rxml_reader_next_sibling(VALUE self)
|
278
|
+
{
|
279
|
+
return INT2FIX(xmlTextReaderNextSibling(rxml_text_reader_get(self)));
|
280
|
+
}
|
281
|
+
|
282
|
+
/*
|
283
|
+
* call-seq:
|
284
|
+
* reader.node_type -> type
|
285
|
+
*
|
286
|
+
* Get the node type of the current node. Reference:
|
287
|
+
* http://dotgnu.org/pnetlib-doc/System/Xml/XmlNodeType.html
|
288
|
+
*/
|
289
|
+
static VALUE
|
290
|
+
rxml_reader_node_type(VALUE self)
|
291
|
+
{
|
292
|
+
return INT2FIX(xmlTextReaderNodeType(rxml_text_reader_get(self)));
|
293
|
+
}
|
294
|
+
|
295
|
+
/*
|
296
|
+
* call-seq:
|
297
|
+
* reader.normalization -> value
|
298
|
+
*
|
299
|
+
* The value indicating whether to normalize white space and attribute values.
|
300
|
+
* Since attribute value and end of line normalizations are a MUST in the XML
|
301
|
+
* specification only the value true is accepted. The broken bahaviour of
|
302
|
+
* accepting out of range character entities like � is of course not
|
303
|
+
* supported either.
|
304
|
+
*
|
305
|
+
* Return 1 or -1 in case of error.
|
306
|
+
*/
|
307
|
+
static VALUE
|
308
|
+
rxml_reader_normalization(VALUE self)
|
309
|
+
{
|
310
|
+
return INT2FIX(xmlTextReaderNormalization(rxml_text_reader_get(self)));
|
311
|
+
}
|
312
|
+
|
313
|
+
/*
|
314
|
+
* call-seq:
|
315
|
+
* reader.read -> code
|
316
|
+
*
|
317
|
+
* Move the position of the current instance to the next node in the stream,
|
318
|
+
* exposing its properties.
|
319
|
+
*
|
320
|
+
* Return 1 if the node was read successfully, 0 if there is no more nodes to
|
321
|
+
* read, or -1 in case of error.
|
322
|
+
*/
|
323
|
+
static VALUE
|
324
|
+
rxml_reader_read(VALUE self)
|
325
|
+
{
|
326
|
+
return INT2FIX(xmlTextReaderRead(rxml_text_reader_get(self)));
|
327
|
+
}
|
328
|
+
|
329
|
+
/*
|
330
|
+
* call-seq:
|
331
|
+
* reader.read_attribute_value -> code
|
332
|
+
*
|
333
|
+
* Parse an attribute value into one or more Text and EntityReference nodes.
|
334
|
+
*
|
335
|
+
* Return 1 in case of success, 0 if the reader was not positionned on an
|
336
|
+
* attribute node or all the attribute values have been read, or -1 in case of
|
337
|
+
* error.
|
338
|
+
*/
|
339
|
+
static VALUE
|
340
|
+
rxml_reader_read_attr_value(VALUE self)
|
341
|
+
{
|
342
|
+
return INT2FIX(xmlTextReaderReadAttributeValue(rxml_text_reader_get(self)));
|
343
|
+
}
|
344
|
+
|
345
|
+
/*
|
346
|
+
* call-seq:
|
347
|
+
* reader.read_inner_xml -> data
|
348
|
+
*
|
349
|
+
* Read the contents of the current node, including child nodes and markup.
|
350
|
+
*
|
351
|
+
* Return a string containing the XML content, or nil if the current node is
|
352
|
+
* neither an element nor attribute, or has no child nodes.
|
353
|
+
*/
|
354
|
+
static VALUE
|
355
|
+
rxml_reader_read_inner_xml(VALUE self)
|
356
|
+
{
|
357
|
+
const xmlChar *result = xmlTextReaderReadInnerXml(rxml_text_reader_get(self));
|
358
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
359
|
+
}
|
360
|
+
|
361
|
+
/*
|
362
|
+
* call-seq:
|
363
|
+
* reader.read_outer_xml -> data
|
364
|
+
*
|
365
|
+
* Read the contents of the current node, including child nodes and markup.
|
366
|
+
*
|
367
|
+
* Return a string containing the XML content, or nil if the current node is
|
368
|
+
* neither an element nor attribute, or has no child nodes.
|
369
|
+
*/
|
370
|
+
static VALUE
|
371
|
+
rxml_reader_read_outer_xml(VALUE self)
|
372
|
+
{
|
373
|
+
const xmlChar *result = xmlTextReaderReadOuterXml(rxml_text_reader_get(self));
|
374
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
375
|
+
}
|
376
|
+
|
377
|
+
/*
|
378
|
+
* call-seq:
|
379
|
+
* reader.read_state -> state
|
380
|
+
*
|
381
|
+
* Get the read state of the reader.
|
382
|
+
*/
|
383
|
+
static VALUE
|
384
|
+
rxml_reader_read_state(VALUE self)
|
385
|
+
{
|
386
|
+
return INT2FIX(xmlTextReaderReadState(rxml_text_reader_get(self)));
|
387
|
+
}
|
388
|
+
|
389
|
+
/*
|
390
|
+
* call-seq:
|
391
|
+
* reader.read_string -> string
|
392
|
+
*
|
393
|
+
* Read the contents of an element or a text node as a string.
|
394
|
+
*
|
395
|
+
* Return a string containing the contents of the Element or Text node, or nil
|
396
|
+
* if the reader is positioned on any other type of node.
|
397
|
+
*/
|
398
|
+
static VALUE
|
399
|
+
rxml_reader_read_string(VALUE self)
|
400
|
+
{
|
401
|
+
const xmlChar *result = xmlTextReaderReadString(rxml_text_reader_get(self));
|
402
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
403
|
+
}
|
404
|
+
|
405
|
+
/*
|
406
|
+
* call-seq:
|
407
|
+
* reader.relax_ng_validate(rng) -> code
|
408
|
+
*
|
409
|
+
* Use RelaxNG to validate the document as it is processed. Activation is only
|
410
|
+
* possible before the first read. If +rng+ is nil, the RelaxNG validation is
|
411
|
+
* desactivated.
|
412
|
+
*
|
413
|
+
* Return 0 in case the RelaxNG validation could be (des)activated and -1 in
|
414
|
+
* case of error.
|
415
|
+
*/
|
416
|
+
static VALUE
|
417
|
+
rxml_reader_relax_ng_validate(VALUE self, VALUE rng)
|
418
|
+
{
|
419
|
+
char *xrng = NIL_P(rng) ? NULL : StringValueCStr(rng);
|
420
|
+
return INT2FIX(xmlTextReaderRelaxNGValidate(rxml_text_reader_get(self), xrng));
|
421
|
+
}
|
422
|
+
|
423
|
+
#if LIBXML_VERSION >= 20620
|
424
|
+
/*
|
425
|
+
* call-seq:
|
426
|
+
* reader.schema_validate(schema) -> code
|
427
|
+
*
|
428
|
+
* Use W3C XSD schema to validate the document as it is processed. Activation
|
429
|
+
* is only possible before the first read. If +schema+ is nil, then XML Schema
|
430
|
+
* validation is desactivated.
|
431
|
+
*
|
432
|
+
* Return 0 in case the schemas validation could be (de)activated and -1 in
|
433
|
+
* case of error.
|
434
|
+
*/
|
435
|
+
static VALUE
|
436
|
+
rxml_reader_schema_validate(VALUE self, VALUE xsd)
|
437
|
+
{
|
438
|
+
char *xxsd = NIL_P(xsd) ? NULL : StringValueCStr(xsd);
|
439
|
+
int status = xmlTextReaderSchemaValidate(rxml_text_reader_get(self), xxsd);
|
440
|
+
return INT2FIX(status);
|
441
|
+
}
|
442
|
+
#endif
|
443
|
+
|
444
|
+
/*
|
445
|
+
* call-seq:
|
446
|
+
* reader.name -> name
|
447
|
+
*
|
448
|
+
* Return the qualified name of the node.
|
449
|
+
*/
|
450
|
+
static VALUE
|
451
|
+
rxml_reader_name(VALUE self)
|
452
|
+
{
|
453
|
+
const xmlChar *result = xmlTextReaderConstName(rxml_text_reader_get(self));
|
454
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
455
|
+
}
|
456
|
+
|
457
|
+
/*
|
458
|
+
* call-seq:
|
459
|
+
* reader.local_name -> name
|
460
|
+
*
|
461
|
+
* Return the local name of the node.
|
462
|
+
*/
|
463
|
+
static VALUE
|
464
|
+
rxml_reader_local_name(VALUE self)
|
465
|
+
{
|
466
|
+
const xmlChar *result = xmlTextReaderConstLocalName(rxml_text_reader_get(self));
|
467
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
468
|
+
}
|
469
|
+
|
470
|
+
/*
|
471
|
+
* call-seq:
|
472
|
+
* reader.attribute_count -> count
|
473
|
+
*
|
474
|
+
* Provide the number of attributes of the current node.
|
475
|
+
*/
|
476
|
+
static VALUE
|
477
|
+
rxml_reader_attr_count(VALUE self)
|
478
|
+
{
|
479
|
+
return INT2FIX(xmlTextReaderAttributeCount(rxml_text_reader_get(self)));
|
480
|
+
}
|
481
|
+
|
482
|
+
/*
|
483
|
+
* call-seq:
|
484
|
+
* reader.encoding -> encoding
|
485
|
+
*
|
486
|
+
* Determine the encoding of the document being read.
|
487
|
+
*/
|
488
|
+
static VALUE
|
489
|
+
rxml_reader_encoding(VALUE self)
|
490
|
+
{
|
491
|
+
const xmlChar *result = xmlTextReaderConstEncoding(rxml_text_reader_get(self));
|
492
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
493
|
+
}
|
494
|
+
|
495
|
+
/*
|
496
|
+
* call-seq:
|
497
|
+
* reader.base_uri -> URI
|
498
|
+
*
|
499
|
+
* Determine the base URI of the node.
|
500
|
+
*/
|
501
|
+
static VALUE
|
502
|
+
rxml_reader_base_uri(VALUE self)
|
503
|
+
{
|
504
|
+
const xmlChar *result = xmlTextReaderConstBaseUri(rxml_text_reader_get(self));
|
505
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
506
|
+
}
|
507
|
+
|
508
|
+
/*
|
509
|
+
* call-seq:
|
510
|
+
* reader.namespace_uri -> URI
|
511
|
+
*
|
512
|
+
* Determine the namespace URI of the node.
|
513
|
+
*/
|
514
|
+
static VALUE
|
515
|
+
rxml_reader_namespace_uri(VALUE self)
|
516
|
+
{
|
517
|
+
const xmlChar *result = xmlTextReaderConstNamespaceUri(rxml_text_reader_get(self));
|
518
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
519
|
+
}
|
520
|
+
|
521
|
+
/*
|
522
|
+
* call-seq:
|
523
|
+
* reader.value -> text
|
524
|
+
*
|
525
|
+
* Provide the text value of the node if present.
|
526
|
+
*/
|
527
|
+
static VALUE
|
528
|
+
rxml_reader_value(VALUE self)
|
529
|
+
{
|
530
|
+
const xmlChar *result = xmlTextReaderConstValue(rxml_text_reader_get(self));
|
531
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
532
|
+
}
|
533
|
+
|
534
|
+
/*
|
535
|
+
* call-seq:
|
536
|
+
* reader.prefix -> prefix
|
537
|
+
*
|
538
|
+
* Get a shorthand reference to the namespace associated with the node.
|
539
|
+
*/
|
540
|
+
static VALUE
|
541
|
+
rxml_reader_prefix(VALUE self)
|
542
|
+
{
|
543
|
+
const xmlChar *result = xmlTextReaderConstPrefix(rxml_text_reader_get(self));
|
544
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
545
|
+
}
|
546
|
+
|
547
|
+
/*
|
548
|
+
* call-seq:
|
549
|
+
* reader.depth -> depth
|
550
|
+
*
|
551
|
+
* Get the depth of the node in the tree.
|
552
|
+
*/
|
553
|
+
static VALUE
|
554
|
+
rxml_reader_depth(VALUE self)
|
555
|
+
{
|
556
|
+
return INT2FIX(xmlTextReaderDepth(rxml_text_reader_get(self)));
|
557
|
+
}
|
558
|
+
|
559
|
+
/*
|
560
|
+
* call-seq:
|
561
|
+
* reader.quote_char -> char
|
562
|
+
*
|
563
|
+
* Get the quotation mark character used to enclose the value of an attribute,
|
564
|
+
* as an integer value (and -1 in case of error).
|
565
|
+
*/
|
566
|
+
static VALUE
|
567
|
+
rxml_reader_quote_char(VALUE self)
|
568
|
+
{
|
569
|
+
return INT2FIX(xmlTextReaderQuoteChar(rxml_text_reader_get(self)));
|
570
|
+
}
|
571
|
+
|
572
|
+
/*
|
573
|
+
* call-seq:
|
574
|
+
* reader.standalone -> code
|
575
|
+
*
|
576
|
+
* Determine the standalone status of the document being read.
|
577
|
+
*
|
578
|
+
* Return 1 if the document was declared to be standalone, 0 if it was
|
579
|
+
* declared to be not standalone, or -1 if the document did not specify its
|
580
|
+
* standalone status or in case of error.
|
581
|
+
*/
|
582
|
+
static VALUE
|
583
|
+
rxml_reader_standalone(VALUE self)
|
584
|
+
{
|
585
|
+
return INT2FIX(xmlTextReaderStandalone(rxml_text_reader_get(self)));
|
586
|
+
}
|
587
|
+
|
588
|
+
/*
|
589
|
+
* call-seq:
|
590
|
+
* reader.xml_lang -> value
|
591
|
+
*
|
592
|
+
* Get the xml:lang scope within which the node resides.
|
593
|
+
*/
|
594
|
+
static VALUE
|
595
|
+
rxml_reader_xml_lang(VALUE self)
|
596
|
+
{
|
597
|
+
const xmlChar *result = xmlTextReaderConstXmlLang(rxml_text_reader_get(self));
|
598
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
599
|
+
}
|
600
|
+
|
601
|
+
/*
|
602
|
+
* call-seq:
|
603
|
+
* reader.xml_version -> version
|
604
|
+
*
|
605
|
+
* Determine the XML version of the document being read.
|
606
|
+
*/
|
607
|
+
static VALUE
|
608
|
+
rxml_reader_xml_version(VALUE self)
|
609
|
+
{
|
610
|
+
const xmlChar *result = xmlTextReaderConstXmlVersion(rxml_text_reader_get(self));
|
611
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
612
|
+
}
|
613
|
+
|
614
|
+
/*
|
615
|
+
* call-seq:
|
616
|
+
* reader.has_attributes? -> bool
|
617
|
+
*
|
618
|
+
* Get whether the node has attributes.
|
619
|
+
*/
|
620
|
+
static VALUE
|
621
|
+
rxml_reader_has_attributes(VALUE self)
|
622
|
+
{
|
623
|
+
return xmlTextReaderHasAttributes(rxml_text_reader_get(self)) ? Qtrue : Qfalse;
|
624
|
+
}
|
625
|
+
|
626
|
+
/*
|
627
|
+
* call-seq:
|
628
|
+
* reader.has_value? -> bool
|
629
|
+
*
|
630
|
+
* Get whether the node can have a text value.
|
631
|
+
*/
|
632
|
+
static VALUE
|
633
|
+
rxml_reader_has_value(VALUE self)
|
634
|
+
{
|
635
|
+
return xmlTextReaderHasValue(rxml_text_reader_get(self)) ? Qtrue : Qfalse;
|
636
|
+
}
|
637
|
+
|
638
|
+
/*
|
639
|
+
* call-seq:
|
640
|
+
* reader[key] -> value
|
641
|
+
*
|
642
|
+
* Provide the value of the attribute with the specified index (if +key+ is an
|
643
|
+
* integer) or with the specified name (if +key+ is a string) relative to the
|
644
|
+
* containing element, as a string.
|
645
|
+
*/
|
646
|
+
static VALUE
|
647
|
+
rxml_reader_attribute(VALUE self, VALUE key)
|
648
|
+
{
|
649
|
+
xmlTextReaderPtr reader;
|
650
|
+
xmlChar *attr;
|
651
|
+
|
652
|
+
reader = rxml_text_reader_get(self);
|
653
|
+
|
654
|
+
if (TYPE(key) == T_FIXNUM) {
|
655
|
+
attr = xmlTextReaderGetAttributeNo(reader, FIX2INT(key));
|
656
|
+
}
|
657
|
+
else {
|
658
|
+
attr = xmlTextReaderGetAttribute(reader, (const xmlChar *)StringValueCStr(key));
|
659
|
+
}
|
660
|
+
return (attr == NULL ? Qnil : rb_str_new2(attr));
|
661
|
+
}
|
662
|
+
|
663
|
+
/*
|
664
|
+
* call-seq:
|
665
|
+
* reader.lookup_namespace(prefix) -> value
|
666
|
+
*
|
667
|
+
* Resolve a namespace prefix in the scope of the current element.
|
668
|
+
* To return the default namespace, specify nil as +prefix+.
|
669
|
+
*/
|
670
|
+
static VALUE
|
671
|
+
rxml_reader_lookup_namespace(VALUE self, VALUE prefix)
|
672
|
+
{
|
673
|
+
const xmlChar *result = xmlTextReaderLookupNamespace(rxml_text_reader_get(self), (const xmlChar *)StringValueCStr(prefix));
|
674
|
+
return (result == NULL ? Qnil : rb_str_new2(result));
|
675
|
+
}
|
676
|
+
|
677
|
+
/*
|
678
|
+
* call-seq:
|
679
|
+
* reader.expand -> node
|
680
|
+
*
|
681
|
+
* Read the contents of the current node and the full subtree. It then makes
|
682
|
+
* the subtree available until the next read call.
|
683
|
+
*
|
684
|
+
* Return an XML::Node object, or nil in case of error.
|
685
|
+
*/
|
686
|
+
static VALUE
|
687
|
+
rxml_reader_expand(VALUE self)
|
688
|
+
{
|
689
|
+
xmlNodePtr node;
|
690
|
+
xmlDocPtr doc;
|
691
|
+
xmlTextReaderPtr reader = rxml_text_reader_get(self);
|
692
|
+
node = xmlTextReaderExpand(reader);
|
693
|
+
|
694
|
+
if (!node)
|
695
|
+
return Qnil;
|
696
|
+
|
697
|
+
/* Okay this is tricky. By accessing the returned node, we
|
698
|
+
take ownership of the reader's document. Thus we need to
|
699
|
+
tell the reader to not free it. Otherwise it will be
|
700
|
+
freed twice - once when the Ruby document wrapper goes
|
701
|
+
out of scope and once when the reader goes out of scope. */
|
702
|
+
|
703
|
+
xmlTextReaderPreserve(reader);
|
704
|
+
doc = xmlTextReaderCurrentDoc(reader);
|
705
|
+
rxml_document_wrap(doc);
|
706
|
+
|
707
|
+
return rxml_node2_wrap(cXMLNode, node);
|
708
|
+
}
|
709
|
+
|
710
|
+
#if LIBXML_VERSION >= 20618
|
711
|
+
/*
|
712
|
+
* call-seq:
|
713
|
+
* reader.byte_consumed -> value
|
714
|
+
*
|
715
|
+
* This method provides the current index of the parser used by the reader,
|
716
|
+
* relative to the start of the current entity.
|
717
|
+
*/
|
718
|
+
static VALUE
|
719
|
+
rxml_reader_byte_consumed(VALUE self)
|
720
|
+
{
|
721
|
+
return INT2NUM(xmlTextReaderByteConsumed(rxml_text_reader_get(self)));
|
722
|
+
}
|
723
|
+
#endif
|
724
|
+
|
725
|
+
#if LIBXML_VERSION >= 20617
|
726
|
+
/*
|
727
|
+
* call-seq:
|
728
|
+
* reader.column_number -> number
|
729
|
+
*
|
730
|
+
* Provide the column number of the current parsing point.
|
731
|
+
*/
|
732
|
+
static VALUE
|
733
|
+
rxml_reader_column_number(VALUE self)
|
734
|
+
{
|
735
|
+
return INT2NUM(xmlTextReaderGetParserColumnNumber(rxml_text_reader_get(self)));
|
736
|
+
}
|
737
|
+
|
738
|
+
/*
|
739
|
+
* call-seq:
|
740
|
+
* reader.line_number -> number
|
741
|
+
*
|
742
|
+
* Provide the line number of the current parsing point.
|
743
|
+
*/
|
744
|
+
static VALUE
|
745
|
+
rxml_reader_line_number(VALUE self)
|
746
|
+
{
|
747
|
+
return INT2NUM(xmlTextReaderGetParserLineNumber(rxml_text_reader_get(self)));
|
748
|
+
}
|
749
|
+
#endif
|
750
|
+
|
751
|
+
/*
|
752
|
+
* call-seq:
|
753
|
+
* reader.default? -> bool
|
754
|
+
*
|
755
|
+
* Return whether an Attribute node was generated from the default value
|
756
|
+
* defined in the DTD or schema.
|
757
|
+
*/
|
758
|
+
static VALUE
|
759
|
+
rxml_reader_default(VALUE self)
|
760
|
+
{
|
761
|
+
return xmlTextReaderIsDefault(rxml_text_reader_get(self)) ? Qtrue : Qfalse;
|
762
|
+
}
|
763
|
+
|
764
|
+
/*
|
765
|
+
* call-seq:
|
766
|
+
* reader.namespace_declaration? -> bool
|
767
|
+
*
|
768
|
+
* Determine whether the current node is a namespace declaration rather than a
|
769
|
+
* regular attribute.
|
770
|
+
*/
|
771
|
+
static VALUE
|
772
|
+
rxml_reader_namespace_declaration(VALUE self)
|
773
|
+
{
|
774
|
+
return xmlTextReaderIsNamespaceDecl(rxml_text_reader_get(self)) ? Qtrue : Qfalse;
|
775
|
+
}
|
776
|
+
|
777
|
+
/*
|
778
|
+
* call-seq:
|
779
|
+
* reader.empty_element? -> bool
|
780
|
+
*
|
781
|
+
* Check if the current node is empty.
|
782
|
+
*/
|
783
|
+
static VALUE
|
784
|
+
rxml_reader_empty_element(VALUE self)
|
785
|
+
{
|
786
|
+
return xmlTextReaderIsEmptyElement(rxml_text_reader_get(self)) ? Qtrue : Qfalse;
|
787
|
+
}
|
788
|
+
|
789
|
+
/*
|
790
|
+
* call-seq:
|
791
|
+
* reader.valid? -> bool
|
792
|
+
*
|
793
|
+
* Retrieve the validity status from the parser context.
|
794
|
+
*/
|
795
|
+
static VALUE
|
796
|
+
rxml_reader_valid(VALUE self)
|
797
|
+
{
|
798
|
+
return xmlTextReaderIsValid(rxml_text_reader_get(self)) ? Qtrue : Qfalse;
|
799
|
+
}
|
800
|
+
|
801
|
+
/* Rdoc needs to know. */
|
802
|
+
#ifdef RDOC_NEVER_DEFINED
|
803
|
+
mLibXML = rb_define_module("LibXML");
|
804
|
+
mXML = rb_define_module_under(mLibXML, "XML");
|
805
|
+
#endif
|
806
|
+
|
807
|
+
void
|
808
|
+
ruby_init_xml_reader(void)
|
809
|
+
{
|
810
|
+
cXMLReader = rb_define_class_under(mXML, "Reader", rb_cObject);
|
811
|
+
|
812
|
+
rb_define_singleton_method(cXMLReader, "file", rxml_reader_new_file, -1);
|
813
|
+
rb_define_singleton_method(cXMLReader, "io", rxml_reader_new_io, -1);
|
814
|
+
rb_define_singleton_method(cXMLReader, "walker", rxml_reader_new_walker, 1);
|
815
|
+
rb_define_alias(CLASS_OF(cXMLReader), "document", "walker");
|
816
|
+
rb_define_singleton_method(cXMLReader, "new", rxml_reader_new_data, -1);
|
817
|
+
rb_define_alias(CLASS_OF(cXMLReader), "string", "new");
|
818
|
+
|
819
|
+
rb_define_method(cXMLReader, "close", rxml_reader_close, 0);
|
820
|
+
|
821
|
+
rb_define_method(cXMLReader, "move_to_attribute", rxml_reader_move_to_attr, 1);
|
822
|
+
rb_define_method(cXMLReader, "move_to_first_attribute", rxml_reader_move_to_first_attr, 0);
|
823
|
+
rb_define_method(cXMLReader, "move_to_next_attribute", rxml_reader_move_to_next_attr, 0);
|
824
|
+
rb_define_method(cXMLReader, "move_to_element", rxml_reader_move_to_element, 0);
|
825
|
+
rb_define_method(cXMLReader, "next", rxml_reader_next, 0);
|
826
|
+
rb_define_method(cXMLReader, "next_sibling", rxml_reader_next_sibling, 0);
|
827
|
+
rb_define_method(cXMLReader, "read", rxml_reader_read, 0);
|
828
|
+
rb_define_method(cXMLReader, "read_attribute_value", rxml_reader_read_attr_value, 0);
|
829
|
+
rb_define_method(cXMLReader, "read_inner_xml", rxml_reader_read_inner_xml, 0);
|
830
|
+
rb_define_method(cXMLReader, "read_outer_xml", rxml_reader_read_outer_xml, 0);
|
831
|
+
rb_define_method(cXMLReader, "read_state", rxml_reader_read_state, 0);
|
832
|
+
rb_define_method(cXMLReader, "read_string", rxml_reader_read_string, 0);
|
833
|
+
|
834
|
+
rb_define_method(cXMLReader, "relax_ng_validate", rxml_reader_relax_ng_validate, 1);
|
835
|
+
#if LIBXML_VERSION >= 20620
|
836
|
+
rb_define_method(cXMLReader, "schema_validate", rxml_reader_schema_validate, 1);
|
837
|
+
#endif
|
838
|
+
|
839
|
+
rb_define_method(cXMLReader, "node_type", rxml_reader_node_type, 0);
|
840
|
+
rb_define_method(cXMLReader, "normalization", rxml_reader_normalization, 0);
|
841
|
+
rb_define_method(cXMLReader, "attribute_count", rxml_reader_attr_count, 0);
|
842
|
+
rb_define_method(cXMLReader, "name", rxml_reader_name, 0);
|
843
|
+
rb_define_method(cXMLReader, "local_name", rxml_reader_local_name, 0);
|
844
|
+
rb_define_method(cXMLReader, "encoding", rxml_reader_encoding, 0);
|
845
|
+
rb_define_method(cXMLReader, "base_uri", rxml_reader_base_uri, 0);
|
846
|
+
rb_define_method(cXMLReader, "namespace_uri", rxml_reader_namespace_uri, 0);
|
847
|
+
rb_define_method(cXMLReader, "xml_lang", rxml_reader_xml_lang, 0);
|
848
|
+
rb_define_method(cXMLReader, "xml_version", rxml_reader_xml_version, 0);
|
849
|
+
rb_define_method(cXMLReader, "prefix", rxml_reader_prefix, 0);
|
850
|
+
rb_define_method(cXMLReader, "depth", rxml_reader_depth, 0);
|
851
|
+
rb_define_method(cXMLReader, "quote_char", rxml_reader_quote_char, 0);
|
852
|
+
rb_define_method(cXMLReader, "standalone", rxml_reader_standalone, 0);
|
853
|
+
|
854
|
+
rb_define_method(cXMLReader, "has_attributes?", rxml_reader_has_attributes, 0);
|
855
|
+
rb_define_method(cXMLReader, "[]", rxml_reader_attribute, 1);
|
856
|
+
rb_define_method(cXMLReader, "has_value?", rxml_reader_has_value, 0);
|
857
|
+
rb_define_method(cXMLReader, "value", rxml_reader_value, 0);
|
858
|
+
|
859
|
+
rb_define_method(cXMLReader, "lookup_namespace", rxml_reader_lookup_namespace, 1);
|
860
|
+
rb_define_method(cXMLReader, "expand", rxml_reader_expand, 0);
|
861
|
+
|
862
|
+
#if LIBXML_VERSION >= 20618
|
863
|
+
rb_define_method(cXMLReader, "byte_consumed", rxml_reader_byte_consumed, 0);
|
864
|
+
#endif
|
865
|
+
#if LIBXML_VERSION >= 20617
|
866
|
+
rb_define_method(cXMLReader, "column_number", rxml_reader_column_number, 0);
|
867
|
+
rb_define_method(cXMLReader, "line_number", rxml_reader_line_number, 0);
|
868
|
+
#endif
|
869
|
+
rb_define_method(cXMLReader, "default?", rxml_reader_default, 0);
|
870
|
+
rb_define_method(cXMLReader, "empty_element?", rxml_reader_empty_element, 0);
|
871
|
+
rb_define_method(cXMLReader, "namespace_declaration?", rxml_reader_namespace_declaration, 0);
|
872
|
+
rb_define_method(cXMLReader, "valid?", rxml_reader_valid, 0);
|
873
|
+
|
874
|
+
rb_define_const(cXMLReader, "LOADDTD", INT2FIX(XML_PARSER_LOADDTD));
|
875
|
+
rb_define_const(cXMLReader, "DEFAULTATTRS", INT2FIX(XML_PARSER_DEFAULTATTRS));
|
876
|
+
rb_define_const(cXMLReader, "VALIDATE", INT2FIX(XML_PARSER_VALIDATE));
|
877
|
+
rb_define_const(cXMLReader, "SUBST_ENTITIES", INT2FIX(XML_PARSER_SUBST_ENTITIES));
|
878
|
+
|
879
|
+
rb_define_const(cXMLReader, "SEVERITY_VALIDITY_WARNING", INT2FIX(XML_PARSER_SEVERITY_VALIDITY_WARNING));
|
880
|
+
rb_define_const(cXMLReader, "SEVERITY_VALIDITY_ERROR", INT2FIX(XML_PARSER_SEVERITY_VALIDITY_ERROR));
|
881
|
+
rb_define_const(cXMLReader, "SEVERITY_WARNING", INT2FIX(XML_PARSER_SEVERITY_WARNING));
|
882
|
+
rb_define_const(cXMLReader, "SEVERITY_ERROR", INT2FIX(XML_PARSER_SEVERITY_ERROR));
|
883
|
+
|
884
|
+
rb_define_const(cXMLReader, "TYPE_NONE", INT2FIX(XML_READER_TYPE_NONE));
|
885
|
+
rb_define_const(cXMLReader, "TYPE_ELEMENT", INT2FIX(XML_READER_TYPE_ELEMENT));
|
886
|
+
rb_define_const(cXMLReader, "TYPE_ATTRIBUTE", INT2FIX(XML_READER_TYPE_ATTRIBUTE));
|
887
|
+
rb_define_const(cXMLReader, "TYPE_TEXT", INT2FIX(XML_READER_TYPE_TEXT));
|
888
|
+
rb_define_const(cXMLReader, "TYPE_CDATA", INT2FIX(XML_READER_TYPE_CDATA));
|
889
|
+
rb_define_const(cXMLReader, "TYPE_ENTITY_REFERENCE", INT2FIX(XML_READER_TYPE_ENTITY_REFERENCE));
|
890
|
+
rb_define_const(cXMLReader, "TYPE_ENTITY", INT2FIX(XML_READER_TYPE_ENTITY));
|
891
|
+
rb_define_const(cXMLReader, "TYPE_PROCESSING_INSTRUCTION", INT2FIX(XML_READER_TYPE_PROCESSING_INSTRUCTION));
|
892
|
+
rb_define_const(cXMLReader, "TYPE_COMMENT", INT2FIX(XML_READER_TYPE_COMMENT));
|
893
|
+
rb_define_const(cXMLReader, "TYPE_DOCUMENT", INT2FIX(XML_READER_TYPE_DOCUMENT));
|
894
|
+
rb_define_const(cXMLReader, "TYPE_DOCUMENT_TYPE", INT2FIX(XML_READER_TYPE_DOCUMENT_TYPE));
|
895
|
+
rb_define_const(cXMLReader, "TYPE_DOCUMENT_FRAGMENT", INT2FIX(XML_READER_TYPE_DOCUMENT_FRAGMENT));
|
896
|
+
rb_define_const(cXMLReader, "TYPE_NOTATION", INT2FIX(XML_READER_TYPE_NOTATION));
|
897
|
+
rb_define_const(cXMLReader, "TYPE_WHITESPACE", INT2FIX(XML_READER_TYPE_WHITESPACE));
|
898
|
+
rb_define_const(cXMLReader, "TYPE_SIGNIFICANT_WHITESPACE", INT2FIX(XML_READER_TYPE_SIGNIFICANT_WHITESPACE));
|
899
|
+
rb_define_const(cXMLReader, "TYPE_END_ELEMENT", INT2FIX(XML_READER_TYPE_END_ELEMENT));
|
900
|
+
rb_define_const(cXMLReader, "TYPE_END_ENTITY", INT2FIX(XML_READER_TYPE_END_ENTITY));
|
901
|
+
rb_define_const(cXMLReader, "TYPE_XML_DECLARATION", INT2FIX(XML_READER_TYPE_XML_DECLARATION));
|
902
|
+
|
903
|
+
/* Read states */
|
904
|
+
rb_define_const(cXMLReader, "MODE_INITIAL", INT2FIX(XML_TEXTREADER_MODE_INITIAL));
|
905
|
+
rb_define_const(cXMLReader, "MODE_INTERACTIVE", INT2FIX(XML_TEXTREADER_MODE_INTERACTIVE));
|
906
|
+
rb_define_const(cXMLReader, "MODE_ERROR", INT2FIX(XML_TEXTREADER_MODE_ERROR));
|
907
|
+
rb_define_const(cXMLReader, "MODE_EOF", INT2FIX(XML_TEXTREADER_MODE_EOF));
|
908
|
+
rb_define_const(cXMLReader, "MODE_CLOSED", INT2FIX(XML_TEXTREADER_MODE_CLOSED));
|
909
|
+
rb_define_const(cXMLReader, "MODE_READING", INT2FIX(XML_TEXTREADER_MODE_READING));
|
910
|
+
}
|