ObjFW  Check-in [b236334eaa]

Overview
Comment:Fix partial parsing when only half of an UTF-8 char has been received.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA3-256: b236334eaa647fe55a913b8ba8965bef39113eae51ab64f32250a069131faa38
User & Date: js on 2010-06-06 22:38:31
Other Links: manifest | tags
Context
2010-06-06
23:31
Fix a warning with Apple GCC 4.0.1. check-in: 889f63ab01 user: js tags: trunk
22:38
Fix partial parsing when only half of an UTF-8 char has been received. check-in: b236334eaa user: js tags: trunk
22:08
Correctly handle section 2.10 and 2.11 of XML 1.0. check-in: bbbb105f5f user: js tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Modified src/OFXMLParser.m from [20f13b2456] to [11560d2f92].

︙︙
174
175
176
177
178
179
180
181

182
183
184
185
186
187
188
189

		switch (state) {
		/* Not in a tag */
		case OF_XMLPARSER_OUTSIDE_TAG:
			if (buf[i] == '<') {
				len = i - last;

				if (len > 0)
					[cache appendCString: buf + last

						  withLength: len];

				if ([cache cStringLength] > 0) {
					OFString *str;

					pool = [[OFAutoreleasePool alloc] init];
					str = transform_string(cache, self);
					[delegate xmlParser: self








|
>
|








174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190

		switch (state) {
		/* Not in a tag */
		case OF_XMLPARSER_OUTSIDE_TAG:
			if (buf[i] == '<') {
				len = i - last;

				if (len > 0)
					[cache appendCStringWithoutUTF8Checking:
					    buf + last
					    length: len];

				if ([cache cStringLength] > 0) {
					OFString *str;

					pool = [[OFAutoreleasePool alloc] init];
					str = transform_string(cache, self);
					[delegate xmlParser: self
︙︙
217
218
219
220
221
222
223
224

225
226
227
228
229
230
231
232

			if (buf[i] == ' ' || buf[i] == '\n' || buf[i] == '\r' ||
			    buf[i] == '>' || buf[i] == '/') {
				const char *cache_c, *tmp;
				size_t cache_len;

				len = i - last;
				if (len > 0)
					[cache appendCString: buf + last

						  withLength: len];
				cache_c = [cache cString];
				cache_len = [cache cStringLength];

				if ((tmp = memchr(cache_c, ':',
				    cache_len)) != NULL) {
					name = [[OFString alloc]
					    initWithCString: tmp + 1








|
>
|








218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234

			if (buf[i] == ' ' || buf[i] == '\n' || buf[i] == '\r' ||
			    buf[i] == '>' || buf[i] == '/') {
				const char *cache_c, *tmp;
				size_t cache_len;

				len = i - last;
				if (len > 0)
					[cache appendCStringWithoutUTF8Checking:
					    buf + last
					    length: len];
				cache_c = [cache cString];
				cache_len = [cache cStringLength];

				if ((tmp = memchr(cache_c, ':',
				    cache_len)) != NULL) {
					name = [[OFString alloc]
					    initWithCString: tmp + 1
︙︙
298
299
300
301
302
303
304
305

306
307
308
309
310
311
312
313

			    buf[i] == '>') {
				const char *cache_c, *tmp;
				size_t cache_len;
				OFString *ns;

				len = i - last;
				if (len > 0)
					[cache appendCString: buf + last

						  withLength: len];
				cache_c = [cache cString];
				cache_len = [cache cStringLength];

				if ((tmp = memchr(cache_c, ':',
				    cache_len)) != NULL) {
					name = [[OFString alloc]
					    initWithCString: tmp + 1








|
>
|








300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316

			    buf[i] == '>') {
				const char *cache_c, *tmp;
				size_t cache_len;
				OFString *ns;

				len = i - last;
				if (len > 0)
					[cache appendCStringWithoutUTF8Checking:
					    buf + last
					    length: len];
				cache_c = [cache cString];
				cache_len = [cache cStringLength];

				if ((tmp = memchr(cache_c, ':',
				    cache_len)) != NULL) {
					name = [[OFString alloc]
					    initWithCString: tmp + 1
︙︙
413
414
415
416
417
418
419
420

421
422
423
424
425
426
427
428

		case OF_XMLPARSER_IN_ATTR_NAME:
			if (buf[i] == '=') {
				const char *cache_c, *tmp;
				size_t cache_len;

				len = i - last;
				if (len > 0)
					[cache appendCString: buf + last

						  withLength: len];

				cache_c = [cache cString];
				cache_len = [cache cStringLength];

				if ((tmp = memchr(cache_c, ':',
				    cache_len)) != NULL ) {
					attrName = [[OFString alloc]








|
>
|








416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432

		case OF_XMLPARSER_IN_ATTR_NAME:
			if (buf[i] == '=') {
				const char *cache_c, *tmp;
				size_t cache_len;

				len = i - last;
				if (len > 0)
					[cache appendCStringWithoutUTF8Checking:
					    buf + last
					    length: len];

				cache_c = [cache cString];
				cache_len = [cache cStringLength];

				if ((tmp = memchr(cache_c, ':',
				    cache_len)) != NULL ) {
					attrName = [[OFString alloc]
︙︙
459
460
461
462
463
464
465
466

467
468
469
470
471
472
473
474

		case OF_XMLPARSER_IN_ATTR_VALUE:
			if (buf[i] == delim) {
				OFString *attr_ns;
				OFString *attr_val;

				len = i - last;
				if (len > 0)
					[cache appendCString: buf + last

						  withLength: len];

				pool = [[OFAutoreleasePool alloc] init];
				attr_ns = namespace_for_prefix(
				    (attrPrefix != nil ? attrPrefix : prefix),
				    namespaces);
				attr_val = transform_string(cache, self);









|
>
|








463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479

		case OF_XMLPARSER_IN_ATTR_VALUE:
			if (buf[i] == delim) {
				OFString *attr_ns;
				OFString *attr_val;

				len = i - last;
				if (len > 0)
					[cache appendCStringWithoutUTF8Checking:
					    buf + last
					    length: len];

				pool = [[OFAutoreleasePool alloc] init];
				attr_ns = namespace_for_prefix(
				    (attrPrefix != nil ? attrPrefix : prefix),
				    namespaces);
				attr_val = transform_string(cache, self);
︙︙
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

			break;
		case OF_XMLPARSER_IN_COMMENT_3:
			if (buf[i] == '-')
				state = OF_XMLPARSER_IN_COMMENT_4;
			break;
		case OF_XMLPARSER_IN_COMMENT_4:
			if (buf[i] == '-') {

				size_t cache_len;



				[cache appendCString: buf + last

					  withLength: i - last];
				cache_len = [cache length];



				pool = [[OFAutoreleasePool alloc] init];
				[cache removeCharactersFromIndex: cache_len - 1
							 toIndex: cache_len];
				[cache removeLeadingAndTrailingWhitespaces];
				[delegate xmlParser: self
				       foundComment: cache];
				[pool release];

				[cache setToCString: ""];

				last = i + 1;
				state = OF_XMLPARSER_EXPECT_CLOSE;
			} else
				state = OF_XMLPARSER_IN_COMMENT_3;

			break;
		}
	}

	len = size - last;
	/* In OF_XMLPARSER_IN_TAG, there can be only spaces */
	if (len > 0 && state != OF_XMLPARSER_IN_TAG)
		[cache appendCString: buf + last
			  withLength: len];
}

- (OFString*)foundUnknownEntityNamed: (OFString*)entity
{
	return [delegate xmlParser: self
	   foundUnknownEntityNamed: entity];
}








>
|

>
>
|
>
|
|
>
>

<
|
|
|

|
















|
|








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

			break;
		case OF_XMLPARSER_IN_COMMENT_3:
			if (buf[i] == '-')
				state = OF_XMLPARSER_IN_COMMENT_4;
			break;
		case OF_XMLPARSER_IN_COMMENT_4:
			if (buf[i] == '-') {
				OFMutableString *comment;
				size_t len;

				pool = [[OFAutoreleasePool alloc] init];

				[cache
				    appendCStringWithoutUTF8Checking: buf + last
							      length: i - last];

				comment = [[cache mutableCopy] autorelease];
				len = [comment length];


				[comment removeCharactersFromIndex: len - 1
							   toIndex: len];
				[comment removeLeadingAndTrailingWhitespaces];
				[delegate xmlParser: self
				       foundComment: comment];
				[pool release];

				[cache setToCString: ""];

				last = i + 1;
				state = OF_XMLPARSER_EXPECT_CLOSE;
			} else
				state = OF_XMLPARSER_IN_COMMENT_3;

			break;
		}
	}

	len = size - last;
	/* In OF_XMLPARSER_IN_TAG, there can be only spaces */
	if (len > 0 && state != OF_XMLPARSER_IN_TAG)
		[cache appendCStringWithoutUTF8Checking: buf + last
						 length: len];
}

- (OFString*)foundUnknownEntityNamed: (OFString*)entity
{
	return [delegate xmlParser: self
	   foundUnknownEntityNamed: entity];
}
︙︙

Modified tests/OFXMLParserTests.m from [43eb47dbde] to [24b4ef9bb7].

︙︙
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294

	OFAutoreleasePool *pool = [[OFAutoreleasePool alloc] init];
	OFXMLParser *parser;
	const char *str = "foo<bar/>\n"
	    "<foobar xmlns='urn:objfw:test:foobar'>\n"
	    " <qux xmlns:foo='urn:objfw:test:foo'>\n"
	    "  <foo:bla foo:bla='&#x62;&#x6c;&#x61;' blafoo='foo'>\n"
	    "   <blup foo:qux='asd' quxqux='test'/>\n"
	    "   <bla:bla\rxmlns:bla=\"urn:objfw:test:bla\" qux='qux'\r\n"
	    "    bla:foo='blafoo'/>\n"
	    "   <abc xmlns='urn:objfw:test:abc' abc='abc' foo:abc='abc'/>\n"
	    "  </foo:bla>\n"
	    "  <!-- commänt -->\n"
	    " </qux>\n"
	    "</foobar>";
	size_t j, len;








|








280
281
282
283
284
285
286
287
288
289
290
291
292
293
294

	OFAutoreleasePool *pool = [[OFAutoreleasePool alloc] init];
	OFXMLParser *parser;
	const char *str = "foo<bar/>\n"
	    "<foobar xmlns='urn:objfw:test:foobar'>\n"
	    " <qux xmlns:foo='urn:objfw:test:foo'>\n"
	    "  <foo:bla foo:bla='&#x62;&#x6c;&#x61;' blafoo='foo'>\n"
	    "   <blup foo:qux='asd' quxqux='test'/>\n"
	    "   <bla:bla\r\rxmlns:bla=\"urn:objfw:test:bla\" qux='qux'\r\n"
	    "    bla:foo='blafoo'/>\n"
	    "   <abc xmlns='urn:objfw:test:abc' abc='abc' foo:abc='abc'/>\n"
	    "  </foo:bla>\n"
	    "  <!-- commänt -->\n"
	    " </qux>\n"
	    "</foobar>";
	size_t j, len;
︙︙

Fossil 2.23 [47362306a7] 2023-11-01 18:56:47
This page was generated in about 0.018s