ObjFW  Check-in [bf45e02951]

Overview
Comment:Parse XML processing instructions.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA3-256: bf45e02951d973ecb58ab9fd28b9d5512f19bad068a93321e7ce121cfab87b44
User & Date: js on 2011-04-09 20:05:15
Other Links: manifest | tags
Context
2011-04-09
20:11
Increase library version. check-in: 5927fbebb7 user: js tags: trunk
20:05
Parse XML processing instructions. check-in: bf45e02951 user: js tags: trunk
15:43
OFXMLParser: Allow processing instructions after the document. check-in: 2d7725aa27 user: js tags: trunk
Changes

Modified src/OFXMLParser.h from [4bfeec9a6e] to [0a269ecfb6].

175
176
177
178
179
180
181

182
183
184
185
186
187
188
	of_xml_parser_element_end_block_t elementEndHandler;
	of_xml_parser_string_block_t charactersHandler;
	of_xml_parser_string_block_t CDATAHandler;
	of_xml_parser_string_block_t commentHandler;
	of_xml_parser_unknown_entity_block_t unknownEntityHandler;
#endif
	size_t level;

	size_t lineNumber;
	BOOL lastCarriageReturn;
	BOOL finishedParsing;
}

#ifdef OF_HAVE_PROPERTIES
@property (retain) id <OFXMLParserDelegate> delegate;







>







175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
	of_xml_parser_element_end_block_t elementEndHandler;
	of_xml_parser_string_block_t charactersHandler;
	of_xml_parser_string_block_t CDATAHandler;
	of_xml_parser_string_block_t commentHandler;
	of_xml_parser_unknown_entity_block_t unknownEntityHandler;
#endif
	size_t level;
	BOOL acceptProlog;
	size_t lineNumber;
	BOOL lastCarriageReturn;
	BOOL finishedParsing;
}

#ifdef OF_HAVE_PROPERTIES
@property (retain) id <OFXMLParserDelegate> delegate;

Modified src/OFXMLParser.m from [1c1c18bcbe] to [669b34c861].

47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
	[cache replaceOccurrencesOfString: @"\r\n"
			       withString: @"\n"];
	[cache replaceOccurrencesOfString: @"\r"
			       withString: @"\n"];
	return [cache stringByXMLUnescapingWithDelegate: delegate];
}

static OF_INLINE OFString*
namespace_for_prefix(OFString *prefix, OFArray *namespaces)
{
	OFDictionary **carray = [namespaces cArray];
	ssize_t i;

	if (prefix == nil)
		prefix = @"";







|







47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
	[cache replaceOccurrencesOfString: @"\r\n"
			       withString: @"\n"];
	[cache replaceOccurrencesOfString: @"\r"
			       withString: @"\n"];
	return [cache stringByXMLUnescapingWithDelegate: delegate];
}

static OFString*
namespace_for_prefix(OFString *prefix, OFArray *namespaces)
{
	OFDictionary **carray = [namespaces cArray];
	ssize_t i;

	if (prefix == nil)
		prefix = @"";
153
154
155
156
157
158
159

160
161
162
163
164
165
166

		pool = [[OFAutoreleasePool alloc] init];
		dict = [OFMutableDictionary dictionaryWithKeysAndObjects:
		    @"xml", @"http://www.w3.org/XML/1998/namespace",
		    @"xmlns", @"http://www.w3.org/2000/xmlns/", nil];
		[namespaces addObject: dict];


		lineNumber = 1;

		[pool release];
	} @catch (id e) {
		[self release];
		@throw e;
	}







>







153
154
155
156
157
158
159
160
161
162
163
164
165
166
167

		pool = [[OFAutoreleasePool alloc] init];
		dict = [OFMutableDictionary dictionaryWithKeysAndObjects:
		    @"xml", @"http://www.w3.org/XML/1998/namespace",
		    @"xmlns", @"http://www.w3.org/2000/xmlns/", nil];
		[namespaces addObject: dict];

		acceptProlog = YES;
		lineNumber = 1;

		[pool release];
	} @catch (id e) {
		[self release];
		@throw e;
	}
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
	} @finally {
		[file release];
	}
}

/*
 * The following methods handle the different states of the parser. They are
 * lookup up in +[initialize] and put in a lookup table to speed things up.
 * One dispatch for every character would be way too slow!
 */

/* Not in a tag */
- (void)_parseOutsideTagWithBuffer: (const char*)buf
				 i: (size_t*)i
			      last: (size_t*)last







|







264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
	} @finally {
		[file release];
	}
}

/*
 * The following methods handle the different states of the parser. They are
 * looked up in +[initialize] and put in a lookup table to speed things up.
 * One dispatch for every character would be way too slow!
 */

/* Not in a tag */
- (void)_parseOutsideTagWithBuffer: (const char*)buf
				 i: (size_t*)i
			      last: (size_t*)last
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
			*last = *i + 1;
			state = OF_XMLPARSER_IN_PROCESSING_INSTRUCTIONS;
			level = 0;
			break;
		case '/':
			*last = *i + 1;
			state = OF_XMLPARSER_IN_CLOSE_TAG_NAME;

			break;
		case '!':
			*last = *i + 1;
			state = OF_XMLPARSER_IN_EXCLAMATIONMARK;

			break;
		default:
			state = OF_XMLPARSER_IN_TAG_NAME;

			(*i)--;
			break;
	}
}



















































































/* Inside prolog */
- (void)_parseInProcessingInstructionsWithBuffer: (const char*)buf
					       i: (size_t*)i
					    last: (size_t*)last
{
	if (buf[*i] == '?')
		level = 1;
	else if (level == 1 && buf[*i] == '>') {







>




>



>





>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
|







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
			*last = *i + 1;
			state = OF_XMLPARSER_IN_PROCESSING_INSTRUCTIONS;
			level = 0;
			break;
		case '/':
			*last = *i + 1;
			state = OF_XMLPARSER_IN_CLOSE_TAG_NAME;
			acceptProlog = NO;
			break;
		case '!':
			*last = *i + 1;
			state = OF_XMLPARSER_IN_EXCLAMATIONMARK;
			acceptProlog = NO;
			break;
		default:
			state = OF_XMLPARSER_IN_TAG_NAME;
			acceptProlog = NO;
			(*i)--;
			break;
	}
}

/* <?xml […]?> */
- (BOOL)_parseXMLProcessingInstructions: (OFString*)pi
{
	const char *pi_c;
	size_t i, last, pi_len;
	int xstate = 0;
	OFString *attr = nil;
	OFString *val = nil;
	char xdelim = 0;

	if (!acceptProlog)
		return NO;

	acceptProlog = NO;

	pi = [pi substringFromIndex: 3
			    toIndex: [pi length]];
	pi = [pi stringByDeletingLeadingAndTrailingWhitespaces];

	pi_c = [pi cString];
	pi_len = [pi cStringLength];

	for (i = last = 0; i < pi_len; i++) {
		switch (xstate) {
		case 0:
			if (pi_c[i] == ' ' || pi_c[i] == '\t' ||
			    pi_c[i] == '\r' || pi_c[i] == '\n')
				continue;

			last = i;
			xstate = 1;
			i--;

			break;
		case 1:
			if (pi_c[i] != '=')
				continue;

			attr = [OFString stringWithCString: pi_c + last
						    length: i - last];
			last = i + 1;
			xstate = 2;

			break;
		case 2:
			if (pi_c[i] != '\'' && pi_c[i] != '"')
				return NO;

			xdelim = pi_c[i];
			last = i + 1;
			xstate = 3;

			break;
		case 3:
			if (pi_c[i] != xdelim)
				continue;

			val = [OFString stringWithCString: pi_c + last
						   length: i - last];

			if ([attr isEqual: @"version"])
				if (![val hasPrefix: @"1."])
					return NO;

			if ([attr isEqual: @"encoding"])
				if ([val caseInsensitiveCompare: @"utf-8"] !=
				    OF_ORDERED_SAME)
					return NO;

			last = i + 1;
			xstate = 0;

			break;
		}
	}

	if (xstate != 0)
		return NO;

	return YES;
}

/* Inside processing instructions */
- (void)_parseInProcessingInstructionsWithBuffer: (const char*)buf
					       i: (size_t*)i
					    last: (size_t*)last
{
	if (buf[*i] == '?')
		level = 1;
	else if (level == 1 && buf[*i] == '>') {
367
368
369
370
371
372
373








374
375
376
377
378
379
380

		/*
		 * Class swizzle the string to be immutable. We pass it as
		 * OFString*, so it can't be modified anyway. But not swizzling
		 * it would create a real copy each time -[copy] is called.
		 */
		pi->isa = [OFString class];









		[delegate parser: self
		    foundProcessingInstructions: pi];

		[pool release];

		[cache setToCString: ""];







>
>
>
>
>
>
>
>







453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474

		/*
		 * Class swizzle the string to be immutable. We pass it as
		 * OFString*, so it can't be modified anyway. But not swizzling
		 * it would create a real copy each time -[copy] is called.
		 */
		pi->isa = [OFString class];

		if ([pi isEqual: @"xml"] || [pi hasPrefix: @"xml "] ||
		    [pi hasPrefix: @"xml\t"] || [pi hasPrefix: @"xml\r"] ||
		    [pi hasPrefix: @"xml\n"])
			if (![self _parseXMLProcessingInstructions: pi])
				@throw [OFMalformedXMLException
				    newWithClass: isa
					parser: self];

		[delegate parser: self
		    foundProcessingInstructions: pi];

		[pool release];

		[cache setToCString: ""];

Modified tests/OFXMLParserTests.m from [1ef5f83edf] to [1a865f21e8].

372
373
374
375
376
377
378















379
380
381

	EXPECT_EXCEPTION(@"Detection of junk after the document #1",
	    OFMalformedXMLException, [parser parseString: @"a"])

	EXPECT_EXCEPTION(@"Detection of junk after the document #2",
	    OFMalformedXMLException, [parser parseString: @"<!["])
















	[pool drain];
}
@end







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>



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

	EXPECT_EXCEPTION(@"Detection of junk after the document #1",
	    OFMalformedXMLException, [parser parseString: @"a"])

	EXPECT_EXCEPTION(@"Detection of junk after the document #2",
	    OFMalformedXMLException, [parser parseString: @"<!["])

	parser = [OFXMLParser parser];
	EXPECT_EXCEPTION(@"Detection of invalid XML processing instructions #1",
	    OFMalformedXMLException,
	    [parser parseString: @"<?xml version='2.0'?>"])

	parser = [OFXMLParser parser];
	EXPECT_EXCEPTION(@"Detection of invalid XML processing instructions #2",
	    OFMalformedXMLException,
	    [parser parseString: @"<?xml encoding='UTF-7'?>"])

	parser = [OFXMLParser parser];
	EXPECT_EXCEPTION(@"Detection of invalid XML processing instructions #3",
	    OFMalformedXMLException,
	    [parser parseString: @"<x><?xml?></x>"])

	[pool drain];
}
@end