|
1 /* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd |
|
2 See the file COPYING for copying permission. |
|
3 |
|
4 runtest.c : run the Expat test suite |
|
5 */ |
|
6 |
|
7 #ifdef HAVE_EXPAT_CONFIG_H |
|
8 #include <expat_config.h> |
|
9 #endif |
|
10 |
|
11 #include <assert.h> |
|
12 #include <stdlib.h> |
|
13 #include <stdio.h> |
|
14 #include <string.h> |
|
15 |
|
16 #include "expat.h" |
|
17 #include "chardata.h" |
|
18 #include "minicheck.h" |
|
19 |
|
20 #ifdef AMIGA_SHARED_LIB |
|
21 #include <proto/expat.h> |
|
22 #endif |
|
23 |
|
24 #ifdef XML_LARGE_SIZE |
|
25 #define XML_FMT_INT_MOD "ll" |
|
26 #else |
|
27 #define XML_FMT_INT_MOD "l" |
|
28 #endif |
|
29 |
|
30 static XML_Parser parser; |
|
31 |
|
32 |
|
33 static void |
|
34 basic_setup(void) |
|
35 { |
|
36 parser = XML_ParserCreate(NULL); |
|
37 if (parser == NULL) |
|
38 fail("Parser not created."); |
|
39 } |
|
40 |
|
41 static void |
|
42 basic_teardown(void) |
|
43 { |
|
44 if (parser != NULL) |
|
45 XML_ParserFree(parser); |
|
46 } |
|
47 |
|
48 /* Generate a failure using the parser state to create an error message; |
|
49 this should be used when the parser reports an error we weren't |
|
50 expecting. |
|
51 */ |
|
52 static void |
|
53 _xml_failure(XML_Parser parser, const char *file, int line) |
|
54 { |
|
55 char buffer[1024]; |
|
56 sprintf(buffer, |
|
57 "\n %s (line %" XML_FMT_INT_MOD "u, offset %"\ |
|
58 XML_FMT_INT_MOD "u)\n reported from %s, line %d", |
|
59 XML_ErrorString(XML_GetErrorCode(parser)), |
|
60 XML_GetCurrentLineNumber(parser), |
|
61 XML_GetCurrentColumnNumber(parser), |
|
62 file, line); |
|
63 _fail_unless(0, file, line, buffer); |
|
64 } |
|
65 |
|
66 #define xml_failure(parser) _xml_failure((parser), __FILE__, __LINE__) |
|
67 |
|
68 static void |
|
69 _expect_failure(char *text, enum XML_Error errorCode, char *errorMessage, |
|
70 char *file, int lineno) |
|
71 { |
|
72 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_OK) |
|
73 /* Hackish use of _fail_unless() macro, but let's us report |
|
74 the right filename and line number. */ |
|
75 _fail_unless(0, file, lineno, errorMessage); |
|
76 if (XML_GetErrorCode(parser) != errorCode) |
|
77 _xml_failure(parser, file, lineno); |
|
78 } |
|
79 |
|
80 #define expect_failure(text, errorCode, errorMessage) \ |
|
81 _expect_failure((text), (errorCode), (errorMessage), \ |
|
82 __FILE__, __LINE__) |
|
83 |
|
84 /* Dummy handlers for when we need to set a handler to tickle a bug, |
|
85 but it doesn't need to do anything. |
|
86 */ |
|
87 |
|
88 static void XMLCALL |
|
89 dummy_start_doctype_handler(void *userData, |
|
90 const XML_Char *doctypeName, |
|
91 const XML_Char *sysid, |
|
92 const XML_Char *pubid, |
|
93 int has_internal_subset) |
|
94 {} |
|
95 |
|
96 static void XMLCALL |
|
97 dummy_end_doctype_handler(void *userData) |
|
98 {} |
|
99 |
|
100 static void XMLCALL |
|
101 dummy_entity_decl_handler(void *userData, |
|
102 const XML_Char *entityName, |
|
103 int is_parameter_entity, |
|
104 const XML_Char *value, |
|
105 int value_length, |
|
106 const XML_Char *base, |
|
107 const XML_Char *systemId, |
|
108 const XML_Char *publicId, |
|
109 const XML_Char *notationName) |
|
110 {} |
|
111 |
|
112 static void XMLCALL |
|
113 dummy_notation_decl_handler(void *userData, |
|
114 const XML_Char *notationName, |
|
115 const XML_Char *base, |
|
116 const XML_Char *systemId, |
|
117 const XML_Char *publicId) |
|
118 {} |
|
119 |
|
120 static void XMLCALL |
|
121 dummy_element_decl_handler(void *userData, |
|
122 const XML_Char *name, |
|
123 XML_Content *model) |
|
124 {} |
|
125 |
|
126 static void XMLCALL |
|
127 dummy_attlist_decl_handler(void *userData, |
|
128 const XML_Char *elname, |
|
129 const XML_Char *attname, |
|
130 const XML_Char *att_type, |
|
131 const XML_Char *dflt, |
|
132 int isrequired) |
|
133 {} |
|
134 |
|
135 static void XMLCALL |
|
136 dummy_comment_handler(void *userData, const XML_Char *data) |
|
137 {} |
|
138 |
|
139 static void XMLCALL |
|
140 dummy_pi_handler(void *userData, const XML_Char *target, const XML_Char *data) |
|
141 {} |
|
142 |
|
143 static void XMLCALL |
|
144 dummy_start_element(void *userData, |
|
145 const XML_Char *name, const XML_Char **atts) |
|
146 {} |
|
147 |
|
148 |
|
149 /* |
|
150 * Character & encoding tests. |
|
151 */ |
|
152 |
|
153 START_TEST(test_nul_byte) |
|
154 { |
|
155 char text[] = "<doc>\0</doc>"; |
|
156 |
|
157 /* test that a NUL byte (in US-ASCII data) is an error */ |
|
158 if (XML_Parse(parser, text, sizeof(text) - 1, XML_TRUE) == XML_STATUS_OK) |
|
159 fail("Parser did not report error on NUL-byte."); |
|
160 if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_TOKEN) |
|
161 xml_failure(parser); |
|
162 } |
|
163 END_TEST |
|
164 |
|
165 |
|
166 START_TEST(test_u0000_char) |
|
167 { |
|
168 /* test that a NUL byte (in US-ASCII data) is an error */ |
|
169 expect_failure("<doc>�</doc>", |
|
170 XML_ERROR_BAD_CHAR_REF, |
|
171 "Parser did not report error on NUL-byte."); |
|
172 } |
|
173 END_TEST |
|
174 |
|
175 START_TEST(test_bom_utf8) |
|
176 { |
|
177 /* This test is really just making sure we don't core on a UTF-8 BOM. */ |
|
178 char *text = "\357\273\277<e/>"; |
|
179 |
|
180 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
181 xml_failure(parser); |
|
182 } |
|
183 END_TEST |
|
184 |
|
185 START_TEST(test_bom_utf16_be) |
|
186 { |
|
187 char text[] = "\376\377\0<\0e\0/\0>"; |
|
188 |
|
189 if (XML_Parse(parser, text, sizeof(text)-1, XML_TRUE) == XML_STATUS_ERROR) |
|
190 xml_failure(parser); |
|
191 } |
|
192 END_TEST |
|
193 |
|
194 START_TEST(test_bom_utf16_le) |
|
195 { |
|
196 char text[] = "\377\376<\0e\0/\0>\0"; |
|
197 |
|
198 if (XML_Parse(parser, text, sizeof(text)-1, XML_TRUE) == XML_STATUS_ERROR) |
|
199 xml_failure(parser); |
|
200 } |
|
201 END_TEST |
|
202 |
|
203 static void XMLCALL |
|
204 accumulate_characters(void *userData, const XML_Char *s, int len) |
|
205 { |
|
206 CharData_AppendXMLChars((CharData *)userData, s, len); |
|
207 } |
|
208 |
|
209 static void XMLCALL |
|
210 accumulate_attribute(void *userData, const XML_Char *name, |
|
211 const XML_Char **atts) |
|
212 { |
|
213 CharData *storage = (CharData *)userData; |
|
214 if (storage->count < 0 && atts != NULL && atts[0] != NULL) { |
|
215 /* "accumulate" the value of the first attribute we see */ |
|
216 CharData_AppendXMLChars(storage, atts[1], -1); |
|
217 } |
|
218 } |
|
219 |
|
220 |
|
221 static void |
|
222 _run_character_check(XML_Char *text, XML_Char *expected, |
|
223 const char *file, int line) |
|
224 { |
|
225 CharData storage; |
|
226 |
|
227 CharData_Init(&storage); |
|
228 XML_SetUserData(parser, &storage); |
|
229 XML_SetCharacterDataHandler(parser, accumulate_characters); |
|
230 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
231 _xml_failure(parser, file, line); |
|
232 CharData_CheckXMLChars(&storage, expected); |
|
233 } |
|
234 |
|
235 #define run_character_check(text, expected) \ |
|
236 _run_character_check(text, expected, __FILE__, __LINE__) |
|
237 |
|
238 static void |
|
239 _run_attribute_check(XML_Char *text, XML_Char *expected, |
|
240 const char *file, int line) |
|
241 { |
|
242 CharData storage; |
|
243 |
|
244 CharData_Init(&storage); |
|
245 XML_SetUserData(parser, &storage); |
|
246 XML_SetStartElementHandler(parser, accumulate_attribute); |
|
247 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
248 _xml_failure(parser, file, line); |
|
249 CharData_CheckXMLChars(&storage, expected); |
|
250 } |
|
251 |
|
252 #define run_attribute_check(text, expected) \ |
|
253 _run_attribute_check(text, expected, __FILE__, __LINE__) |
|
254 |
|
255 /* Regression test for SF bug #491986. */ |
|
256 START_TEST(test_danish_latin1) |
|
257 { |
|
258 char *text = |
|
259 "<?xml version='1.0' encoding='iso-8859-1'?>\n" |
|
260 "<e>J\xF8rgen \xE6\xF8\xE5\xC6\xD8\xC5</e>"; |
|
261 run_character_check(text, |
|
262 "J\xC3\xB8rgen \xC3\xA6\xC3\xB8\xC3\xA5\xC3\x86\xC3\x98\xC3\x85"); |
|
263 } |
|
264 END_TEST |
|
265 |
|
266 |
|
267 /* Regression test for SF bug #514281. */ |
|
268 START_TEST(test_french_charref_hexidecimal) |
|
269 { |
|
270 char *text = |
|
271 "<?xml version='1.0' encoding='iso-8859-1'?>\n" |
|
272 "<doc>éèàçêÈ</doc>"; |
|
273 run_character_check(text, |
|
274 "\xC3\xA9\xC3\xA8\xC3\xA0\xC3\xA7\xC3\xAA\xC3\x88"); |
|
275 } |
|
276 END_TEST |
|
277 |
|
278 START_TEST(test_french_charref_decimal) |
|
279 { |
|
280 char *text = |
|
281 "<?xml version='1.0' encoding='iso-8859-1'?>\n" |
|
282 "<doc>éèàçêÈ</doc>"; |
|
283 run_character_check(text, |
|
284 "\xC3\xA9\xC3\xA8\xC3\xA0\xC3\xA7\xC3\xAA\xC3\x88"); |
|
285 } |
|
286 END_TEST |
|
287 |
|
288 START_TEST(test_french_latin1) |
|
289 { |
|
290 char *text = |
|
291 "<?xml version='1.0' encoding='iso-8859-1'?>\n" |
|
292 "<doc>\xE9\xE8\xE0\xE7\xEa\xC8</doc>"; |
|
293 run_character_check(text, |
|
294 "\xC3\xA9\xC3\xA8\xC3\xA0\xC3\xA7\xC3\xAA\xC3\x88"); |
|
295 } |
|
296 END_TEST |
|
297 |
|
298 START_TEST(test_french_utf8) |
|
299 { |
|
300 char *text = |
|
301 "<?xml version='1.0' encoding='utf-8'?>\n" |
|
302 "<doc>\xC3\xA9</doc>"; |
|
303 run_character_check(text, "\xC3\xA9"); |
|
304 } |
|
305 END_TEST |
|
306 |
|
307 /* Regression test for SF bug #600479. |
|
308 XXX There should be a test that exercises all legal XML Unicode |
|
309 characters as PCDATA and attribute value content, and XML Name |
|
310 characters as part of element and attribute names. |
|
311 */ |
|
312 START_TEST(test_utf8_false_rejection) |
|
313 { |
|
314 char *text = "<doc>\xEF\xBA\xBF</doc>"; |
|
315 run_character_check(text, "\xEF\xBA\xBF"); |
|
316 } |
|
317 END_TEST |
|
318 |
|
319 /* Regression test for SF bug #477667. |
|
320 This test assures that any 8-bit character followed by a 7-bit |
|
321 character will not be mistakenly interpreted as a valid UTF-8 |
|
322 sequence. |
|
323 */ |
|
324 START_TEST(test_illegal_utf8) |
|
325 { |
|
326 char text[100]; |
|
327 int i; |
|
328 |
|
329 for (i = 128; i <= 255; ++i) { |
|
330 sprintf(text, "<e>%ccd</e>", i); |
|
331 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_OK) { |
|
332 sprintf(text, |
|
333 "expected token error for '%c' (ordinal %d) in UTF-8 text", |
|
334 i, i); |
|
335 fail(text); |
|
336 } |
|
337 else if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_TOKEN) |
|
338 xml_failure(parser); |
|
339 /* Reset the parser since we use the same parser repeatedly. */ |
|
340 XML_ParserReset(parser, NULL); |
|
341 } |
|
342 } |
|
343 END_TEST |
|
344 |
|
345 START_TEST(test_utf16) |
|
346 { |
|
347 /* <?xml version="1.0" encoding="UTF-16"?> |
|
348 <doc a='123'>some text</doc> |
|
349 */ |
|
350 char text[] = |
|
351 "\000<\000?\000x\000m\000\154\000 \000v\000e\000r\000s\000i\000o" |
|
352 "\000n\000=\000'\0001\000.\000\060\000'\000 \000e\000n\000c\000o" |
|
353 "\000d\000i\000n\000g\000=\000'\000U\000T\000F\000-\0001\000\066" |
|
354 "\000'\000?\000>\000\n" |
|
355 "\000<\000d\000o\000c\000 \000a\000=\000'\0001\0002\0003\000'" |
|
356 "\000>\000s\000o\000m\000e\000 \000t\000e\000x\000t\000<\000/" |
|
357 "\000d\000o\000c\000>"; |
|
358 if (XML_Parse(parser, text, sizeof(text)-1, XML_TRUE) == XML_STATUS_ERROR) |
|
359 xml_failure(parser); |
|
360 } |
|
361 END_TEST |
|
362 |
|
363 START_TEST(test_utf16_le_epilog_newline) |
|
364 { |
|
365 int first_chunk_bytes = 17; |
|
366 char text[] = |
|
367 "\xFF\xFE" /* BOM */ |
|
368 "<\000e\000/\000>\000" /* document element */ |
|
369 "\r\000\n\000\r\000\n\000"; /* epilog */ |
|
370 |
|
371 if (first_chunk_bytes >= sizeof(text) - 1) |
|
372 fail("bad value of first_chunk_bytes"); |
|
373 if ( XML_Parse(parser, text, first_chunk_bytes, XML_FALSE) |
|
374 == XML_STATUS_ERROR) |
|
375 xml_failure(parser); |
|
376 else { |
|
377 enum XML_Status rc; |
|
378 rc = XML_Parse(parser, text + first_chunk_bytes, |
|
379 sizeof(text) - first_chunk_bytes - 1, XML_TRUE); |
|
380 if (rc == XML_STATUS_ERROR) |
|
381 xml_failure(parser); |
|
382 } |
|
383 } |
|
384 END_TEST |
|
385 |
|
386 /* Regression test for SF bug #481609, #774028. */ |
|
387 START_TEST(test_latin1_umlauts) |
|
388 { |
|
389 char *text = |
|
390 "<?xml version='1.0' encoding='iso-8859-1'?>\n" |
|
391 "<e a='\xE4 \xF6 \xFC ä ö ü ä ö ü >'\n" |
|
392 " >\xE4 \xF6 \xFC ä ö ü ä ö ü ></e>"; |
|
393 char *utf8 = |
|
394 "\xC3\xA4 \xC3\xB6 \xC3\xBC " |
|
395 "\xC3\xA4 \xC3\xB6 \xC3\xBC " |
|
396 "\xC3\xA4 \xC3\xB6 \xC3\xBC >"; |
|
397 run_character_check(text, utf8); |
|
398 XML_ParserReset(parser, NULL); |
|
399 run_attribute_check(text, utf8); |
|
400 } |
|
401 END_TEST |
|
402 |
|
403 /* Regression test #1 for SF bug #653180. */ |
|
404 START_TEST(test_line_number_after_parse) |
|
405 { |
|
406 char *text = |
|
407 "<tag>\n" |
|
408 "\n" |
|
409 "\n</tag>"; |
|
410 XML_Size lineno; |
|
411 |
|
412 if (XML_Parse(parser, text, strlen(text), XML_FALSE) == XML_STATUS_ERROR) |
|
413 xml_failure(parser); |
|
414 lineno = XML_GetCurrentLineNumber(parser); |
|
415 if (lineno != 4) { |
|
416 char buffer[100]; |
|
417 sprintf(buffer, |
|
418 "expected 4 lines, saw %" XML_FMT_INT_MOD "u", lineno); |
|
419 fail(buffer); |
|
420 } |
|
421 } |
|
422 END_TEST |
|
423 |
|
424 /* Regression test #2 for SF bug #653180. */ |
|
425 START_TEST(test_column_number_after_parse) |
|
426 { |
|
427 char *text = "<tag></tag>"; |
|
428 XML_Size colno; |
|
429 |
|
430 if (XML_Parse(parser, text, strlen(text), XML_FALSE) == XML_STATUS_ERROR) |
|
431 xml_failure(parser); |
|
432 colno = XML_GetCurrentColumnNumber(parser); |
|
433 if (colno != 11) { |
|
434 char buffer[100]; |
|
435 sprintf(buffer, |
|
436 "expected 11 columns, saw %" XML_FMT_INT_MOD "u", colno); |
|
437 fail(buffer); |
|
438 } |
|
439 } |
|
440 END_TEST |
|
441 |
|
442 static void XMLCALL |
|
443 start_element_event_handler2(void *userData, const XML_Char *name, |
|
444 const XML_Char **attr) |
|
445 { |
|
446 CharData *storage = (CharData *) userData; |
|
447 char buffer[100]; |
|
448 |
|
449 sprintf(buffer, |
|
450 "<%s> at col:%" XML_FMT_INT_MOD "u line:%"\ |
|
451 XML_FMT_INT_MOD "u\n", name, |
|
452 XML_GetCurrentColumnNumber(parser), |
|
453 XML_GetCurrentLineNumber(parser)); |
|
454 CharData_AppendString(storage, buffer); |
|
455 } |
|
456 |
|
457 static void XMLCALL |
|
458 end_element_event_handler2(void *userData, const XML_Char *name) |
|
459 { |
|
460 CharData *storage = (CharData *) userData; |
|
461 char buffer[100]; |
|
462 |
|
463 sprintf(buffer, |
|
464 "</%s> at col:%" XML_FMT_INT_MOD "u line:%"\ |
|
465 XML_FMT_INT_MOD "u\n", name, |
|
466 XML_GetCurrentColumnNumber(parser), |
|
467 XML_GetCurrentLineNumber(parser)); |
|
468 CharData_AppendString(storage, buffer); |
|
469 } |
|
470 |
|
471 /* Regression test #3 for SF bug #653180. */ |
|
472 START_TEST(test_line_and_column_numbers_inside_handlers) |
|
473 { |
|
474 char *text = |
|
475 "<a>\n" /* Unix end-of-line */ |
|
476 " <b>\r\n" /* Windows end-of-line */ |
|
477 " <c/>\r" /* Mac OS end-of-line */ |
|
478 " </b>\n" |
|
479 " <d>\n" |
|
480 " <f/>\n" |
|
481 " </d>\n" |
|
482 "</a>"; |
|
483 char *expected = |
|
484 "<a> at col:0 line:1\n" |
|
485 "<b> at col:2 line:2\n" |
|
486 "<c> at col:4 line:3\n" |
|
487 "</c> at col:8 line:3\n" |
|
488 "</b> at col:2 line:4\n" |
|
489 "<d> at col:2 line:5\n" |
|
490 "<f> at col:4 line:6\n" |
|
491 "</f> at col:8 line:6\n" |
|
492 "</d> at col:2 line:7\n" |
|
493 "</a> at col:0 line:8\n"; |
|
494 CharData storage; |
|
495 |
|
496 CharData_Init(&storage); |
|
497 XML_SetUserData(parser, &storage); |
|
498 XML_SetStartElementHandler(parser, start_element_event_handler2); |
|
499 XML_SetEndElementHandler(parser, end_element_event_handler2); |
|
500 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
501 xml_failure(parser); |
|
502 |
|
503 CharData_CheckString(&storage, expected); |
|
504 } |
|
505 END_TEST |
|
506 |
|
507 /* Regression test #4 for SF bug #653180. */ |
|
508 START_TEST(test_line_number_after_error) |
|
509 { |
|
510 char *text = |
|
511 "<a>\n" |
|
512 " <b>\n" |
|
513 " </a>"; /* missing </b> */ |
|
514 XML_Size lineno; |
|
515 if (XML_Parse(parser, text, strlen(text), XML_FALSE) != XML_STATUS_ERROR) |
|
516 fail("Expected a parse error"); |
|
517 |
|
518 lineno = XML_GetCurrentLineNumber(parser); |
|
519 if (lineno != 3) { |
|
520 char buffer[100]; |
|
521 sprintf(buffer, "expected 3 lines, saw %" XML_FMT_INT_MOD "u", lineno); |
|
522 fail(buffer); |
|
523 } |
|
524 } |
|
525 END_TEST |
|
526 |
|
527 /* Regression test #5 for SF bug #653180. */ |
|
528 START_TEST(test_column_number_after_error) |
|
529 { |
|
530 char *text = |
|
531 "<a>\n" |
|
532 " <b>\n" |
|
533 " </a>"; /* missing </b> */ |
|
534 XML_Size colno; |
|
535 if (XML_Parse(parser, text, strlen(text), XML_FALSE) != XML_STATUS_ERROR) |
|
536 fail("Expected a parse error"); |
|
537 |
|
538 colno = XML_GetCurrentColumnNumber(parser); |
|
539 if (colno != 4) { |
|
540 char buffer[100]; |
|
541 sprintf(buffer, |
|
542 "expected 4 columns, saw %" XML_FMT_INT_MOD "u", colno); |
|
543 fail(buffer); |
|
544 } |
|
545 } |
|
546 END_TEST |
|
547 |
|
548 /* Regression test for SF bug #478332. */ |
|
549 START_TEST(test_really_long_lines) |
|
550 { |
|
551 /* This parses an input line longer than INIT_DATA_BUF_SIZE |
|
552 characters long (defined to be 1024 in xmlparse.c). We take a |
|
553 really cheesy approach to building the input buffer, because |
|
554 this avoids writing bugs in buffer-filling code. |
|
555 */ |
|
556 char *text = |
|
557 "<e>" |
|
558 /* 64 chars */ |
|
559 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
560 /* until we have at least 1024 characters on the line: */ |
|
561 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
562 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
563 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
564 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
565 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
566 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
567 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
568 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
569 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
570 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
571 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
572 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
573 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
574 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
575 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
576 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-+" |
|
577 "</e>"; |
|
578 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
579 xml_failure(parser); |
|
580 } |
|
581 END_TEST |
|
582 |
|
583 |
|
584 /* |
|
585 * Element event tests. |
|
586 */ |
|
587 |
|
588 static void XMLCALL |
|
589 end_element_event_handler(void *userData, const XML_Char *name) |
|
590 { |
|
591 CharData *storage = (CharData *) userData; |
|
592 CharData_AppendString(storage, "/"); |
|
593 CharData_AppendXMLChars(storage, name, -1); |
|
594 } |
|
595 |
|
596 START_TEST(test_end_element_events) |
|
597 { |
|
598 char *text = "<a><b><c/></b><d><f/></d></a>"; |
|
599 char *expected = "/c/b/f/d/a"; |
|
600 CharData storage; |
|
601 |
|
602 CharData_Init(&storage); |
|
603 XML_SetUserData(parser, &storage); |
|
604 XML_SetEndElementHandler(parser, end_element_event_handler); |
|
605 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
606 xml_failure(parser); |
|
607 CharData_CheckString(&storage, expected); |
|
608 } |
|
609 END_TEST |
|
610 |
|
611 |
|
612 /* |
|
613 * Attribute tests. |
|
614 */ |
|
615 |
|
616 /* Helpers used by the following test; this checks any "attr" and "refs" |
|
617 attributes to make sure whitespace has been normalized. |
|
618 |
|
619 Return true if whitespace has been normalized in a string, using |
|
620 the rules for attribute value normalization. The 'is_cdata' flag |
|
621 is needed since CDATA attributes don't need to have multiple |
|
622 whitespace characters collapsed to a single space, while other |
|
623 attribute data types do. (Section 3.3.3 of the recommendation.) |
|
624 */ |
|
625 static int |
|
626 is_whitespace_normalized(const XML_Char *s, int is_cdata) |
|
627 { |
|
628 int blanks = 0; |
|
629 int at_start = 1; |
|
630 while (*s) { |
|
631 if (*s == ' ') |
|
632 ++blanks; |
|
633 else if (*s == '\t' || *s == '\n' || *s == '\r') |
|
634 return 0; |
|
635 else { |
|
636 if (at_start) { |
|
637 at_start = 0; |
|
638 if (blanks && !is_cdata) |
|
639 /* illegal leading blanks */ |
|
640 return 0; |
|
641 } |
|
642 else if (blanks > 1 && !is_cdata) |
|
643 return 0; |
|
644 blanks = 0; |
|
645 } |
|
646 ++s; |
|
647 } |
|
648 if (blanks && !is_cdata) |
|
649 return 0; |
|
650 return 1; |
|
651 } |
|
652 |
|
653 /* Check the attribute whitespace checker: */ |
|
654 static void |
|
655 testhelper_is_whitespace_normalized(void) |
|
656 { |
|
657 assert(is_whitespace_normalized("abc", 0)); |
|
658 assert(is_whitespace_normalized("abc", 1)); |
|
659 assert(is_whitespace_normalized("abc def ghi", 0)); |
|
660 assert(is_whitespace_normalized("abc def ghi", 1)); |
|
661 assert(!is_whitespace_normalized(" abc def ghi", 0)); |
|
662 assert(is_whitespace_normalized(" abc def ghi", 1)); |
|
663 assert(!is_whitespace_normalized("abc def ghi", 0)); |
|
664 assert(is_whitespace_normalized("abc def ghi", 1)); |
|
665 assert(!is_whitespace_normalized("abc def ghi ", 0)); |
|
666 assert(is_whitespace_normalized("abc def ghi ", 1)); |
|
667 assert(!is_whitespace_normalized(" ", 0)); |
|
668 assert(is_whitespace_normalized(" ", 1)); |
|
669 assert(!is_whitespace_normalized("\t", 0)); |
|
670 assert(!is_whitespace_normalized("\t", 1)); |
|
671 assert(!is_whitespace_normalized("\n", 0)); |
|
672 assert(!is_whitespace_normalized("\n", 1)); |
|
673 assert(!is_whitespace_normalized("\r", 0)); |
|
674 assert(!is_whitespace_normalized("\r", 1)); |
|
675 assert(!is_whitespace_normalized("abc\t def", 1)); |
|
676 } |
|
677 |
|
678 static void XMLCALL |
|
679 check_attr_contains_normalized_whitespace(void *userData, |
|
680 const XML_Char *name, |
|
681 const XML_Char **atts) |
|
682 { |
|
683 int i; |
|
684 for (i = 0; atts[i] != NULL; i += 2) { |
|
685 const XML_Char *attrname = atts[i]; |
|
686 const XML_Char *value = atts[i + 1]; |
|
687 if (strcmp("attr", attrname) == 0 |
|
688 || strcmp("ents", attrname) == 0 |
|
689 || strcmp("refs", attrname) == 0) { |
|
690 if (!is_whitespace_normalized(value, 0)) { |
|
691 char buffer[256]; |
|
692 sprintf(buffer, "attribute value not normalized: %s='%s'", |
|
693 attrname, value); |
|
694 fail(buffer); |
|
695 } |
|
696 } |
|
697 } |
|
698 } |
|
699 |
|
700 START_TEST(test_attr_whitespace_normalization) |
|
701 { |
|
702 char *text = |
|
703 "<!DOCTYPE doc [\n" |
|
704 " <!ATTLIST doc\n" |
|
705 " attr NMTOKENS #REQUIRED\n" |
|
706 " ents ENTITIES #REQUIRED\n" |
|
707 " refs IDREFS #REQUIRED>\n" |
|
708 "]>\n" |
|
709 "<doc attr=' a b c\t\td\te\t' refs=' id-1 \t id-2\t\t' \n" |
|
710 " ents=' ent-1 \t\r\n" |
|
711 " ent-2 ' >\n" |
|
712 " <e id='id-1'/>\n" |
|
713 " <e id='id-2'/>\n" |
|
714 "</doc>"; |
|
715 |
|
716 XML_SetStartElementHandler(parser, |
|
717 check_attr_contains_normalized_whitespace); |
|
718 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
719 xml_failure(parser); |
|
720 } |
|
721 END_TEST |
|
722 |
|
723 |
|
724 /* |
|
725 * XML declaration tests. |
|
726 */ |
|
727 |
|
728 START_TEST(test_xmldecl_misplaced) |
|
729 { |
|
730 expect_failure("\n" |
|
731 "<?xml version='1.0'?>\n" |
|
732 "<a/>", |
|
733 XML_ERROR_MISPLACED_XML_PI, |
|
734 "failed to report misplaced XML declaration"); |
|
735 } |
|
736 END_TEST |
|
737 |
|
738 /* Regression test for SF bug #584832. */ |
|
739 static int XMLCALL |
|
740 UnknownEncodingHandler(void *data,const XML_Char *encoding,XML_Encoding *info) |
|
741 { |
|
742 if (strcmp(encoding,"unsupported-encoding") == 0) { |
|
743 int i; |
|
744 for (i = 0; i < 256; ++i) |
|
745 info->map[i] = i; |
|
746 info->data = NULL; |
|
747 info->convert = NULL; |
|
748 info->release = NULL; |
|
749 return XML_STATUS_OK; |
|
750 } |
|
751 return XML_STATUS_ERROR; |
|
752 } |
|
753 |
|
754 START_TEST(test_unknown_encoding_internal_entity) |
|
755 { |
|
756 char *text = |
|
757 "<?xml version='1.0' encoding='unsupported-encoding'?>\n" |
|
758 "<!DOCTYPE test [<!ENTITY foo 'bar'>]>\n" |
|
759 "<test a='&foo;'/>"; |
|
760 |
|
761 XML_SetUnknownEncodingHandler(parser, UnknownEncodingHandler, NULL); |
|
762 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
763 xml_failure(parser); |
|
764 } |
|
765 END_TEST |
|
766 |
|
767 /* Regression test for SF bug #620106. */ |
|
768 static int XMLCALL |
|
769 external_entity_loader_set_encoding(XML_Parser parser, |
|
770 const XML_Char *context, |
|
771 const XML_Char *base, |
|
772 const XML_Char *systemId, |
|
773 const XML_Char *publicId) |
|
774 { |
|
775 /* This text says it's an unsupported encoding, but it's really |
|
776 UTF-8, which we tell Expat using XML_SetEncoding(). |
|
777 */ |
|
778 char *text = |
|
779 "<?xml encoding='iso-8859-3'?>" |
|
780 "\xC3\xA9"; |
|
781 XML_Parser extparser; |
|
782 |
|
783 extparser = XML_ExternalEntityParserCreate(parser, context, NULL); |
|
784 if (extparser == NULL) |
|
785 fail("Could not create external entity parser."); |
|
786 if (!XML_SetEncoding(extparser, "utf-8")) |
|
787 fail("XML_SetEncoding() ignored for external entity"); |
|
788 if ( XML_Parse(extparser, text, strlen(text), XML_TRUE) |
|
789 == XML_STATUS_ERROR) { |
|
790 xml_failure(parser); |
|
791 return 0; |
|
792 } |
|
793 return 1; |
|
794 } |
|
795 |
|
796 START_TEST(test_ext_entity_set_encoding) |
|
797 { |
|
798 char *text = |
|
799 "<!DOCTYPE doc [\n" |
|
800 " <!ENTITY en SYSTEM 'http://xml.libexpat.org/dummy.ent'>\n" |
|
801 "]>\n" |
|
802 "<doc>&en;</doc>"; |
|
803 |
|
804 XML_SetExternalEntityRefHandler(parser, |
|
805 external_entity_loader_set_encoding); |
|
806 run_character_check(text, "\xC3\xA9"); |
|
807 } |
|
808 END_TEST |
|
809 |
|
810 /* Test that no error is reported for unknown entities if we don't |
|
811 read an external subset. This was fixed in Expat 1.95.5. |
|
812 */ |
|
813 START_TEST(test_wfc_undeclared_entity_unread_external_subset) { |
|
814 char *text = |
|
815 "<!DOCTYPE doc SYSTEM 'foo'>\n" |
|
816 "<doc>&entity;</doc>"; |
|
817 |
|
818 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
819 xml_failure(parser); |
|
820 } |
|
821 END_TEST |
|
822 |
|
823 /* Test that an error is reported for unknown entities if we don't |
|
824 have an external subset. |
|
825 */ |
|
826 START_TEST(test_wfc_undeclared_entity_no_external_subset) { |
|
827 expect_failure("<doc>&entity;</doc>", |
|
828 XML_ERROR_UNDEFINED_ENTITY, |
|
829 "Parser did not report undefined entity w/out a DTD."); |
|
830 } |
|
831 END_TEST |
|
832 |
|
833 /* Test that an error is reported for unknown entities if we don't |
|
834 read an external subset, but have been declared standalone. |
|
835 */ |
|
836 START_TEST(test_wfc_undeclared_entity_standalone) { |
|
837 char *text = |
|
838 "<?xml version='1.0' encoding='us-ascii' standalone='yes'?>\n" |
|
839 "<!DOCTYPE doc SYSTEM 'foo'>\n" |
|
840 "<doc>&entity;</doc>"; |
|
841 |
|
842 expect_failure(text, |
|
843 XML_ERROR_UNDEFINED_ENTITY, |
|
844 "Parser did not report undefined entity (standalone)."); |
|
845 } |
|
846 END_TEST |
|
847 |
|
848 static int XMLCALL |
|
849 external_entity_loader(XML_Parser parser, |
|
850 const XML_Char *context, |
|
851 const XML_Char *base, |
|
852 const XML_Char *systemId, |
|
853 const XML_Char *publicId) |
|
854 { |
|
855 char *text = (char *)XML_GetUserData(parser); |
|
856 XML_Parser extparser; |
|
857 |
|
858 extparser = XML_ExternalEntityParserCreate(parser, context, NULL); |
|
859 if (extparser == NULL) |
|
860 fail("Could not create external entity parser."); |
|
861 if ( XML_Parse(extparser, text, strlen(text), XML_TRUE) |
|
862 == XML_STATUS_ERROR) { |
|
863 xml_failure(parser); |
|
864 return XML_STATUS_ERROR; |
|
865 } |
|
866 return XML_STATUS_OK; |
|
867 } |
|
868 |
|
869 /* Test that an error is reported for unknown entities if we have read |
|
870 an external subset, and standalone is true. |
|
871 */ |
|
872 START_TEST(test_wfc_undeclared_entity_with_external_subset_standalone) { |
|
873 char *text = |
|
874 "<?xml version='1.0' encoding='us-ascii' standalone='yes'?>\n" |
|
875 "<!DOCTYPE doc SYSTEM 'foo'>\n" |
|
876 "<doc>&entity;</doc>"; |
|
877 char *foo_text = |
|
878 "<!ELEMENT doc (#PCDATA)*>"; |
|
879 |
|
880 XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS); |
|
881 XML_SetUserData(parser, foo_text); |
|
882 XML_SetExternalEntityRefHandler(parser, external_entity_loader); |
|
883 expect_failure(text, |
|
884 XML_ERROR_UNDEFINED_ENTITY, |
|
885 "Parser did not report undefined entity (external DTD)."); |
|
886 } |
|
887 END_TEST |
|
888 |
|
889 /* Test that no error is reported for unknown entities if we have read |
|
890 an external subset, and standalone is false. |
|
891 */ |
|
892 START_TEST(test_wfc_undeclared_entity_with_external_subset) { |
|
893 char *text = |
|
894 "<?xml version='1.0' encoding='us-ascii'?>\n" |
|
895 "<!DOCTYPE doc SYSTEM 'foo'>\n" |
|
896 "<doc>&entity;</doc>"; |
|
897 char *foo_text = |
|
898 "<!ELEMENT doc (#PCDATA)*>"; |
|
899 |
|
900 XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS); |
|
901 XML_SetUserData(parser, foo_text); |
|
902 XML_SetExternalEntityRefHandler(parser, external_entity_loader); |
|
903 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
904 xml_failure(parser); |
|
905 } |
|
906 END_TEST |
|
907 |
|
908 START_TEST(test_wfc_no_recursive_entity_refs) |
|
909 { |
|
910 char *text = |
|
911 "<!DOCTYPE doc [\n" |
|
912 " <!ENTITY entity '&entity;'>\n" |
|
913 "]>\n" |
|
914 "<doc>&entity;</doc>"; |
|
915 |
|
916 expect_failure(text, |
|
917 XML_ERROR_RECURSIVE_ENTITY_REF, |
|
918 "Parser did not report recursive entity reference."); |
|
919 } |
|
920 END_TEST |
|
921 |
|
922 /* Regression test for SF bug #483514. */ |
|
923 START_TEST(test_dtd_default_handling) |
|
924 { |
|
925 char *text = |
|
926 "<!DOCTYPE doc [\n" |
|
927 "<!ENTITY e SYSTEM 'http://xml.libexpat.org/e'>\n" |
|
928 "<!NOTATION n SYSTEM 'http://xml.libexpat.org/n'>\n" |
|
929 "<!ELEMENT doc EMPTY>\n" |
|
930 "<!ATTLIST doc a CDATA #IMPLIED>\n" |
|
931 "<?pi in dtd?>\n" |
|
932 "<!--comment in dtd-->\n" |
|
933 "]><doc/>"; |
|
934 |
|
935 XML_SetDefaultHandler(parser, accumulate_characters); |
|
936 XML_SetDoctypeDeclHandler(parser, |
|
937 dummy_start_doctype_handler, |
|
938 dummy_end_doctype_handler); |
|
939 XML_SetEntityDeclHandler(parser, dummy_entity_decl_handler); |
|
940 XML_SetNotationDeclHandler(parser, dummy_notation_decl_handler); |
|
941 XML_SetElementDeclHandler(parser, dummy_element_decl_handler); |
|
942 XML_SetAttlistDeclHandler(parser, dummy_attlist_decl_handler); |
|
943 XML_SetProcessingInstructionHandler(parser, dummy_pi_handler); |
|
944 XML_SetCommentHandler(parser, dummy_comment_handler); |
|
945 run_character_check(text, "\n\n\n\n\n\n\n<doc/>"); |
|
946 } |
|
947 END_TEST |
|
948 |
|
949 /* See related SF bug #673791. |
|
950 When namespace processing is enabled, setting the namespace URI for |
|
951 a prefix is not allowed; this test ensures that it *is* allowed |
|
952 when namespace processing is not enabled. |
|
953 (See Namespaces in XML, section 2.) |
|
954 */ |
|
955 START_TEST(test_empty_ns_without_namespaces) |
|
956 { |
|
957 char *text = |
|
958 "<doc xmlns:prefix='http://www.example.com/'>\n" |
|
959 " <e xmlns:prefix=''/>\n" |
|
960 "</doc>"; |
|
961 |
|
962 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
963 xml_failure(parser); |
|
964 } |
|
965 END_TEST |
|
966 |
|
967 /* Regression test for SF bug #824420. |
|
968 Checks that an xmlns:prefix attribute set in an attribute's default |
|
969 value isn't misinterpreted. |
|
970 */ |
|
971 START_TEST(test_ns_in_attribute_default_without_namespaces) |
|
972 { |
|
973 char *text = |
|
974 "<!DOCTYPE e:element [\n" |
|
975 " <!ATTLIST e:element\n" |
|
976 " xmlns:e CDATA 'http://example.com/'>\n" |
|
977 " ]>\n" |
|
978 "<e:element/>"; |
|
979 |
|
980 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
981 xml_failure(parser); |
|
982 } |
|
983 END_TEST |
|
984 |
|
985 |
|
986 /* |
|
987 * Namespaces tests. |
|
988 */ |
|
989 |
|
990 static void |
|
991 namespace_setup(void) |
|
992 { |
|
993 parser = XML_ParserCreateNS(NULL, ' '); |
|
994 if (parser == NULL) |
|
995 fail("Parser not created."); |
|
996 } |
|
997 |
|
998 static void |
|
999 namespace_teardown(void) |
|
1000 { |
|
1001 basic_teardown(); |
|
1002 } |
|
1003 |
|
1004 /* Check that an element name and attribute name match the expected values. |
|
1005 The expected values are passed as an array reference of string pointers |
|
1006 provided as the userData argument; the first is the expected |
|
1007 element name, and the second is the expected attribute name. |
|
1008 */ |
|
1009 static void XMLCALL |
|
1010 triplet_start_checker(void *userData, const XML_Char *name, |
|
1011 const XML_Char **atts) |
|
1012 { |
|
1013 char **elemstr = (char **)userData; |
|
1014 char buffer[1024]; |
|
1015 if (strcmp(elemstr[0], name) != 0) { |
|
1016 sprintf(buffer, "unexpected start string: '%s'", name); |
|
1017 fail(buffer); |
|
1018 } |
|
1019 if (strcmp(elemstr[1], atts[0]) != 0) { |
|
1020 sprintf(buffer, "unexpected attribute string: '%s'", atts[0]); |
|
1021 fail(buffer); |
|
1022 } |
|
1023 } |
|
1024 |
|
1025 /* Check that the element name passed to the end-element handler matches |
|
1026 the expected value. The expected value is passed as the first element |
|
1027 in an array of strings passed as the userData argument. |
|
1028 */ |
|
1029 static void XMLCALL |
|
1030 triplet_end_checker(void *userData, const XML_Char *name) |
|
1031 { |
|
1032 char **elemstr = (char **)userData; |
|
1033 if (strcmp(elemstr[0], name) != 0) { |
|
1034 char buffer[1024]; |
|
1035 sprintf(buffer, "unexpected end string: '%s'", name); |
|
1036 fail(buffer); |
|
1037 } |
|
1038 } |
|
1039 |
|
1040 START_TEST(test_return_ns_triplet) |
|
1041 { |
|
1042 char *text = |
|
1043 "<foo:e xmlns:foo='http://expat.sf.net/' bar:a='12'\n" |
|
1044 " xmlns:bar='http://expat.sf.net/'></foo:e>"; |
|
1045 char *elemstr[] = { |
|
1046 "http://expat.sf.net/ e foo", |
|
1047 "http://expat.sf.net/ a bar" |
|
1048 }; |
|
1049 XML_SetReturnNSTriplet(parser, XML_TRUE); |
|
1050 XML_SetUserData(parser, elemstr); |
|
1051 XML_SetElementHandler(parser, triplet_start_checker, triplet_end_checker); |
|
1052 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
1053 xml_failure(parser); |
|
1054 } |
|
1055 END_TEST |
|
1056 |
|
1057 static void XMLCALL |
|
1058 overwrite_start_checker(void *userData, const XML_Char *name, |
|
1059 const XML_Char **atts) |
|
1060 { |
|
1061 CharData *storage = (CharData *) userData; |
|
1062 CharData_AppendString(storage, "start "); |
|
1063 CharData_AppendXMLChars(storage, name, -1); |
|
1064 while (*atts != NULL) { |
|
1065 CharData_AppendString(storage, "\nattribute "); |
|
1066 CharData_AppendXMLChars(storage, *atts, -1); |
|
1067 atts += 2; |
|
1068 } |
|
1069 CharData_AppendString(storage, "\n"); |
|
1070 } |
|
1071 |
|
1072 static void XMLCALL |
|
1073 overwrite_end_checker(void *userData, const XML_Char *name) |
|
1074 { |
|
1075 CharData *storage = (CharData *) userData; |
|
1076 CharData_AppendString(storage, "end "); |
|
1077 CharData_AppendXMLChars(storage, name, -1); |
|
1078 CharData_AppendString(storage, "\n"); |
|
1079 } |
|
1080 |
|
1081 static void |
|
1082 run_ns_tagname_overwrite_test(char *text, char *result) |
|
1083 { |
|
1084 CharData storage; |
|
1085 CharData_Init(&storage); |
|
1086 XML_SetUserData(parser, &storage); |
|
1087 XML_SetElementHandler(parser, |
|
1088 overwrite_start_checker, overwrite_end_checker); |
|
1089 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
1090 xml_failure(parser); |
|
1091 CharData_CheckString(&storage, result); |
|
1092 } |
|
1093 |
|
1094 /* Regression test for SF bug #566334. */ |
|
1095 START_TEST(test_ns_tagname_overwrite) |
|
1096 { |
|
1097 char *text = |
|
1098 "<n:e xmlns:n='http://xml.libexpat.org/'>\n" |
|
1099 " <n:f n:attr='foo'/>\n" |
|
1100 " <n:g n:attr2='bar'/>\n" |
|
1101 "</n:e>"; |
|
1102 char *result = |
|
1103 "start http://xml.libexpat.org/ e\n" |
|
1104 "start http://xml.libexpat.org/ f\n" |
|
1105 "attribute http://xml.libexpat.org/ attr\n" |
|
1106 "end http://xml.libexpat.org/ f\n" |
|
1107 "start http://xml.libexpat.org/ g\n" |
|
1108 "attribute http://xml.libexpat.org/ attr2\n" |
|
1109 "end http://xml.libexpat.org/ g\n" |
|
1110 "end http://xml.libexpat.org/ e\n"; |
|
1111 run_ns_tagname_overwrite_test(text, result); |
|
1112 } |
|
1113 END_TEST |
|
1114 |
|
1115 /* Regression test for SF bug #566334. */ |
|
1116 START_TEST(test_ns_tagname_overwrite_triplet) |
|
1117 { |
|
1118 char *text = |
|
1119 "<n:e xmlns:n='http://xml.libexpat.org/'>\n" |
|
1120 " <n:f n:attr='foo'/>\n" |
|
1121 " <n:g n:attr2='bar'/>\n" |
|
1122 "</n:e>"; |
|
1123 char *result = |
|
1124 "start http://xml.libexpat.org/ e n\n" |
|
1125 "start http://xml.libexpat.org/ f n\n" |
|
1126 "attribute http://xml.libexpat.org/ attr n\n" |
|
1127 "end http://xml.libexpat.org/ f n\n" |
|
1128 "start http://xml.libexpat.org/ g n\n" |
|
1129 "attribute http://xml.libexpat.org/ attr2 n\n" |
|
1130 "end http://xml.libexpat.org/ g n\n" |
|
1131 "end http://xml.libexpat.org/ e n\n"; |
|
1132 XML_SetReturnNSTriplet(parser, XML_TRUE); |
|
1133 run_ns_tagname_overwrite_test(text, result); |
|
1134 } |
|
1135 END_TEST |
|
1136 |
|
1137 |
|
1138 /* Regression test for SF bug #620343. */ |
|
1139 static void XMLCALL |
|
1140 start_element_fail(void *userData, |
|
1141 const XML_Char *name, const XML_Char **atts) |
|
1142 { |
|
1143 /* We should never get here. */ |
|
1144 fail("should never reach start_element_fail()"); |
|
1145 } |
|
1146 |
|
1147 static void XMLCALL |
|
1148 start_ns_clearing_start_element(void *userData, |
|
1149 const XML_Char *prefix, |
|
1150 const XML_Char *uri) |
|
1151 { |
|
1152 XML_SetStartElementHandler((XML_Parser) userData, NULL); |
|
1153 } |
|
1154 |
|
1155 START_TEST(test_start_ns_clears_start_element) |
|
1156 { |
|
1157 /* This needs to use separate start/end tags; using the empty tag |
|
1158 syntax doesn't cause the problematic path through Expat to be |
|
1159 taken. |
|
1160 */ |
|
1161 char *text = "<e xmlns='http://xml.libexpat.org/'></e>"; |
|
1162 |
|
1163 XML_SetStartElementHandler(parser, start_element_fail); |
|
1164 XML_SetStartNamespaceDeclHandler(parser, start_ns_clearing_start_element); |
|
1165 XML_UseParserAsHandlerArg(parser); |
|
1166 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
1167 xml_failure(parser); |
|
1168 } |
|
1169 END_TEST |
|
1170 |
|
1171 /* Regression test for SF bug #616863. */ |
|
1172 static int XMLCALL |
|
1173 external_entity_handler(XML_Parser parser, |
|
1174 const XML_Char *context, |
|
1175 const XML_Char *base, |
|
1176 const XML_Char *systemId, |
|
1177 const XML_Char *publicId) |
|
1178 { |
|
1179 int callno = 1 + (int)XML_GetUserData(parser); |
|
1180 char *text; |
|
1181 XML_Parser p2; |
|
1182 |
|
1183 if (callno == 1) |
|
1184 text = ("<!ELEMENT doc (e+)>\n" |
|
1185 "<!ATTLIST doc xmlns CDATA #IMPLIED>\n" |
|
1186 "<!ELEMENT e EMPTY>\n"); |
|
1187 else |
|
1188 text = ("<?xml version='1.0' encoding='us-ascii'?>" |
|
1189 "<e/>"); |
|
1190 |
|
1191 XML_SetUserData(parser, (void *) callno); |
|
1192 p2 = XML_ExternalEntityParserCreate(parser, context, NULL); |
|
1193 if (XML_Parse(p2, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) { |
|
1194 xml_failure(p2); |
|
1195 return 0; |
|
1196 } |
|
1197 XML_ParserFree(p2); |
|
1198 return 1; |
|
1199 } |
|
1200 |
|
1201 START_TEST(test_default_ns_from_ext_subset_and_ext_ge) |
|
1202 { |
|
1203 char *text = |
|
1204 "<?xml version='1.0'?>\n" |
|
1205 "<!DOCTYPE doc SYSTEM 'http://xml.libexpat.org/doc.dtd' [\n" |
|
1206 " <!ENTITY en SYSTEM 'http://xml.libexpat.org/entity.ent'>\n" |
|
1207 "]>\n" |
|
1208 "<doc xmlns='http://xml.libexpat.org/ns1'>\n" |
|
1209 "&en;\n" |
|
1210 "</doc>"; |
|
1211 |
|
1212 XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS); |
|
1213 XML_SetExternalEntityRefHandler(parser, external_entity_handler); |
|
1214 /* We actually need to set this handler to tickle this bug. */ |
|
1215 XML_SetStartElementHandler(parser, dummy_start_element); |
|
1216 XML_SetUserData(parser, NULL); |
|
1217 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
1218 xml_failure(parser); |
|
1219 } |
|
1220 END_TEST |
|
1221 |
|
1222 /* Regression test #1 for SF bug #673791. */ |
|
1223 START_TEST(test_ns_prefix_with_empty_uri_1) |
|
1224 { |
|
1225 char *text = |
|
1226 "<doc xmlns:prefix='http://xml.libexpat.org/'>\n" |
|
1227 " <e xmlns:prefix=''/>\n" |
|
1228 "</doc>"; |
|
1229 |
|
1230 expect_failure(text, |
|
1231 XML_ERROR_UNDECLARING_PREFIX, |
|
1232 "Did not report re-setting namespace" |
|
1233 " URI with prefix to ''."); |
|
1234 } |
|
1235 END_TEST |
|
1236 |
|
1237 /* Regression test #2 for SF bug #673791. */ |
|
1238 START_TEST(test_ns_prefix_with_empty_uri_2) |
|
1239 { |
|
1240 char *text = |
|
1241 "<?xml version='1.0'?>\n" |
|
1242 "<docelem xmlns:pre=''/>"; |
|
1243 |
|
1244 expect_failure(text, |
|
1245 XML_ERROR_UNDECLARING_PREFIX, |
|
1246 "Did not report setting namespace URI with prefix to ''."); |
|
1247 } |
|
1248 END_TEST |
|
1249 |
|
1250 /* Regression test #3 for SF bug #673791. */ |
|
1251 START_TEST(test_ns_prefix_with_empty_uri_3) |
|
1252 { |
|
1253 char *text = |
|
1254 "<!DOCTYPE doc [\n" |
|
1255 " <!ELEMENT doc EMPTY>\n" |
|
1256 " <!ATTLIST doc\n" |
|
1257 " xmlns:prefix CDATA ''>\n" |
|
1258 "]>\n" |
|
1259 "<doc/>"; |
|
1260 |
|
1261 expect_failure(text, |
|
1262 XML_ERROR_UNDECLARING_PREFIX, |
|
1263 "Didn't report attr default setting NS w/ prefix to ''."); |
|
1264 } |
|
1265 END_TEST |
|
1266 |
|
1267 /* Regression test #4 for SF bug #673791. */ |
|
1268 START_TEST(test_ns_prefix_with_empty_uri_4) |
|
1269 { |
|
1270 char *text = |
|
1271 "<!DOCTYPE doc [\n" |
|
1272 " <!ELEMENT prefix:doc EMPTY>\n" |
|
1273 " <!ATTLIST prefix:doc\n" |
|
1274 " xmlns:prefix CDATA 'http://xml.libexpat.org/'>\n" |
|
1275 "]>\n" |
|
1276 "<prefix:doc/>"; |
|
1277 /* Packaged info expected by the end element handler; |
|
1278 the weird structuring lets us re-use the triplet_end_checker() |
|
1279 function also used for another test. */ |
|
1280 char *elemstr[] = { |
|
1281 "http://xml.libexpat.org/ doc prefix" |
|
1282 }; |
|
1283 XML_SetReturnNSTriplet(parser, XML_TRUE); |
|
1284 XML_SetUserData(parser, elemstr); |
|
1285 XML_SetEndElementHandler(parser, triplet_end_checker); |
|
1286 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
1287 xml_failure(parser); |
|
1288 } |
|
1289 END_TEST |
|
1290 |
|
1291 START_TEST(test_ns_default_with_empty_uri) |
|
1292 { |
|
1293 char *text = |
|
1294 "<doc xmlns='http://xml.libexpat.org/'>\n" |
|
1295 " <e xmlns=''/>\n" |
|
1296 "</doc>"; |
|
1297 if (XML_Parse(parser, text, strlen(text), XML_TRUE) == XML_STATUS_ERROR) |
|
1298 xml_failure(parser); |
|
1299 } |
|
1300 END_TEST |
|
1301 |
|
1302 /* Regression test for SF bug #692964: two prefixes for one namespace. */ |
|
1303 START_TEST(test_ns_duplicate_attrs_diff_prefixes) |
|
1304 { |
|
1305 char *text = |
|
1306 "<doc xmlns:a='http://xml.libexpat.org/a'\n" |
|
1307 " xmlns:b='http://xml.libexpat.org/a'\n" |
|
1308 " a:a='v' b:a='v' />"; |
|
1309 expect_failure(text, |
|
1310 XML_ERROR_DUPLICATE_ATTRIBUTE, |
|
1311 "did not report multiple attributes with same URI+name"); |
|
1312 } |
|
1313 END_TEST |
|
1314 |
|
1315 /* Regression test for SF bug #695401: unbound prefix. */ |
|
1316 START_TEST(test_ns_unbound_prefix_on_attribute) |
|
1317 { |
|
1318 char *text = "<doc a:attr=''/>"; |
|
1319 expect_failure(text, |
|
1320 XML_ERROR_UNBOUND_PREFIX, |
|
1321 "did not report unbound prefix on attribute"); |
|
1322 } |
|
1323 END_TEST |
|
1324 |
|
1325 /* Regression test for SF bug #695401: unbound prefix. */ |
|
1326 START_TEST(test_ns_unbound_prefix_on_element) |
|
1327 { |
|
1328 char *text = "<a:doc/>"; |
|
1329 expect_failure(text, |
|
1330 XML_ERROR_UNBOUND_PREFIX, |
|
1331 "did not report unbound prefix on element"); |
|
1332 } |
|
1333 END_TEST |
|
1334 |
|
1335 static Suite * |
|
1336 make_suite(void) |
|
1337 { |
|
1338 Suite *s = suite_create("basic"); |
|
1339 TCase *tc_basic = tcase_create("basic tests"); |
|
1340 TCase *tc_namespace = tcase_create("XML namespaces"); |
|
1341 |
|
1342 suite_add_tcase(s, tc_basic); |
|
1343 tcase_add_checked_fixture(tc_basic, basic_setup, basic_teardown); |
|
1344 tcase_add_test(tc_basic, test_nul_byte); |
|
1345 tcase_add_test(tc_basic, test_u0000_char); |
|
1346 tcase_add_test(tc_basic, test_bom_utf8); |
|
1347 tcase_add_test(tc_basic, test_bom_utf16_be); |
|
1348 tcase_add_test(tc_basic, test_bom_utf16_le); |
|
1349 tcase_add_test(tc_basic, test_illegal_utf8); |
|
1350 tcase_add_test(tc_basic, test_utf16); |
|
1351 tcase_add_test(tc_basic, test_utf16_le_epilog_newline); |
|
1352 tcase_add_test(tc_basic, test_latin1_umlauts); |
|
1353 /* Regression test for SF bug #491986. */ |
|
1354 tcase_add_test(tc_basic, test_danish_latin1); |
|
1355 /* Regression test for SF bug #514281. */ |
|
1356 tcase_add_test(tc_basic, test_french_charref_hexidecimal); |
|
1357 tcase_add_test(tc_basic, test_french_charref_decimal); |
|
1358 tcase_add_test(tc_basic, test_french_latin1); |
|
1359 tcase_add_test(tc_basic, test_french_utf8); |
|
1360 tcase_add_test(tc_basic, test_utf8_false_rejection); |
|
1361 tcase_add_test(tc_basic, test_line_number_after_parse); |
|
1362 tcase_add_test(tc_basic, test_column_number_after_parse); |
|
1363 tcase_add_test(tc_basic, test_line_and_column_numbers_inside_handlers); |
|
1364 tcase_add_test(tc_basic, test_line_number_after_error); |
|
1365 tcase_add_test(tc_basic, test_column_number_after_error); |
|
1366 tcase_add_test(tc_basic, test_really_long_lines); |
|
1367 tcase_add_test(tc_basic, test_end_element_events); |
|
1368 tcase_add_test(tc_basic, test_attr_whitespace_normalization); |
|
1369 tcase_add_test(tc_basic, test_xmldecl_misplaced); |
|
1370 tcase_add_test(tc_basic, test_unknown_encoding_internal_entity); |
|
1371 tcase_add_test(tc_basic, |
|
1372 test_wfc_undeclared_entity_unread_external_subset); |
|
1373 tcase_add_test(tc_basic, test_wfc_undeclared_entity_no_external_subset); |
|
1374 tcase_add_test(tc_basic, test_wfc_undeclared_entity_standalone); |
|
1375 tcase_add_test(tc_basic, test_wfc_undeclared_entity_with_external_subset); |
|
1376 tcase_add_test(tc_basic, |
|
1377 test_wfc_undeclared_entity_with_external_subset_standalone); |
|
1378 tcase_add_test(tc_basic, test_wfc_no_recursive_entity_refs); |
|
1379 tcase_add_test(tc_basic, test_ext_entity_set_encoding); |
|
1380 tcase_add_test(tc_basic, test_dtd_default_handling); |
|
1381 tcase_add_test(tc_basic, test_empty_ns_without_namespaces); |
|
1382 tcase_add_test(tc_basic, test_ns_in_attribute_default_without_namespaces); |
|
1383 |
|
1384 suite_add_tcase(s, tc_namespace); |
|
1385 tcase_add_checked_fixture(tc_namespace, |
|
1386 namespace_setup, namespace_teardown); |
|
1387 tcase_add_test(tc_namespace, test_return_ns_triplet); |
|
1388 tcase_add_test(tc_namespace, test_ns_tagname_overwrite); |
|
1389 tcase_add_test(tc_namespace, test_ns_tagname_overwrite_triplet); |
|
1390 tcase_add_test(tc_namespace, test_start_ns_clears_start_element); |
|
1391 tcase_add_test(tc_namespace, test_default_ns_from_ext_subset_and_ext_ge); |
|
1392 tcase_add_test(tc_namespace, test_ns_prefix_with_empty_uri_1); |
|
1393 tcase_add_test(tc_namespace, test_ns_prefix_with_empty_uri_2); |
|
1394 tcase_add_test(tc_namespace, test_ns_prefix_with_empty_uri_3); |
|
1395 tcase_add_test(tc_namespace, test_ns_prefix_with_empty_uri_4); |
|
1396 tcase_add_test(tc_namespace, test_ns_default_with_empty_uri); |
|
1397 tcase_add_test(tc_namespace, test_ns_duplicate_attrs_diff_prefixes); |
|
1398 tcase_add_test(tc_namespace, test_ns_unbound_prefix_on_attribute); |
|
1399 tcase_add_test(tc_namespace, test_ns_unbound_prefix_on_element); |
|
1400 |
|
1401 return s; |
|
1402 } |
|
1403 |
|
1404 |
|
1405 #ifdef AMIGA_SHARED_LIB |
|
1406 int |
|
1407 amiga_main(int argc, char *argv[]) |
|
1408 #else |
|
1409 int |
|
1410 main(int argc, char *argv[]) |
|
1411 #endif |
|
1412 { |
|
1413 int i, nf; |
|
1414 int forking = 0, forking_set = 0; |
|
1415 int verbosity = CK_NORMAL; |
|
1416 Suite *s = make_suite(); |
|
1417 SRunner *sr = srunner_create(s); |
|
1418 |
|
1419 /* run the tests for internal helper functions */ |
|
1420 testhelper_is_whitespace_normalized(); |
|
1421 |
|
1422 for (i = 1; i < argc; ++i) { |
|
1423 char *opt = argv[i]; |
|
1424 if (strcmp(opt, "-v") == 0 || strcmp(opt, "--verbose") == 0) |
|
1425 verbosity = CK_VERBOSE; |
|
1426 else if (strcmp(opt, "-q") == 0 || strcmp(opt, "--quiet") == 0) |
|
1427 verbosity = CK_SILENT; |
|
1428 else if (strcmp(opt, "-f") == 0 || strcmp(opt, "--fork") == 0) { |
|
1429 forking = 1; |
|
1430 forking_set = 1; |
|
1431 } |
|
1432 else if (strcmp(opt, "-n") == 0 || strcmp(opt, "--no-fork") == 0) { |
|
1433 forking = 0; |
|
1434 forking_set = 1; |
|
1435 } |
|
1436 else { |
|
1437 fprintf(stderr, "runtests: unknown option '%s'\n", opt); |
|
1438 return 2; |
|
1439 } |
|
1440 } |
|
1441 if (forking_set) |
|
1442 srunner_set_fork_status(sr, forking ? CK_FORK : CK_NOFORK); |
|
1443 if (verbosity != CK_SILENT) |
|
1444 printf("Expat version: %s\n", XML_ExpatVersion()); |
|
1445 srunner_run_all(sr, verbosity); |
|
1446 nf = srunner_ntests_failed(sr); |
|
1447 srunner_free(sr); |
|
1448 |
|
1449 return (nf == 0) ? EXIT_SUCCESS : EXIT_FAILURE; |
|
1450 } |