forked from php/php-src
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathhtml.c
1665 lines (1473 loc) · 46.9 KB
/
html.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
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
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
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
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
519
520
521
522
523
524
525
526
527
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
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
+----------------------------------------------------------------------+
| Copyright (c) The PHP Group |
+----------------------------------------------------------------------+
| This source file is subject to version 3.01 of the PHP license, |
| that is bundled with this package in the file LICENSE, and is |
| available through the world-wide-web at the following url: |
| http://www.php.net/license/3_01.txt |
| If you did not receive a copy of the PHP license and are unable to |
| obtain it through the world-wide-web, please send a note to |
| [email protected] so we can mail you a copy immediately. |
+----------------------------------------------------------------------+
| Authors: Rasmus Lerdorf <[email protected]> |
| Jaakko Hyvätti <[email protected]> |
| Wez Furlong <[email protected]> |
| Gustavo Lopes <[email protected]> |
+----------------------------------------------------------------------+
*/
/*
* HTML entity resources:
*
* http://www.unicode.org/Public/MAPPINGS/OBSOLETE/UNI2SGML.TXT
*
* XHTML 1.0 DTD
* http://www.w3.org/TR/2002/REC-xhtml1-20020801/dtds.html#h-A2
*
* From HTML 4.01 strict DTD:
* http://www.w3.org/TR/html4/HTMLlat1.ent
* http://www.w3.org/TR/html4/HTMLsymbol.ent
* http://www.w3.org/TR/html4/HTMLspecial.ent
*
* HTML 5:
* http://dev.w3.org/html5/spec/Overview.html#named-character-references
*/
#include "php.h"
#ifdef PHP_WIN32
#include "config.w32.h"
#else
#include <php_config.h>
#endif
#include "php_standard.h"
#include "php_string.h"
#include "SAPI.h"
#include <locale.h>
#if HAVE_LANGINFO_H
#include <langinfo.h>
#endif
#include <zend_hash.h>
#include "html_tables.h"
/* Macro for disabling flag of translation of non-basic entities where this isn't supported.
* Not appropriate for html_entity_decode/htmlspecialchars_decode */
#define LIMIT_ALL(all, doctype, charset) do { \
(all) = (all) && !CHARSET_PARTIAL_SUPPORT((charset)) && ((doctype) != ENT_HTML_DOC_XML1); \
} while (0)
#define MB_FAILURE(pos, advance) do { \
*cursor = pos + (advance); \
*status = FAILURE; \
return 0; \
} while (0)
#define CHECK_LEN(pos, chars_need) ((str_len - (pos)) >= (chars_need))
/* valid as single byte character or leading byte */
#define utf8_lead(c) ((c) < 0x80 || ((c) >= 0xC2 && (c) <= 0xF4))
/* whether it's actually valid depends on other stuff;
* this macro cannot check for non-shortest forms, surrogates or
* code points above 0x10FFFF */
#define utf8_trail(c) ((c) >= 0x80 && (c) <= 0xBF)
#define gb2312_lead(c) ((c) != 0x8E && (c) != 0x8F && (c) != 0xA0 && (c) != 0xFF)
#define gb2312_trail(c) ((c) >= 0xA1 && (c) <= 0xFE)
#define sjis_lead(c) ((c) != 0x80 && (c) != 0xA0 && (c) < 0xFD)
#define sjis_trail(c) ((c) >= 0x40 && (c) != 0x7F && (c) < 0xFD)
/* {{{ get_default_charset
*/
static char *get_default_charset(void) {
if (PG(internal_encoding) && PG(internal_encoding)[0]) {
return PG(internal_encoding);
} else if (SG(default_charset) && SG(default_charset)[0] ) {
return SG(default_charset);
}
return NULL;
}
/* }}} */
/* {{{ get_next_char
*/
static inline unsigned int get_next_char(
enum entity_charset charset,
const unsigned char *str,
size_t str_len,
size_t *cursor,
int *status)
{
size_t pos = *cursor;
unsigned int this_char = 0;
*status = SUCCESS;
assert(pos <= str_len);
if (!CHECK_LEN(pos, 1))
MB_FAILURE(pos, 1);
switch (charset) {
case cs_utf_8:
{
/* We'll follow strategy 2. from section 3.6.1 of UTR #36:
* "In a reported illegal byte sequence, do not include any
* non-initial byte that encodes a valid character or is a leading
* byte for a valid sequence." */
unsigned char c;
c = str[pos];
if (c < 0x80) {
this_char = c;
pos++;
} else if (c < 0xc2) {
MB_FAILURE(pos, 1);
} else if (c < 0xe0) {
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
if (!utf8_trail(str[pos + 1])) {
MB_FAILURE(pos, utf8_lead(str[pos + 1]) ? 1 : 2);
}
this_char = ((c & 0x1f) << 6) | (str[pos + 1] & 0x3f);
if (this_char < 0x80) { /* non-shortest form */
MB_FAILURE(pos, 2);
}
pos += 2;
} else if (c < 0xf0) {
size_t avail = str_len - pos;
if (avail < 3 ||
!utf8_trail(str[pos + 1]) || !utf8_trail(str[pos + 2])) {
if (avail < 2 || utf8_lead(str[pos + 1]))
MB_FAILURE(pos, 1);
else if (avail < 3 || utf8_lead(str[pos + 2]))
MB_FAILURE(pos, 2);
else
MB_FAILURE(pos, 3);
}
this_char = ((c & 0x0f) << 12) | ((str[pos + 1] & 0x3f) << 6) | (str[pos + 2] & 0x3f);
if (this_char < 0x800) { /* non-shortest form */
MB_FAILURE(pos, 3);
} else if (this_char >= 0xd800 && this_char <= 0xdfff) { /* surrogate */
MB_FAILURE(pos, 3);
}
pos += 3;
} else if (c < 0xf5) {
size_t avail = str_len - pos;
if (avail < 4 ||
!utf8_trail(str[pos + 1]) || !utf8_trail(str[pos + 2]) ||
!utf8_trail(str[pos + 3])) {
if (avail < 2 || utf8_lead(str[pos + 1]))
MB_FAILURE(pos, 1);
else if (avail < 3 || utf8_lead(str[pos + 2]))
MB_FAILURE(pos, 2);
else if (avail < 4 || utf8_lead(str[pos + 3]))
MB_FAILURE(pos, 3);
else
MB_FAILURE(pos, 4);
}
this_char = ((c & 0x07) << 18) | ((str[pos + 1] & 0x3f) << 12) | ((str[pos + 2] & 0x3f) << 6) | (str[pos + 3] & 0x3f);
if (this_char < 0x10000 || this_char > 0x10FFFF) { /* non-shortest form or outside range */
MB_FAILURE(pos, 4);
}
pos += 4;
} else {
MB_FAILURE(pos, 1);
}
}
break;
case cs_big5:
/* reference http://demo.icu-project.org/icu-bin/convexp?conv=big5 */
{
unsigned char c = str[pos];
if (c >= 0x81 && c <= 0xFE) {
unsigned char next;
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
next = str[pos + 1];
if ((next >= 0x40 && next <= 0x7E) ||
(next >= 0xA1 && next <= 0xFE)) {
this_char = (c << 8) | next;
} else {
MB_FAILURE(pos, 1);
}
pos += 2;
} else {
this_char = c;
pos += 1;
}
}
break;
case cs_big5hkscs:
{
unsigned char c = str[pos];
if (c >= 0x81 && c <= 0xFE) {
unsigned char next;
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
next = str[pos + 1];
if ((next >= 0x40 && next <= 0x7E) ||
(next >= 0xA1 && next <= 0xFE)) {
this_char = (c << 8) | next;
} else if (next != 0x80 && next != 0xFF) {
MB_FAILURE(pos, 1);
} else {
MB_FAILURE(pos, 2);
}
pos += 2;
} else {
this_char = c;
pos += 1;
}
}
break;
case cs_gb2312: /* EUC-CN */
{
unsigned char c = str[pos];
if (c >= 0xA1 && c <= 0xFE) {
unsigned char next;
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
next = str[pos + 1];
if (gb2312_trail(next)) {
this_char = (c << 8) | next;
} else if (gb2312_lead(next)) {
MB_FAILURE(pos, 1);
} else {
MB_FAILURE(pos, 2);
}
pos += 2;
} else if (gb2312_lead(c)) {
this_char = c;
pos += 1;
} else {
MB_FAILURE(pos, 1);
}
}
break;
case cs_sjis:
{
unsigned char c = str[pos];
if ((c >= 0x81 && c <= 0x9F) || (c >= 0xE0 && c <= 0xFC)) {
unsigned char next;
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
next = str[pos + 1];
if (sjis_trail(next)) {
this_char = (c << 8) | next;
} else if (sjis_lead(next)) {
MB_FAILURE(pos, 1);
} else {
MB_FAILURE(pos, 2);
}
pos += 2;
} else if (c < 0x80 || (c >= 0xA1 && c <= 0xDF)) {
this_char = c;
pos += 1;
} else {
MB_FAILURE(pos, 1);
}
}
break;
case cs_eucjp:
{
unsigned char c = str[pos];
if (c >= 0xA1 && c <= 0xFE) {
unsigned next;
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
next = str[pos + 1];
if (next >= 0xA1 && next <= 0xFE) {
/* this a jis kanji char */
this_char = (c << 8) | next;
} else {
MB_FAILURE(pos, (next != 0xA0 && next != 0xFF) ? 1 : 2);
}
pos += 2;
} else if (c == 0x8E) {
unsigned next;
if (!CHECK_LEN(pos, 2))
MB_FAILURE(pos, 1);
next = str[pos + 1];
if (next >= 0xA1 && next <= 0xDF) {
/* JIS X 0201 kana */
this_char = (c << 8) | next;
} else {
MB_FAILURE(pos, (next != 0xA0 && next != 0xFF) ? 1 : 2);
}
pos += 2;
} else if (c == 0x8F) {
size_t avail = str_len - pos;
if (avail < 3 || !(str[pos + 1] >= 0xA1 && str[pos + 1] <= 0xFE) ||
!(str[pos + 2] >= 0xA1 && str[pos + 2] <= 0xFE)) {
if (avail < 2 || (str[pos + 1] != 0xA0 && str[pos + 1] != 0xFF))
MB_FAILURE(pos, 1);
else if (avail < 3 || (str[pos + 2] != 0xA0 && str[pos + 2] != 0xFF))
MB_FAILURE(pos, 2);
else
MB_FAILURE(pos, 3);
} else {
/* JIS X 0212 hojo-kanji */
this_char = (c << 16) | (str[pos + 1] << 8) | str[pos + 2];
}
pos += 3;
} else if (c != 0xA0 && c != 0xFF) {
/* character encoded in 1 code unit */
this_char = c;
pos += 1;
} else {
MB_FAILURE(pos, 1);
}
}
break;
default:
/* single-byte charsets */
this_char = str[pos++];
break;
}
*cursor = pos;
return this_char;
}
/* }}} */
/* {{{ php_next_utf8_char
* Public interface for get_next_char used with UTF-8 */
PHPAPI unsigned int php_next_utf8_char(
const unsigned char *str,
size_t str_len,
size_t *cursor,
int *status)
{
return get_next_char(cs_utf_8, str, str_len, cursor, status);
}
/* }}} */
/* {{{ entity_charset determine_charset
* returns the charset identifier based on current locale or a hint.
* defaults to UTF-8 */
static enum entity_charset determine_charset(char *charset_hint)
{
size_t i;
enum entity_charset charset = cs_utf_8;
size_t len = 0;
const zend_encoding *zenc;
/* Default is now UTF-8 */
if (charset_hint == NULL)
return cs_utf_8;
if ((len = strlen(charset_hint)) != 0) {
goto det_charset;
}
zenc = zend_multibyte_get_internal_encoding();
if (zenc != NULL) {
charset_hint = (char *)zend_multibyte_get_encoding_name(zenc);
if (charset_hint != NULL && (len=strlen(charset_hint)) != 0) {
if ((len == 4) /* sizeof (auto|pass) */ &&
/* XXX should the "wchar" be ignored as well?? */
(!memcmp("pass", charset_hint, 4) ||
!memcmp("auto", charset_hint, 4))) {
charset_hint = NULL;
len = 0;
} else {
goto det_charset;
}
}
}
charset_hint = SG(default_charset);
if (charset_hint != NULL && (len=strlen(charset_hint)) != 0) {
goto det_charset;
}
/* try to detect the charset for the locale */
#if HAVE_NL_LANGINFO && defined(CODESET)
charset_hint = nl_langinfo(CODESET);
if (charset_hint != NULL && (len=strlen(charset_hint)) != 0) {
goto det_charset;
}
#endif
/* try to figure out the charset from the locale */
{
char *localename;
char *dot, *at;
/* lang[_territory][.codeset][@modifier] */
localename = setlocale(LC_CTYPE, NULL);
dot = strchr(localename, '.');
if (dot) {
dot++;
/* locale specifies a codeset */
at = strchr(dot, '@');
if (at)
len = at - dot;
else
len = strlen(dot);
charset_hint = dot;
} else {
/* no explicit name; see if the name itself
* is the charset */
charset_hint = localename;
len = strlen(charset_hint);
}
}
det_charset:
if (charset_hint) {
int found = 0;
/* now walk the charset map and look for the codeset */
for (i = 0; i < sizeof(charset_map)/sizeof(charset_map[0]); i++) {
if (len == charset_map[i].codeset_len &&
zend_binary_strcasecmp(charset_hint, len, charset_map[i].codeset, len) == 0) {
charset = charset_map[i].charset;
found = 1;
break;
}
}
if (!found) {
php_error_docref(NULL, E_WARNING, "Charset `%s' not supported, assuming utf-8",
charset_hint);
}
}
return charset;
}
/* }}} */
/* {{{ php_utf32_utf8 */
static inline size_t php_utf32_utf8(unsigned char *buf, unsigned k)
{
size_t retval = 0;
/* assert(0x0 <= k <= 0x10FFFF); */
if (k < 0x80) {
buf[0] = k;
retval = 1;
} else if (k < 0x800) {
buf[0] = 0xc0 | (k >> 6);
buf[1] = 0x80 | (k & 0x3f);
retval = 2;
} else if (k < 0x10000) {
buf[0] = 0xe0 | (k >> 12);
buf[1] = 0x80 | ((k >> 6) & 0x3f);
buf[2] = 0x80 | (k & 0x3f);
retval = 3;
} else {
buf[0] = 0xf0 | (k >> 18);
buf[1] = 0x80 | ((k >> 12) & 0x3f);
buf[2] = 0x80 | ((k >> 6) & 0x3f);
buf[3] = 0x80 | (k & 0x3f);
retval = 4;
}
/* UTF-8 has been restricted to max 4 bytes since RFC 3629 */
return retval;
}
/* }}} */
/* {{{ unimap_bsearc_cmp
* Binary search of unicode code points in unicode <--> charset mapping.
* Returns the code point in the target charset (whose mapping table was given) or 0 if
* the unicode code point is not in the table.
*/
static inline unsigned char unimap_bsearch(const uni_to_enc *table, unsigned code_key_a, size_t num)
{
const uni_to_enc *l = table,
*h = &table[num-1],
*m;
unsigned short code_key;
/* we have no mappings outside the BMP */
if (code_key_a > 0xFFFFU)
return 0;
code_key = (unsigned short) code_key_a;
while (l <= h) {
m = l + (h - l) / 2;
if (code_key < m->un_code_point)
h = m - 1;
else if (code_key > m->un_code_point)
l = m + 1;
else
return m->cs_code;
}
return 0;
}
/* }}} */
/* {{{ map_from_unicode */
static inline int map_from_unicode(unsigned code, enum entity_charset charset, unsigned *res)
{
unsigned char found;
const uni_to_enc *table;
size_t table_size;
switch (charset) {
case cs_8859_1:
/* identity mapping of code points to unicode */
if (code > 0xFF) {
return FAILURE;
}
*res = code;
break;
case cs_8859_5:
if (code <= 0xA0 || code == 0xAD /* soft hyphen */) {
*res = code;
} else if (code == 0x2116) {
*res = 0xF0; /* numero sign */
} else if (code == 0xA7) {
*res = 0xFD; /* section sign */
} else if (code >= 0x0401 && code <= 0x044F) {
if (code == 0x040D || code == 0x0450 || code == 0x045D)
return FAILURE;
*res = code - 0x360;
} else {
return FAILURE;
}
break;
case cs_8859_15:
if (code < 0xA4 || (code > 0xBE && code <= 0xFF)) {
*res = code;
} else { /* between A4 and 0xBE */
found = unimap_bsearch(unimap_iso885915,
code, sizeof(unimap_iso885915) / sizeof(*unimap_iso885915));
if (found)
*res = found;
else
return FAILURE;
}
break;
case cs_cp1252:
if (code <= 0x7F || (code >= 0xA0 && code <= 0xFF)) {
*res = code;
} else {
found = unimap_bsearch(unimap_win1252,
code, sizeof(unimap_win1252) / sizeof(*unimap_win1252));
if (found)
*res = found;
else
return FAILURE;
}
break;
case cs_macroman:
if (code == 0x7F)
return FAILURE;
table = unimap_macroman;
table_size = sizeof(unimap_macroman) / sizeof(*unimap_macroman);
goto table_over_7F;
case cs_cp1251:
table = unimap_win1251;
table_size = sizeof(unimap_win1251) / sizeof(*unimap_win1251);
goto table_over_7F;
case cs_koi8r:
table = unimap_koi8r;
table_size = sizeof(unimap_koi8r) / sizeof(*unimap_koi8r);
goto table_over_7F;
case cs_cp866:
table = unimap_cp866;
table_size = sizeof(unimap_cp866) / sizeof(*unimap_cp866);
table_over_7F:
if (code <= 0x7F) {
*res = code;
} else {
found = unimap_bsearch(table, code, table_size);
if (found)
*res = found;
else
return FAILURE;
}
break;
/* from here on, only map the possible characters in the ASCII range.
* to improve support here, it's a matter of building the unicode mappings.
* See <http://www.unicode.org/Public/6.0.0/ucd/Unihan.zip> */
case cs_sjis:
case cs_eucjp:
/* we interpret 0x5C as the Yen symbol. This is not universal.
* See <http://www.w3.org/Submission/japanese-xml/#ambiguity_of_yen> */
if (code >= 0x20 && code <= 0x7D) {
if (code == 0x5C)
return FAILURE;
*res = code;
} else {
return FAILURE;
}
break;
case cs_big5:
case cs_big5hkscs:
case cs_gb2312:
if (code >= 0x20 && code <= 0x7D) {
*res = code;
} else {
return FAILURE;
}
break;
default:
return FAILURE;
}
return SUCCESS;
}
/* }}} */
/* {{{ */
static inline void map_to_unicode(unsigned code, const enc_to_uni *table, unsigned *res)
{
/* only single byte encodings are currently supported; assumed code <= 0xFF */
*res = table->inner[ENT_ENC_TO_UNI_STAGE1(code)]->uni_cp[ENT_ENC_TO_UNI_STAGE2(code)];
}
/* }}} */
/* {{{ unicode_cp_is_allowed */
static inline int unicode_cp_is_allowed(unsigned uni_cp, int document_type)
{
/* XML 1.0 HTML 4.01 HTML 5
* 0x09..0x0A 0x09..0x0A 0x09..0x0A
* 0x0D 0x0D 0x0C..0x0D
* 0x0020..0xD7FF 0x20..0x7E 0x20..0x7E
* 0x00A0..0xD7FF 0x00A0..0xD7FF
* 0xE000..0xFFFD 0xE000..0x10FFFF 0xE000..0xFDCF
* 0x010000..0x10FFFF 0xFDF0..0x10FFFF (*)
*
* (*) exclude code points where ((code & 0xFFFF) >= 0xFFFE)
*
* References:
* XML 1.0: <http://www.w3.org/TR/REC-xml/#charsets>
* HTML 4.01: <http://www.w3.org/TR/1999/PR-html40-19990824/sgml/sgmldecl.html>
* HTML 5: <http://dev.w3.org/html5/spec/Overview.html#preprocessing-the-input-stream>
*
* Not sure this is the relevant part for HTML 5, though. I opted to
* disallow the characters that would result in a parse error when
* preprocessing of the input stream. See also section 8.1.3.
*
* It's unclear if XHTML 1.0 allows C1 characters. I'll opt to apply to
* XHTML 1.0 the same rules as for XML 1.0.
* See <http://cmsmcq.com/2007/C1.xml>.
*/
switch (document_type) {
case ENT_HTML_DOC_HTML401:
return (uni_cp >= 0x20 && uni_cp <= 0x7E) ||
(uni_cp == 0x0A || uni_cp == 0x09 || uni_cp == 0x0D) ||
(uni_cp >= 0xA0 && uni_cp <= 0xD7FF) ||
(uni_cp >= 0xE000 && uni_cp <= 0x10FFFF);
case ENT_HTML_DOC_HTML5:
return (uni_cp >= 0x20 && uni_cp <= 0x7E) ||
(uni_cp >= 0x09 && uni_cp <= 0x0D && uni_cp != 0x0B) || /* form feed U+0C allowed */
(uni_cp >= 0xA0 && uni_cp <= 0xD7FF) ||
(uni_cp >= 0xE000 && uni_cp <= 0x10FFFF &&
((uni_cp & 0xFFFF) < 0xFFFE) && /* last two of each plane (nonchars) disallowed */
(uni_cp < 0xFDD0 || uni_cp > 0xFDEF)); /* U+FDD0-U+FDEF (nonchars) disallowed */
case ENT_HTML_DOC_XHTML:
case ENT_HTML_DOC_XML1:
return (uni_cp >= 0x20 && uni_cp <= 0xD7FF) ||
(uni_cp == 0x0A || uni_cp == 0x09 || uni_cp == 0x0D) ||
(uni_cp >= 0xE000 && uni_cp <= 0x10FFFF && uni_cp != 0xFFFE && uni_cp != 0xFFFF);
default:
return 1;
}
}
/* }}} */
/* {{{ unicode_cp_is_allowed */
static inline int numeric_entity_is_allowed(unsigned uni_cp, int document_type)
{
/* less restrictive than unicode_cp_is_allowed */
switch (document_type) {
case ENT_HTML_DOC_HTML401:
/* all non-SGML characters (those marked with UNUSED in DESCSET) should be
* representable with numeric entities */
return uni_cp <= 0x10FFFF;
case ENT_HTML_DOC_HTML5:
/* 8.1.4. The numeric character reference forms described above are allowed to
* reference any Unicode code point other than U+0000, U+000D, permanently
* undefined Unicode characters (noncharacters), and control characters other
* than space characters (U+0009, U+000A, U+000C and U+000D) */
/* seems to allow surrogate characters, then */
return (uni_cp >= 0x20 && uni_cp <= 0x7E) ||
(uni_cp >= 0x09 && uni_cp <= 0x0C && uni_cp != 0x0B) || /* form feed U+0C allowed, but not U+0D */
(uni_cp >= 0xA0 && uni_cp <= 0x10FFFF &&
((uni_cp & 0xFFFF) < 0xFFFE) && /* last two of each plane (nonchars) disallowed */
(uni_cp < 0xFDD0 || uni_cp > 0xFDEF)); /* U+FDD0-U+FDEF (nonchars) disallowed */
case ENT_HTML_DOC_XHTML:
case ENT_HTML_DOC_XML1:
/* OTOH, XML 1.0 requires "character references to match the production for Char
* See <http://www.w3.org/TR/REC-xml/#NT-CharRef> */
return unicode_cp_is_allowed(uni_cp, document_type);
default:
return 1;
}
}
/* }}} */
/* {{{ process_numeric_entity
* Auxiliary function to traverse_for_entities.
* On input, *buf should point to the first character after # and on output, it's the last
* byte read, no matter if there was success or insuccess.
*/
static inline int process_numeric_entity(const char **buf, unsigned *code_point)
{
zend_long code_l;
int hexadecimal = (**buf == 'x' || **buf == 'X'); /* TODO: XML apparently disallows "X" */
char *endptr;
if (hexadecimal && (**buf != '\0'))
(*buf)++;
/* strtol allows whitespace and other stuff in the beginning
* we're not interested */
if ((hexadecimal && !isxdigit(**buf)) ||
(!hexadecimal && !isdigit(**buf))) {
return FAILURE;
}
code_l = ZEND_STRTOL(*buf, &endptr, hexadecimal ? 16 : 10);
/* we're guaranteed there were valid digits, so *endptr > buf */
*buf = endptr;
if (**buf != ';')
return FAILURE;
/* many more are invalid, but that depends on whether it's HTML
* (and which version) or XML. */
if (code_l > Z_L(0x10FFFF))
return FAILURE;
if (code_point != NULL)
*code_point = (unsigned)code_l;
return SUCCESS;
}
/* }}} */
/* {{{ process_named_entity */
static inline int process_named_entity_html(const char **buf, const char **start, size_t *length)
{
*start = *buf;
/* "&" is represented by a 0x26 in all supported encodings. That means
* the byte after represents a character or is the leading byte of an
* sequence of 8-bit code units. If in the ranges below, it represents
* necessarily a alpha character because none of the supported encodings
* has an overlap with ASCII in the leading byte (only on the second one) */
while ((**buf >= 'a' && **buf <= 'z') ||
(**buf >= 'A' && **buf <= 'Z') ||
(**buf >= '0' && **buf <= '9')) {
(*buf)++;
}
if (**buf != ';')
return FAILURE;
/* cast to size_t OK as the quantity is always non-negative */
*length = *buf - *start;
if (*length == 0)
return FAILURE;
return SUCCESS;
}
/* }}} */
/* {{{ resolve_named_entity_html */
static inline int resolve_named_entity_html(const char *start, size_t length, const entity_ht *ht, unsigned *uni_cp1, unsigned *uni_cp2)
{
const entity_cp_map *s;
zend_ulong hash = zend_inline_hash_func(start, length);
s = ht->buckets[hash % ht->num_elems];
while (s->entity) {
if (s->entity_len == length) {
if (memcmp(start, s->entity, length) == 0) {
*uni_cp1 = s->codepoint1;
*uni_cp2 = s->codepoint2;
return SUCCESS;
}
}
s++;
}
return FAILURE;
}
/* }}} */
static inline size_t write_octet_sequence(unsigned char *buf, enum entity_charset charset, unsigned code) {
/* code is not necessarily a unicode code point */
switch (charset) {
case cs_utf_8:
return php_utf32_utf8(buf, code);
case cs_8859_1:
case cs_cp1252:
case cs_8859_15:
case cs_koi8r:
case cs_cp1251:
case cs_8859_5:
case cs_cp866:
case cs_macroman:
/* single byte stuff */
*buf = code;
return 1;
case cs_big5:
case cs_big5hkscs:
case cs_sjis:
case cs_gb2312:
/* we don't have complete unicode mappings for these yet in entity_decode,
* and we opt to pass through the octet sequences for these in htmlentities
* instead of converting to an int and then converting back. */
#if 0
return php_mb2_int_to_char(buf, code);
#else
#if ZEND_DEBUG
assert(code <= 0xFFU);
#endif
*buf = code;
return 1;
#endif
case cs_eucjp:
#if 0 /* idem */
return php_mb2_int_to_char(buf, code);
#else
#if ZEND_DEBUG
assert(code <= 0xFFU);
#endif
*buf = code;
return 1;
#endif
default:
assert(0);
return 0;
}
}
/* {{{ traverse_for_entities
* Auxiliary function to php_unescape_html_entities().
* - The argument "all" determines if all numeric entities are decode or only those
* that correspond to quotes (depending on quote_style).
*/
/* maximum expansion (factor 1.2) for HTML 5 with ≫⃒ and ≪⃒ */
/* +2 is 1 because of rest (probably unnecessary), 1 because of terminating 0 */
#define TRAVERSE_FOR_ENTITIES_EXPAND_SIZE(oldlen) ((oldlen) + (oldlen) / 5 + 2)
static void traverse_for_entities(
const char *old,
size_t oldlen,
zend_string *ret, /* should have allocated TRAVERSE_FOR_ENTITIES_EXPAND_SIZE(olden) */
int all,
int flags,
const entity_ht *inv_map,
enum entity_charset charset)
{
const char *p,
*lim;
char *q;
int doctype = flags & ENT_HTML_DOC_TYPE_MASK;
lim = old + oldlen; /* terminator address */
assert(*lim == '\0');
for (p = old, q = ZSTR_VAL(ret); p < lim;) {
unsigned code, code2 = 0;
const char *next = NULL; /* when set, next > p, otherwise possible inf loop */
/* Shift JIS, Big5 and HKSCS use multi-byte encodings where an
* ASCII range byte can be part of a multi-byte sequence.
* However, they start at 0x40, therefore if we find a 0x26 byte,
* we're sure it represents the '&' character. */
/* assumes there are no single-char entities */
if (p[0] != '&' || (p + 3 >= lim)) {
*(q++) = *(p++);
continue;
}
/* now p[3] is surely valid and is no terminator */
/* numerical entity */
if (p[1] == '#') {
next = &p[2];
if (process_numeric_entity(&next, &code) == FAILURE)
goto invalid_code;
/* If we're in htmlspecialchars_decode, we're only decoding entities
* that represent &, <, >, " and '. Is this one of them? */
if (!all && (code > 63U ||
stage3_table_be_apos_00000[code].data.ent.entity == NULL))
goto invalid_code;
/* are we allowed to decode this entity in this document type?
* HTML 5 is the only that has a character that cannot be used in
* a numeric entity but is allowed literally (U+000D). The
* unoptimized version would be ... || !numeric_entity_is_allowed(code) */
if (!unicode_cp_is_allowed(code, doctype) ||
(doctype == ENT_HTML_DOC_HTML5 && code == 0x0D))
goto invalid_code;
} else {
const char *start;
size_t ent_len;
next = &p[1];
start = next;
if (process_named_entity_html(&next, &start, &ent_len) == FAILURE)
goto invalid_code;
if (resolve_named_entity_html(start, ent_len, inv_map, &code, &code2) == FAILURE) {
if (doctype == ENT_HTML_DOC_XHTML && ent_len == 4 && start[0] == 'a'
&& start[1] == 'p' && start[2] == 'o' && start[3] == 's') {
/* uses html4 inv_map, which doesn't include apos;. This is a
* hack to support it */
code = (unsigned) '\'';
} else {
goto invalid_code;
}
}
}
assert(*next == ';');
if (((code == '\'' && !(flags & ENT_HTML_QUOTE_SINGLE)) ||
(code == '"' && !(flags & ENT_HTML_QUOTE_DOUBLE)))
/* && code2 == '\0' always true for current maps */)
goto invalid_code;
/* UTF-8 doesn't need mapping (ISO-8859-1 doesn't either, but
* the call is needed to ensure the codepoint <= U+00FF) */
if (charset != cs_utf_8) {
/* replace unicode code point */
if (map_from_unicode(code, charset, &code) == FAILURE || code2 != 0)
goto invalid_code; /* not representable in target charset */
}
q += write_octet_sequence((unsigned char*)q, charset, code);
if (code2) {
q += write_octet_sequence((unsigned char*)q, charset, code2);
}
/* jump over the valid entity; may go beyond size of buffer; np */
p = next + 1;
continue;
invalid_code:
for (; p < next; p++) {
*(q++) = *p;
}
}
*q = '\0';
ZSTR_LEN(ret) = (size_t)(q - ZSTR_VAL(ret));
}
/* }}} */
/* {{{ unescape_inverse_map */
static const entity_ht *unescape_inverse_map(int all, int flags)
{