1 /*
2 * encoding.c : implements the encoding conversion functions needed for XML
3 *
4 * Related specs:
5 * rfc2044 (UTF-8 and UTF-16) F. Yergeau Alis Technologies
6 * rfc2781 UTF-16, an encoding of ISO 10646, P. Hoffman, F. Yergeau
7 * [ISO-10646] UTF-8 and UTF-16 in Annexes
8 * [ISO-8859-1] ISO Latin-1 characters codes.
9 * [UNICODE] The Unicode Consortium, "The Unicode Standard --
10 * Worldwide Character Encoding -- Version 1.0", Addison-
11 * Wesley, Volume 1, 1991, Volume 2, 1992. UTF-8 is
12 * described in Unicode Technical Report #4.
13 * [US-ASCII] Coded Character Set--7-bit American Standard Code for
14 * Information Interchange, ANSI X3.4-1986.
15 *
16 * See Copyright for the status of this software.
17 *
18 * daniel@veillard.com
19 *
20 * Original code for IsoLatin1 and UTF-16 by "Martin J. Duerst" <duerst@w3.org>
21 */
22
23 #define IN_LIBXML
24 #include "libxml.h"
25
26 #include <string.h>
27 #include <limits.h>
28 #include <ctype.h>
29 #include <stdlib.h>
30
31 #ifdef LIBXML_ICONV_ENABLED
32 #include <errno.h>
33 #endif
34
35 #include <libxml/encoding.h>
36 #include <libxml/xmlmemory.h>
37 #ifdef LIBXML_HTML_ENABLED
38 #include <libxml/HTMLparser.h>
39 #endif
40 #include <libxml/globals.h>
41 #include <libxml/xmlerror.h>
42
43 #include "buf.h"
44 #include "enc.h"
45
46 #ifdef LIBXML_ICU_ENABLED
47 #include <unicode/ucnv.h>
48 /* Size of pivot buffer, same as icu/source/common/ucnv.cpp CHUNK_SIZE */
49 #define ICU_PIVOT_BUF_SIZE 1024
50 typedef struct _uconv_t uconv_t;
51 struct _uconv_t {
52 UConverter *uconv; /* for conversion between an encoding and UTF-16 */
53 UConverter *utf8; /* for conversion between UTF-8 and UTF-16 */
54 UChar pivot_buf[ICU_PIVOT_BUF_SIZE];
55 UChar *pivot_source;
56 UChar *pivot_target;
57 };
58 #endif
59
60 static xmlCharEncodingHandlerPtr xmlUTF16LEHandler = NULL;
61 static xmlCharEncodingHandlerPtr xmlUTF16BEHandler = NULL;
62
63 typedef struct _xmlCharEncodingAlias xmlCharEncodingAlias;
64 typedef xmlCharEncodingAlias *xmlCharEncodingAliasPtr;
65 struct _xmlCharEncodingAlias {
66 const char *name;
67 const char *alias;
68 };
69
70 static xmlCharEncodingAliasPtr xmlCharEncodingAliases = NULL;
71 static int xmlCharEncodingAliasesNb = 0;
72 static int xmlCharEncodingAliasesMax = 0;
73
74 #if defined(LIBXML_ICONV_ENABLED) || defined(LIBXML_ICU_ENABLED)
75 #if 0
76 #define DEBUG_ENCODING /* Define this to get encoding traces */
77 #endif
78 #else
79 #ifdef LIBXML_ISO8859X_ENABLED
80 static void xmlRegisterCharEncodingHandlersISO8859x (void);
81 #endif
82 #endif
83
84 static int xmlLittleEndian = 1;
85
86 /**
87 * xmlEncodingErrMemory:
88 * @extra: extra information
89 *
90 * Handle an out of memory condition
91 */
92 static void
xmlEncodingErrMemory(const char * extra)93 xmlEncodingErrMemory(const char *extra)
94 {
95 __xmlSimpleError(XML_FROM_I18N, XML_ERR_NO_MEMORY, NULL, NULL, extra);
96 }
97
98 /**
99 * xmlErrEncoding:
100 * @error: the error number
101 * @msg: the error message
102 *
103 * n encoding error
104 */
105 static void LIBXML_ATTR_FORMAT(2,0)
xmlEncodingErr(xmlParserErrors error,const char * msg,const char * val)106 xmlEncodingErr(xmlParserErrors error, const char *msg, const char *val)
107 {
108 __xmlRaiseError(NULL, NULL, NULL, NULL, NULL,
109 XML_FROM_I18N, error, XML_ERR_FATAL,
110 NULL, 0, val, NULL, NULL, 0, 0, msg, val);
111 }
112
113 #ifdef LIBXML_ICU_ENABLED
114 static uconv_t*
openIcuConverter(const char * name,int toUnicode)115 openIcuConverter(const char* name, int toUnicode)
116 {
117 UErrorCode status = U_ZERO_ERROR;
118 uconv_t *conv = (uconv_t *) xmlMalloc(sizeof(uconv_t));
119 if (conv == NULL)
120 return NULL;
121
122 conv->pivot_source = conv->pivot_buf;
123 conv->pivot_target = conv->pivot_buf;
124
125 conv->uconv = ucnv_open(name, &status);
126 if (U_FAILURE(status))
127 goto error;
128
129 status = U_ZERO_ERROR;
130 if (toUnicode) {
131 ucnv_setToUCallBack(conv->uconv, UCNV_TO_U_CALLBACK_STOP,
132 NULL, NULL, NULL, &status);
133 }
134 else {
135 ucnv_setFromUCallBack(conv->uconv, UCNV_FROM_U_CALLBACK_STOP,
136 NULL, NULL, NULL, &status);
137 }
138 if (U_FAILURE(status))
139 goto error;
140
141 status = U_ZERO_ERROR;
142 conv->utf8 = ucnv_open("UTF-8", &status);
143 if (U_SUCCESS(status))
144 return conv;
145
146 error:
147 if (conv->uconv)
148 ucnv_close(conv->uconv);
149 xmlFree(conv);
150 return NULL;
151 }
152
153 static void
closeIcuConverter(uconv_t * conv)154 closeIcuConverter(uconv_t *conv)
155 {
156 if (conv != NULL) {
157 ucnv_close(conv->uconv);
158 ucnv_close(conv->utf8);
159 xmlFree(conv);
160 }
161 }
162 #endif /* LIBXML_ICU_ENABLED */
163
164 /************************************************************************
165 * *
166 * Conversions To/From UTF8 encoding *
167 * *
168 ************************************************************************/
169
170 /**
171 * asciiToUTF8:
172 * @out: a pointer to an array of bytes to store the result
173 * @outlen: the length of @out
174 * @in: a pointer to an array of ASCII chars
175 * @inlen: the length of @in
176 *
177 * Take a block of ASCII chars in and try to convert it to an UTF-8
178 * block of chars out.
179 * Returns 0 if success, or -1 otherwise
180 * The value of @inlen after return is the number of octets consumed
181 * if the return value is positive, else unpredictable.
182 * The value of @outlen after return is the number of octets produced.
183 */
184 static int
asciiToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)185 asciiToUTF8(unsigned char* out, int *outlen,
186 const unsigned char* in, int *inlen) {
187 unsigned char* outstart = out;
188 const unsigned char* base = in;
189 const unsigned char* processed = in;
190 unsigned char* outend = out + *outlen;
191 const unsigned char* inend;
192 unsigned int c;
193
194 inend = in + (*inlen);
195 while ((in < inend) && (out - outstart + 5 < *outlen)) {
196 c= *in++;
197
198 if (out >= outend)
199 break;
200 if (c < 0x80) {
201 *out++ = c;
202 } else {
203 *outlen = out - outstart;
204 *inlen = processed - base;
205 return(-1);
206 }
207
208 processed = (const unsigned char*) in;
209 }
210 *outlen = out - outstart;
211 *inlen = processed - base;
212 return(*outlen);
213 }
214
215 #ifdef LIBXML_OUTPUT_ENABLED
216 /**
217 * UTF8Toascii:
218 * @out: a pointer to an array of bytes to store the result
219 * @outlen: the length of @out
220 * @in: a pointer to an array of UTF-8 chars
221 * @inlen: the length of @in
222 *
223 * Take a block of UTF-8 chars in and try to convert it to an ASCII
224 * block of chars out.
225 *
226 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
227 * The value of @inlen after return is the number of octets consumed
228 * if the return value is positive, else unpredictable.
229 * The value of @outlen after return is the number of octets produced.
230 */
231 static int
UTF8Toascii(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)232 UTF8Toascii(unsigned char* out, int *outlen,
233 const unsigned char* in, int *inlen) {
234 const unsigned char* processed = in;
235 const unsigned char* outend;
236 const unsigned char* outstart = out;
237 const unsigned char* instart = in;
238 const unsigned char* inend;
239 unsigned int c, d;
240 int trailing;
241
242 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
243 if (in == NULL) {
244 /*
245 * initialization nothing to do
246 */
247 *outlen = 0;
248 *inlen = 0;
249 return(0);
250 }
251 inend = in + (*inlen);
252 outend = out + (*outlen);
253 while (in < inend) {
254 d = *in++;
255 if (d < 0x80) { c= d; trailing= 0; }
256 else if (d < 0xC0) {
257 /* trailing byte in leading position */
258 *outlen = out - outstart;
259 *inlen = processed - instart;
260 return(-2);
261 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
262 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
263 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
264 else {
265 /* no chance for this in Ascii */
266 *outlen = out - outstart;
267 *inlen = processed - instart;
268 return(-2);
269 }
270
271 if (inend - in < trailing) {
272 break;
273 }
274
275 for ( ; trailing; trailing--) {
276 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
277 break;
278 c <<= 6;
279 c |= d & 0x3F;
280 }
281
282 /* assertion: c is a single UTF-4 value */
283 if (c < 0x80) {
284 if (out >= outend)
285 break;
286 *out++ = c;
287 } else {
288 /* no chance for this in Ascii */
289 *outlen = out - outstart;
290 *inlen = processed - instart;
291 return(-2);
292 }
293 processed = in;
294 }
295 *outlen = out - outstart;
296 *inlen = processed - instart;
297 return(*outlen);
298 }
299 #endif /* LIBXML_OUTPUT_ENABLED */
300
301 /**
302 * isolat1ToUTF8:
303 * @out: a pointer to an array of bytes to store the result
304 * @outlen: the length of @out
305 * @in: a pointer to an array of ISO Latin 1 chars
306 * @inlen: the length of @in
307 *
308 * Take a block of ISO Latin 1 chars in and try to convert it to an UTF-8
309 * block of chars out.
310 * Returns the number of bytes written if success, or -1 otherwise
311 * The value of @inlen after return is the number of octets consumed
312 * if the return value is positive, else unpredictable.
313 * The value of @outlen after return is the number of octets produced.
314 */
315 int
isolat1ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)316 isolat1ToUTF8(unsigned char* out, int *outlen,
317 const unsigned char* in, int *inlen) {
318 unsigned char* outstart = out;
319 const unsigned char* base = in;
320 unsigned char* outend;
321 const unsigned char* inend;
322 const unsigned char* instop;
323
324 if ((out == NULL) || (in == NULL) || (outlen == NULL) || (inlen == NULL))
325 return(-1);
326
327 outend = out + *outlen;
328 inend = in + (*inlen);
329 instop = inend;
330
331 while ((in < inend) && (out < outend - 1)) {
332 if (*in >= 0x80) {
333 *out++ = (((*in) >> 6) & 0x1F) | 0xC0;
334 *out++ = ((*in) & 0x3F) | 0x80;
335 ++in;
336 }
337 if ((instop - in) > (outend - out)) instop = in + (outend - out);
338 while ((in < instop) && (*in < 0x80)) {
339 *out++ = *in++;
340 }
341 }
342 if ((in < inend) && (out < outend) && (*in < 0x80)) {
343 *out++ = *in++;
344 }
345 *outlen = out - outstart;
346 *inlen = in - base;
347 return(*outlen);
348 }
349
350 /**
351 * UTF8ToUTF8:
352 * @out: a pointer to an array of bytes to store the result
353 * @outlen: the length of @out
354 * @inb: a pointer to an array of UTF-8 chars
355 * @inlenb: the length of @in in UTF-8 chars
356 *
357 * No op copy operation for UTF8 handling.
358 *
359 * Returns the number of bytes written, or -1 if lack of space.
360 * The value of *inlen after return is the number of octets consumed
361 * if the return value is positive, else unpredictable.
362 */
363 static int
UTF8ToUTF8(unsigned char * out,int * outlen,const unsigned char * inb,int * inlenb)364 UTF8ToUTF8(unsigned char* out, int *outlen,
365 const unsigned char* inb, int *inlenb)
366 {
367 int len;
368
369 if ((out == NULL) || (outlen == NULL) || (inlenb == NULL))
370 return(-1);
371 if (inb == NULL) {
372 /* inb == NULL means output is initialized. */
373 *outlen = 0;
374 *inlenb = 0;
375 return(0);
376 }
377 if (*outlen > *inlenb) {
378 len = *inlenb;
379 } else {
380 len = *outlen;
381 }
382 if (len < 0)
383 return(-1);
384
385 /*
386 * FIXME: Conversion functions must assure valid UTF-8, so we have
387 * to check for UTF-8 validity. Preferably, this converter shouldn't
388 * be used at all.
389 */
390 memcpy(out, inb, len);
391
392 *outlen = len;
393 *inlenb = len;
394 return(*outlen);
395 }
396
397
398 #ifdef LIBXML_OUTPUT_ENABLED
399 /**
400 * UTF8Toisolat1:
401 * @out: a pointer to an array of bytes to store the result
402 * @outlen: the length of @out
403 * @in: a pointer to an array of UTF-8 chars
404 * @inlen: the length of @in
405 *
406 * Take a block of UTF-8 chars in and try to convert it to an ISO Latin 1
407 * block of chars out.
408 *
409 * Returns the number of bytes written if success, -2 if the transcoding fails,
410 or -1 otherwise
411 * The value of @inlen after return is the number of octets consumed
412 * if the return value is positive, else unpredictable.
413 * The value of @outlen after return is the number of octets produced.
414 */
415 int
UTF8Toisolat1(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)416 UTF8Toisolat1(unsigned char* out, int *outlen,
417 const unsigned char* in, int *inlen) {
418 const unsigned char* processed = in;
419 const unsigned char* outend;
420 const unsigned char* outstart = out;
421 const unsigned char* instart = in;
422 const unsigned char* inend;
423 unsigned int c, d;
424 int trailing;
425
426 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
427 if (in == NULL) {
428 /*
429 * initialization nothing to do
430 */
431 *outlen = 0;
432 *inlen = 0;
433 return(0);
434 }
435 inend = in + (*inlen);
436 outend = out + (*outlen);
437 while (in < inend) {
438 d = *in++;
439 if (d < 0x80) { c= d; trailing= 0; }
440 else if (d < 0xC0) {
441 /* trailing byte in leading position */
442 *outlen = out - outstart;
443 *inlen = processed - instart;
444 return(-2);
445 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
446 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
447 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
448 else {
449 /* no chance for this in IsoLat1 */
450 *outlen = out - outstart;
451 *inlen = processed - instart;
452 return(-2);
453 }
454
455 if (inend - in < trailing) {
456 break;
457 }
458
459 for ( ; trailing; trailing--) {
460 if (in >= inend)
461 break;
462 if (((d= *in++) & 0xC0) != 0x80) {
463 *outlen = out - outstart;
464 *inlen = processed - instart;
465 return(-2);
466 }
467 c <<= 6;
468 c |= d & 0x3F;
469 }
470
471 /* assertion: c is a single UTF-4 value */
472 if (c <= 0xFF) {
473 if (out >= outend)
474 break;
475 *out++ = c;
476 } else {
477 /* no chance for this in IsoLat1 */
478 *outlen = out - outstart;
479 *inlen = processed - instart;
480 return(-2);
481 }
482 processed = in;
483 }
484 *outlen = out - outstart;
485 *inlen = processed - instart;
486 return(*outlen);
487 }
488 #endif /* LIBXML_OUTPUT_ENABLED */
489
490 /**
491 * UTF16LEToUTF8:
492 * @out: a pointer to an array of bytes to store the result
493 * @outlen: the length of @out
494 * @inb: a pointer to an array of UTF-16LE passwd as a byte array
495 * @inlenb: the length of @in in UTF-16LE chars
496 *
497 * Take a block of UTF-16LE ushorts in and try to convert it to an UTF-8
498 * block of chars out. This function assumes the endian property
499 * is the same between the native type of this machine and the
500 * inputed one.
501 *
502 * Returns the number of bytes written, or -1 if lack of space, or -2
503 * if the transcoding fails (if *in is not a valid utf16 string)
504 * The value of *inlen after return is the number of octets consumed
505 * if the return value is positive, else unpredictable.
506 */
507 static int
UTF16LEToUTF8(unsigned char * out,int * outlen,const unsigned char * inb,int * inlenb)508 UTF16LEToUTF8(unsigned char* out, int *outlen,
509 const unsigned char* inb, int *inlenb)
510 {
511 unsigned char* outstart = out;
512 const unsigned char* processed = inb;
513 unsigned char* outend;
514 unsigned short* in = (unsigned short*) inb;
515 unsigned short* inend;
516 unsigned int c, d, inlen;
517 unsigned char *tmp;
518 int bits;
519
520 if (*outlen == 0) {
521 *inlenb = 0;
522 return(0);
523 }
524 outend = out + *outlen;
525 if ((*inlenb % 2) == 1)
526 (*inlenb)--;
527 inlen = *inlenb / 2;
528 inend = in + inlen;
529 while ((in < inend) && (out - outstart + 5 < *outlen)) {
530 if (xmlLittleEndian) {
531 c= *in++;
532 } else {
533 tmp = (unsigned char *) in;
534 c = *tmp++;
535 c = c | (((unsigned int)*tmp) << 8);
536 in++;
537 }
538 if ((c & 0xFC00) == 0xD800) { /* surrogates */
539 if (in >= inend) { /* handle split mutli-byte characters */
540 break;
541 }
542 if (xmlLittleEndian) {
543 d = *in++;
544 } else {
545 tmp = (unsigned char *) in;
546 d = *tmp++;
547 d = d | (((unsigned int)*tmp) << 8);
548 in++;
549 }
550 if ((d & 0xFC00) == 0xDC00) {
551 c &= 0x03FF;
552 c <<= 10;
553 c |= d & 0x03FF;
554 c += 0x10000;
555 }
556 else {
557 *outlen = out - outstart;
558 *inlenb = processed - inb;
559 return(-2);
560 }
561 }
562
563 /* assertion: c is a single UTF-4 value */
564 if (out >= outend)
565 break;
566 if (c < 0x80) { *out++= c; bits= -6; }
567 else if (c < 0x800) { *out++= ((c >> 6) & 0x1F) | 0xC0; bits= 0; }
568 else if (c < 0x10000) { *out++= ((c >> 12) & 0x0F) | 0xE0; bits= 6; }
569 else { *out++= ((c >> 18) & 0x07) | 0xF0; bits= 12; }
570
571 for ( ; bits >= 0; bits-= 6) {
572 if (out >= outend)
573 break;
574 *out++= ((c >> bits) & 0x3F) | 0x80;
575 }
576 processed = (const unsigned char*) in;
577 }
578 *outlen = out - outstart;
579 *inlenb = processed - inb;
580 return(*outlen);
581 }
582
583 #ifdef LIBXML_OUTPUT_ENABLED
584 /**
585 * UTF8ToUTF16LE:
586 * @outb: a pointer to an array of bytes to store the result
587 * @outlen: the length of @outb
588 * @in: a pointer to an array of UTF-8 chars
589 * @inlen: the length of @in
590 *
591 * Take a block of UTF-8 chars in and try to convert it to an UTF-16LE
592 * block of chars out.
593 *
594 * Returns the number of bytes written, or -1 if lack of space, or -2
595 * if the transcoding failed.
596 */
597 static int
UTF8ToUTF16LE(unsigned char * outb,int * outlen,const unsigned char * in,int * inlen)598 UTF8ToUTF16LE(unsigned char* outb, int *outlen,
599 const unsigned char* in, int *inlen)
600 {
601 unsigned short* out = (unsigned short*) outb;
602 const unsigned char* processed = in;
603 const unsigned char *const instart = in;
604 unsigned short* outstart= out;
605 unsigned short* outend;
606 const unsigned char* inend;
607 unsigned int c, d;
608 int trailing;
609 unsigned char *tmp;
610 unsigned short tmp1, tmp2;
611
612 /* UTF16LE encoding has no BOM */
613 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
614 if (in == NULL) {
615 *outlen = 0;
616 *inlen = 0;
617 return(0);
618 }
619 inend= in + *inlen;
620 outend = out + (*outlen / 2);
621 while (in < inend) {
622 d= *in++;
623 if (d < 0x80) { c= d; trailing= 0; }
624 else if (d < 0xC0) {
625 /* trailing byte in leading position */
626 *outlen = (out - outstart) * 2;
627 *inlen = processed - instart;
628 return(-2);
629 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
630 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
631 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
632 else {
633 /* no chance for this in UTF-16 */
634 *outlen = (out - outstart) * 2;
635 *inlen = processed - instart;
636 return(-2);
637 }
638
639 if (inend - in < trailing) {
640 break;
641 }
642
643 for ( ; trailing; trailing--) {
644 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
645 break;
646 c <<= 6;
647 c |= d & 0x3F;
648 }
649
650 /* assertion: c is a single UTF-4 value */
651 if (c < 0x10000) {
652 if (out >= outend)
653 break;
654 if (xmlLittleEndian) {
655 *out++ = c;
656 } else {
657 tmp = (unsigned char *) out;
658 *tmp = c ;
659 *(tmp + 1) = c >> 8 ;
660 out++;
661 }
662 }
663 else if (c < 0x110000) {
664 if (out+1 >= outend)
665 break;
666 c -= 0x10000;
667 if (xmlLittleEndian) {
668 *out++ = 0xD800 | (c >> 10);
669 *out++ = 0xDC00 | (c & 0x03FF);
670 } else {
671 tmp1 = 0xD800 | (c >> 10);
672 tmp = (unsigned char *) out;
673 *tmp = (unsigned char) tmp1;
674 *(tmp + 1) = tmp1 >> 8;
675 out++;
676
677 tmp2 = 0xDC00 | (c & 0x03FF);
678 tmp = (unsigned char *) out;
679 *tmp = (unsigned char) tmp2;
680 *(tmp + 1) = tmp2 >> 8;
681 out++;
682 }
683 }
684 else
685 break;
686 processed = in;
687 }
688 *outlen = (out - outstart) * 2;
689 *inlen = processed - instart;
690 return(*outlen);
691 }
692
693 /**
694 * UTF8ToUTF16:
695 * @outb: a pointer to an array of bytes to store the result
696 * @outlen: the length of @outb
697 * @in: a pointer to an array of UTF-8 chars
698 * @inlen: the length of @in
699 *
700 * Take a block of UTF-8 chars in and try to convert it to an UTF-16
701 * block of chars out.
702 *
703 * Returns the number of bytes written, or -1 if lack of space, or -2
704 * if the transcoding failed.
705 */
706 static int
UTF8ToUTF16(unsigned char * outb,int * outlen,const unsigned char * in,int * inlen)707 UTF8ToUTF16(unsigned char* outb, int *outlen,
708 const unsigned char* in, int *inlen)
709 {
710 if (in == NULL) {
711 /*
712 * initialization, add the Byte Order Mark for UTF-16LE
713 */
714 if (*outlen >= 2) {
715 outb[0] = 0xFF;
716 outb[1] = 0xFE;
717 *outlen = 2;
718 *inlen = 0;
719 #ifdef DEBUG_ENCODING
720 xmlGenericError(xmlGenericErrorContext,
721 "Added FFFE Byte Order Mark\n");
722 #endif
723 return(2);
724 }
725 *outlen = 0;
726 *inlen = 0;
727 return(0);
728 }
729 return (UTF8ToUTF16LE(outb, outlen, in, inlen));
730 }
731 #endif /* LIBXML_OUTPUT_ENABLED */
732
733 /**
734 * UTF16BEToUTF8:
735 * @out: a pointer to an array of bytes to store the result
736 * @outlen: the length of @out
737 * @inb: a pointer to an array of UTF-16 passed as a byte array
738 * @inlenb: the length of @in in UTF-16 chars
739 *
740 * Take a block of UTF-16 ushorts in and try to convert it to an UTF-8
741 * block of chars out. This function assumes the endian property
742 * is the same between the native type of this machine and the
743 * inputed one.
744 *
745 * Returns the number of bytes written, or -1 if lack of space, or -2
746 * if the transcoding fails (if *in is not a valid utf16 string)
747 * The value of *inlen after return is the number of octets consumed
748 * if the return value is positive, else unpredictable.
749 */
750 static int
UTF16BEToUTF8(unsigned char * out,int * outlen,const unsigned char * inb,int * inlenb)751 UTF16BEToUTF8(unsigned char* out, int *outlen,
752 const unsigned char* inb, int *inlenb)
753 {
754 unsigned char* outstart = out;
755 const unsigned char* processed = inb;
756 unsigned char* outend;
757 unsigned short* in = (unsigned short*) inb;
758 unsigned short* inend;
759 unsigned int c, d, inlen;
760 unsigned char *tmp;
761 int bits;
762
763 if (*outlen == 0) {
764 *inlenb = 0;
765 return(0);
766 }
767 outend = out + *outlen;
768 if ((*inlenb % 2) == 1)
769 (*inlenb)--;
770 inlen = *inlenb / 2;
771 inend= in + inlen;
772 while ((in < inend) && (out - outstart + 5 < *outlen)) {
773 if (xmlLittleEndian) {
774 tmp = (unsigned char *) in;
775 c = *tmp++;
776 c = (c << 8) | (unsigned int) *tmp;
777 in++;
778 } else {
779 c= *in++;
780 }
781 if ((c & 0xFC00) == 0xD800) { /* surrogates */
782 if (in >= inend) { /* handle split mutli-byte characters */
783 break;
784 }
785 if (xmlLittleEndian) {
786 tmp = (unsigned char *) in;
787 d = *tmp++;
788 d = (d << 8) | (unsigned int) *tmp;
789 in++;
790 } else {
791 d= *in++;
792 }
793 if ((d & 0xFC00) == 0xDC00) {
794 c &= 0x03FF;
795 c <<= 10;
796 c |= d & 0x03FF;
797 c += 0x10000;
798 }
799 else {
800 *outlen = out - outstart;
801 *inlenb = processed - inb;
802 return(-2);
803 }
804 }
805
806 /* assertion: c is a single UTF-4 value */
807 if (out >= outend)
808 break;
809 if (c < 0x80) { *out++= c; bits= -6; }
810 else if (c < 0x800) { *out++= ((c >> 6) & 0x1F) | 0xC0; bits= 0; }
811 else if (c < 0x10000) { *out++= ((c >> 12) & 0x0F) | 0xE0; bits= 6; }
812 else { *out++= ((c >> 18) & 0x07) | 0xF0; bits= 12; }
813
814 for ( ; bits >= 0; bits-= 6) {
815 if (out >= outend)
816 break;
817 *out++= ((c >> bits) & 0x3F) | 0x80;
818 }
819 processed = (const unsigned char*) in;
820 }
821 *outlen = out - outstart;
822 *inlenb = processed - inb;
823 return(*outlen);
824 }
825
826 #ifdef LIBXML_OUTPUT_ENABLED
827 /**
828 * UTF8ToUTF16BE:
829 * @outb: a pointer to an array of bytes to store the result
830 * @outlen: the length of @outb
831 * @in: a pointer to an array of UTF-8 chars
832 * @inlen: the length of @in
833 *
834 * Take a block of UTF-8 chars in and try to convert it to an UTF-16BE
835 * block of chars out.
836 *
837 * Returns the number of byte written, or -1 by lack of space, or -2
838 * if the transcoding failed.
839 */
840 static int
UTF8ToUTF16BE(unsigned char * outb,int * outlen,const unsigned char * in,int * inlen)841 UTF8ToUTF16BE(unsigned char* outb, int *outlen,
842 const unsigned char* in, int *inlen)
843 {
844 unsigned short* out = (unsigned short*) outb;
845 const unsigned char* processed = in;
846 const unsigned char *const instart = in;
847 unsigned short* outstart= out;
848 unsigned short* outend;
849 const unsigned char* inend;
850 unsigned int c, d;
851 int trailing;
852 unsigned char *tmp;
853 unsigned short tmp1, tmp2;
854
855 /* UTF-16BE has no BOM */
856 if ((outb == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
857 if (in == NULL) {
858 *outlen = 0;
859 *inlen = 0;
860 return(0);
861 }
862 inend= in + *inlen;
863 outend = out + (*outlen / 2);
864 while (in < inend) {
865 d= *in++;
866 if (d < 0x80) { c= d; trailing= 0; }
867 else if (d < 0xC0) {
868 /* trailing byte in leading position */
869 *outlen = out - outstart;
870 *inlen = processed - instart;
871 return(-2);
872 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
873 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
874 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
875 else {
876 /* no chance for this in UTF-16 */
877 *outlen = out - outstart;
878 *inlen = processed - instart;
879 return(-2);
880 }
881
882 if (inend - in < trailing) {
883 break;
884 }
885
886 for ( ; trailing; trailing--) {
887 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80)) break;
888 c <<= 6;
889 c |= d & 0x3F;
890 }
891
892 /* assertion: c is a single UTF-4 value */
893 if (c < 0x10000) {
894 if (out >= outend) break;
895 if (xmlLittleEndian) {
896 tmp = (unsigned char *) out;
897 *tmp = c >> 8;
898 *(tmp + 1) = c;
899 out++;
900 } else {
901 *out++ = c;
902 }
903 }
904 else if (c < 0x110000) {
905 if (out+1 >= outend) break;
906 c -= 0x10000;
907 if (xmlLittleEndian) {
908 tmp1 = 0xD800 | (c >> 10);
909 tmp = (unsigned char *) out;
910 *tmp = tmp1 >> 8;
911 *(tmp + 1) = (unsigned char) tmp1;
912 out++;
913
914 tmp2 = 0xDC00 | (c & 0x03FF);
915 tmp = (unsigned char *) out;
916 *tmp = tmp2 >> 8;
917 *(tmp + 1) = (unsigned char) tmp2;
918 out++;
919 } else {
920 *out++ = 0xD800 | (c >> 10);
921 *out++ = 0xDC00 | (c & 0x03FF);
922 }
923 }
924 else
925 break;
926 processed = in;
927 }
928 *outlen = (out - outstart) * 2;
929 *inlen = processed - instart;
930 return(*outlen);
931 }
932 #endif /* LIBXML_OUTPUT_ENABLED */
933
934 /************************************************************************
935 * *
936 * Generic encoding handling routines *
937 * *
938 ************************************************************************/
939
940 /**
941 * xmlDetectCharEncoding:
942 * @in: a pointer to the first bytes of the XML entity, must be at least
943 * 2 bytes long (at least 4 if encoding is UTF4 variant).
944 * @len: pointer to the length of the buffer
945 *
946 * Guess the encoding of the entity using the first bytes of the entity content
947 * according to the non-normative appendix F of the XML-1.0 recommendation.
948 *
949 * Returns one of the XML_CHAR_ENCODING_... values.
950 */
951 xmlCharEncoding
xmlDetectCharEncoding(const unsigned char * in,int len)952 xmlDetectCharEncoding(const unsigned char* in, int len)
953 {
954 if (in == NULL)
955 return(XML_CHAR_ENCODING_NONE);
956 if (len >= 4) {
957 if ((in[0] == 0x00) && (in[1] == 0x00) &&
958 (in[2] == 0x00) && (in[3] == 0x3C))
959 return(XML_CHAR_ENCODING_UCS4BE);
960 if ((in[0] == 0x3C) && (in[1] == 0x00) &&
961 (in[2] == 0x00) && (in[3] == 0x00))
962 return(XML_CHAR_ENCODING_UCS4LE);
963 if ((in[0] == 0x00) && (in[1] == 0x00) &&
964 (in[2] == 0x3C) && (in[3] == 0x00))
965 return(XML_CHAR_ENCODING_UCS4_2143);
966 if ((in[0] == 0x00) && (in[1] == 0x3C) &&
967 (in[2] == 0x00) && (in[3] == 0x00))
968 return(XML_CHAR_ENCODING_UCS4_3412);
969 if ((in[0] == 0x4C) && (in[1] == 0x6F) &&
970 (in[2] == 0xA7) && (in[3] == 0x94))
971 return(XML_CHAR_ENCODING_EBCDIC);
972 if ((in[0] == 0x3C) && (in[1] == 0x3F) &&
973 (in[2] == 0x78) && (in[3] == 0x6D))
974 return(XML_CHAR_ENCODING_UTF8);
975 /*
976 * Although not part of the recommendation, we also
977 * attempt an "auto-recognition" of UTF-16LE and
978 * UTF-16BE encodings.
979 */
980 if ((in[0] == 0x3C) && (in[1] == 0x00) &&
981 (in[2] == 0x3F) && (in[3] == 0x00))
982 return(XML_CHAR_ENCODING_UTF16LE);
983 if ((in[0] == 0x00) && (in[1] == 0x3C) &&
984 (in[2] == 0x00) && (in[3] == 0x3F))
985 return(XML_CHAR_ENCODING_UTF16BE);
986 }
987 if (len >= 3) {
988 /*
989 * Errata on XML-1.0 June 20 2001
990 * We now allow an UTF8 encoded BOM
991 */
992 if ((in[0] == 0xEF) && (in[1] == 0xBB) &&
993 (in[2] == 0xBF))
994 return(XML_CHAR_ENCODING_UTF8);
995 }
996 /* For UTF-16 we can recognize by the BOM */
997 if (len >= 2) {
998 if ((in[0] == 0xFE) && (in[1] == 0xFF))
999 return(XML_CHAR_ENCODING_UTF16BE);
1000 if ((in[0] == 0xFF) && (in[1] == 0xFE))
1001 return(XML_CHAR_ENCODING_UTF16LE);
1002 }
1003 return(XML_CHAR_ENCODING_NONE);
1004 }
1005
1006 /**
1007 * xmlCleanupEncodingAliases:
1008 *
1009 * Unregisters all aliases
1010 */
1011 void
xmlCleanupEncodingAliases(void)1012 xmlCleanupEncodingAliases(void) {
1013 int i;
1014
1015 if (xmlCharEncodingAliases == NULL)
1016 return;
1017
1018 for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
1019 if (xmlCharEncodingAliases[i].name != NULL)
1020 xmlFree((char *) xmlCharEncodingAliases[i].name);
1021 if (xmlCharEncodingAliases[i].alias != NULL)
1022 xmlFree((char *) xmlCharEncodingAliases[i].alias);
1023 }
1024 xmlCharEncodingAliasesNb = 0;
1025 xmlCharEncodingAliasesMax = 0;
1026 xmlFree(xmlCharEncodingAliases);
1027 xmlCharEncodingAliases = NULL;
1028 }
1029
1030 /**
1031 * xmlGetEncodingAlias:
1032 * @alias: the alias name as parsed, in UTF-8 format (ASCII actually)
1033 *
1034 * Lookup an encoding name for the given alias.
1035 *
1036 * Returns NULL if not found, otherwise the original name
1037 */
1038 const char *
xmlGetEncodingAlias(const char * alias)1039 xmlGetEncodingAlias(const char *alias) {
1040 int i;
1041 char upper[100];
1042
1043 if (alias == NULL)
1044 return(NULL);
1045
1046 if (xmlCharEncodingAliases == NULL)
1047 return(NULL);
1048
1049 for (i = 0;i < 99;i++) {
1050 upper[i] = toupper(alias[i]);
1051 if (upper[i] == 0) break;
1052 }
1053 upper[i] = 0;
1054
1055 /*
1056 * Walk down the list looking for a definition of the alias
1057 */
1058 for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
1059 if (!strcmp(xmlCharEncodingAliases[i].alias, upper)) {
1060 return(xmlCharEncodingAliases[i].name);
1061 }
1062 }
1063 return(NULL);
1064 }
1065
1066 /**
1067 * xmlAddEncodingAlias:
1068 * @name: the encoding name as parsed, in UTF-8 format (ASCII actually)
1069 * @alias: the alias name as parsed, in UTF-8 format (ASCII actually)
1070 *
1071 * Registers an alias @alias for an encoding named @name. Existing alias
1072 * will be overwritten.
1073 *
1074 * Returns 0 in case of success, -1 in case of error
1075 */
1076 int
xmlAddEncodingAlias(const char * name,const char * alias)1077 xmlAddEncodingAlias(const char *name, const char *alias) {
1078 int i;
1079 char upper[100];
1080
1081 if ((name == NULL) || (alias == NULL))
1082 return(-1);
1083
1084 for (i = 0;i < 99;i++) {
1085 upper[i] = toupper(alias[i]);
1086 if (upper[i] == 0) break;
1087 }
1088 upper[i] = 0;
1089
1090 if (xmlCharEncodingAliases == NULL) {
1091 xmlCharEncodingAliasesNb = 0;
1092 xmlCharEncodingAliasesMax = 20;
1093 xmlCharEncodingAliases = (xmlCharEncodingAliasPtr)
1094 xmlMalloc(xmlCharEncodingAliasesMax * sizeof(xmlCharEncodingAlias));
1095 if (xmlCharEncodingAliases == NULL)
1096 return(-1);
1097 } else if (xmlCharEncodingAliasesNb >= xmlCharEncodingAliasesMax) {
1098 xmlCharEncodingAliasesMax *= 2;
1099 xmlCharEncodingAliases = (xmlCharEncodingAliasPtr)
1100 xmlRealloc(xmlCharEncodingAliases,
1101 xmlCharEncodingAliasesMax * sizeof(xmlCharEncodingAlias));
1102 }
1103 /*
1104 * Walk down the list looking for a definition of the alias
1105 */
1106 for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
1107 if (!strcmp(xmlCharEncodingAliases[i].alias, upper)) {
1108 /*
1109 * Replace the definition.
1110 */
1111 xmlFree((char *) xmlCharEncodingAliases[i].name);
1112 xmlCharEncodingAliases[i].name = xmlMemStrdup(name);
1113 return(0);
1114 }
1115 }
1116 /*
1117 * Add the definition
1118 */
1119 xmlCharEncodingAliases[xmlCharEncodingAliasesNb].name = xmlMemStrdup(name);
1120 xmlCharEncodingAliases[xmlCharEncodingAliasesNb].alias = xmlMemStrdup(upper);
1121 xmlCharEncodingAliasesNb++;
1122 return(0);
1123 }
1124
1125 /**
1126 * xmlDelEncodingAlias:
1127 * @alias: the alias name as parsed, in UTF-8 format (ASCII actually)
1128 *
1129 * Unregisters an encoding alias @alias
1130 *
1131 * Returns 0 in case of success, -1 in case of error
1132 */
1133 int
xmlDelEncodingAlias(const char * alias)1134 xmlDelEncodingAlias(const char *alias) {
1135 int i;
1136
1137 if (alias == NULL)
1138 return(-1);
1139
1140 if (xmlCharEncodingAliases == NULL)
1141 return(-1);
1142 /*
1143 * Walk down the list looking for a definition of the alias
1144 */
1145 for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
1146 if (!strcmp(xmlCharEncodingAliases[i].alias, alias)) {
1147 xmlFree((char *) xmlCharEncodingAliases[i].name);
1148 xmlFree((char *) xmlCharEncodingAliases[i].alias);
1149 xmlCharEncodingAliasesNb--;
1150 memmove(&xmlCharEncodingAliases[i], &xmlCharEncodingAliases[i + 1],
1151 sizeof(xmlCharEncodingAlias) * (xmlCharEncodingAliasesNb - i));
1152 return(0);
1153 }
1154 }
1155 return(-1);
1156 }
1157
1158 /**
1159 * xmlParseCharEncoding:
1160 * @name: the encoding name as parsed, in UTF-8 format (ASCII actually)
1161 *
1162 * Compare the string to the encoding schemes already known. Note
1163 * that the comparison is case insensitive accordingly to the section
1164 * [XML] 4.3.3 Character Encoding in Entities.
1165 *
1166 * Returns one of the XML_CHAR_ENCODING_... values or XML_CHAR_ENCODING_NONE
1167 * if not recognized.
1168 */
1169 xmlCharEncoding
xmlParseCharEncoding(const char * name)1170 xmlParseCharEncoding(const char* name)
1171 {
1172 const char *alias;
1173 char upper[500];
1174 int i;
1175
1176 if (name == NULL)
1177 return(XML_CHAR_ENCODING_NONE);
1178
1179 /*
1180 * Do the alias resolution
1181 */
1182 alias = xmlGetEncodingAlias(name);
1183 if (alias != NULL)
1184 name = alias;
1185
1186 for (i = 0;i < 499;i++) {
1187 upper[i] = toupper(name[i]);
1188 if (upper[i] == 0) break;
1189 }
1190 upper[i] = 0;
1191
1192 if (!strcmp(upper, "")) return(XML_CHAR_ENCODING_NONE);
1193 if (!strcmp(upper, "UTF-8")) return(XML_CHAR_ENCODING_UTF8);
1194 if (!strcmp(upper, "UTF8")) return(XML_CHAR_ENCODING_UTF8);
1195
1196 /*
1197 * NOTE: if we were able to parse this, the endianness of UTF16 is
1198 * already found and in use
1199 */
1200 if (!strcmp(upper, "UTF-16")) return(XML_CHAR_ENCODING_UTF16LE);
1201 if (!strcmp(upper, "UTF16")) return(XML_CHAR_ENCODING_UTF16LE);
1202
1203 if (!strcmp(upper, "ISO-10646-UCS-2")) return(XML_CHAR_ENCODING_UCS2);
1204 if (!strcmp(upper, "UCS-2")) return(XML_CHAR_ENCODING_UCS2);
1205 if (!strcmp(upper, "UCS2")) return(XML_CHAR_ENCODING_UCS2);
1206
1207 /*
1208 * NOTE: if we were able to parse this, the endianness of UCS4 is
1209 * already found and in use
1210 */
1211 if (!strcmp(upper, "ISO-10646-UCS-4")) return(XML_CHAR_ENCODING_UCS4LE);
1212 if (!strcmp(upper, "UCS-4")) return(XML_CHAR_ENCODING_UCS4LE);
1213 if (!strcmp(upper, "UCS4")) return(XML_CHAR_ENCODING_UCS4LE);
1214
1215
1216 if (!strcmp(upper, "ISO-8859-1")) return(XML_CHAR_ENCODING_8859_1);
1217 if (!strcmp(upper, "ISO-LATIN-1")) return(XML_CHAR_ENCODING_8859_1);
1218 if (!strcmp(upper, "ISO LATIN 1")) return(XML_CHAR_ENCODING_8859_1);
1219
1220 if (!strcmp(upper, "ISO-8859-2")) return(XML_CHAR_ENCODING_8859_2);
1221 if (!strcmp(upper, "ISO-LATIN-2")) return(XML_CHAR_ENCODING_8859_2);
1222 if (!strcmp(upper, "ISO LATIN 2")) return(XML_CHAR_ENCODING_8859_2);
1223
1224 if (!strcmp(upper, "ISO-8859-3")) return(XML_CHAR_ENCODING_8859_3);
1225 if (!strcmp(upper, "ISO-8859-4")) return(XML_CHAR_ENCODING_8859_4);
1226 if (!strcmp(upper, "ISO-8859-5")) return(XML_CHAR_ENCODING_8859_5);
1227 if (!strcmp(upper, "ISO-8859-6")) return(XML_CHAR_ENCODING_8859_6);
1228 if (!strcmp(upper, "ISO-8859-7")) return(XML_CHAR_ENCODING_8859_7);
1229 if (!strcmp(upper, "ISO-8859-8")) return(XML_CHAR_ENCODING_8859_8);
1230 if (!strcmp(upper, "ISO-8859-9")) return(XML_CHAR_ENCODING_8859_9);
1231
1232 if (!strcmp(upper, "ISO-2022-JP")) return(XML_CHAR_ENCODING_2022_JP);
1233 if (!strcmp(upper, "SHIFT_JIS")) return(XML_CHAR_ENCODING_SHIFT_JIS);
1234 if (!strcmp(upper, "EUC-JP")) return(XML_CHAR_ENCODING_EUC_JP);
1235
1236 #ifdef DEBUG_ENCODING
1237 xmlGenericError(xmlGenericErrorContext, "Unknown encoding %s\n", name);
1238 #endif
1239 return(XML_CHAR_ENCODING_ERROR);
1240 }
1241
1242 /**
1243 * xmlGetCharEncodingName:
1244 * @enc: the encoding
1245 *
1246 * The "canonical" name for XML encoding.
1247 * C.f. http://www.w3.org/TR/REC-xml#charencoding
1248 * Section 4.3.3 Character Encoding in Entities
1249 *
1250 * Returns the canonical name for the given encoding
1251 */
1252
1253 const char*
xmlGetCharEncodingName(xmlCharEncoding enc)1254 xmlGetCharEncodingName(xmlCharEncoding enc) {
1255 switch (enc) {
1256 case XML_CHAR_ENCODING_ERROR:
1257 return(NULL);
1258 case XML_CHAR_ENCODING_NONE:
1259 return(NULL);
1260 case XML_CHAR_ENCODING_UTF8:
1261 return("UTF-8");
1262 case XML_CHAR_ENCODING_UTF16LE:
1263 return("UTF-16");
1264 case XML_CHAR_ENCODING_UTF16BE:
1265 return("UTF-16");
1266 case XML_CHAR_ENCODING_EBCDIC:
1267 return("EBCDIC");
1268 case XML_CHAR_ENCODING_UCS4LE:
1269 return("ISO-10646-UCS-4");
1270 case XML_CHAR_ENCODING_UCS4BE:
1271 return("ISO-10646-UCS-4");
1272 case XML_CHAR_ENCODING_UCS4_2143:
1273 return("ISO-10646-UCS-4");
1274 case XML_CHAR_ENCODING_UCS4_3412:
1275 return("ISO-10646-UCS-4");
1276 case XML_CHAR_ENCODING_UCS2:
1277 return("ISO-10646-UCS-2");
1278 case XML_CHAR_ENCODING_8859_1:
1279 return("ISO-8859-1");
1280 case XML_CHAR_ENCODING_8859_2:
1281 return("ISO-8859-2");
1282 case XML_CHAR_ENCODING_8859_3:
1283 return("ISO-8859-3");
1284 case XML_CHAR_ENCODING_8859_4:
1285 return("ISO-8859-4");
1286 case XML_CHAR_ENCODING_8859_5:
1287 return("ISO-8859-5");
1288 case XML_CHAR_ENCODING_8859_6:
1289 return("ISO-8859-6");
1290 case XML_CHAR_ENCODING_8859_7:
1291 return("ISO-8859-7");
1292 case XML_CHAR_ENCODING_8859_8:
1293 return("ISO-8859-8");
1294 case XML_CHAR_ENCODING_8859_9:
1295 return("ISO-8859-9");
1296 case XML_CHAR_ENCODING_2022_JP:
1297 return("ISO-2022-JP");
1298 case XML_CHAR_ENCODING_SHIFT_JIS:
1299 return("Shift-JIS");
1300 case XML_CHAR_ENCODING_EUC_JP:
1301 return("EUC-JP");
1302 case XML_CHAR_ENCODING_ASCII:
1303 return(NULL);
1304 }
1305 return(NULL);
1306 }
1307
1308 /************************************************************************
1309 * *
1310 * Char encoding handlers *
1311 * *
1312 ************************************************************************/
1313
1314
1315 /* the size should be growable, but it's not a big deal ... */
1316 #define MAX_ENCODING_HANDLERS 50
1317 static xmlCharEncodingHandlerPtr *handlers = NULL;
1318 static int nbCharEncodingHandler = 0;
1319
1320 /*
1321 * The default is UTF-8 for XML, that's also the default used for the
1322 * parser internals, so the default encoding handler is NULL
1323 */
1324
1325 static xmlCharEncodingHandlerPtr xmlDefaultCharEncodingHandler = NULL;
1326
1327 /**
1328 * xmlNewCharEncodingHandler:
1329 * @name: the encoding name, in UTF-8 format (ASCII actually)
1330 * @input: the xmlCharEncodingInputFunc to read that encoding
1331 * @output: the xmlCharEncodingOutputFunc to write that encoding
1332 *
1333 * Create and registers an xmlCharEncodingHandler.
1334 *
1335 * Returns the xmlCharEncodingHandlerPtr created (or NULL in case of error).
1336 */
1337 xmlCharEncodingHandlerPtr
xmlNewCharEncodingHandler(const char * name,xmlCharEncodingInputFunc input,xmlCharEncodingOutputFunc output)1338 xmlNewCharEncodingHandler(const char *name,
1339 xmlCharEncodingInputFunc input,
1340 xmlCharEncodingOutputFunc output) {
1341 xmlCharEncodingHandlerPtr handler;
1342 const char *alias;
1343 char upper[500];
1344 int i;
1345 char *up = NULL;
1346
1347 /*
1348 * Do the alias resolution
1349 */
1350 alias = xmlGetEncodingAlias(name);
1351 if (alias != NULL)
1352 name = alias;
1353
1354 /*
1355 * Keep only the uppercase version of the encoding.
1356 */
1357 if (name == NULL) {
1358 xmlEncodingErr(XML_I18N_NO_NAME,
1359 "xmlNewCharEncodingHandler : no name !\n", NULL);
1360 return(NULL);
1361 }
1362 for (i = 0;i < 499;i++) {
1363 upper[i] = toupper(name[i]);
1364 if (upper[i] == 0) break;
1365 }
1366 upper[i] = 0;
1367 up = xmlMemStrdup(upper);
1368 if (up == NULL) {
1369 xmlEncodingErrMemory("xmlNewCharEncodingHandler : out of memory !\n");
1370 return(NULL);
1371 }
1372
1373 /*
1374 * allocate and fill-up an handler block.
1375 */
1376 handler = (xmlCharEncodingHandlerPtr)
1377 xmlMalloc(sizeof(xmlCharEncodingHandler));
1378 if (handler == NULL) {
1379 xmlFree(up);
1380 xmlEncodingErrMemory("xmlNewCharEncodingHandler : out of memory !\n");
1381 return(NULL);
1382 }
1383 memset(handler, 0, sizeof(xmlCharEncodingHandler));
1384 handler->input = input;
1385 handler->output = output;
1386 handler->name = up;
1387
1388 #ifdef LIBXML_ICONV_ENABLED
1389 handler->iconv_in = NULL;
1390 handler->iconv_out = NULL;
1391 #endif
1392 #ifdef LIBXML_ICU_ENABLED
1393 handler->uconv_in = NULL;
1394 handler->uconv_out = NULL;
1395 #endif
1396
1397 /*
1398 * registers and returns the handler.
1399 */
1400 xmlRegisterCharEncodingHandler(handler);
1401 #ifdef DEBUG_ENCODING
1402 xmlGenericError(xmlGenericErrorContext,
1403 "Registered encoding handler for %s\n", name);
1404 #endif
1405 return(handler);
1406 }
1407
1408 /**
1409 * xmlInitCharEncodingHandlers:
1410 *
1411 * DEPRECATED: This function will be made private. Call xmlInitParser to
1412 * initialize the library.
1413 *
1414 * Initialize the char encoding support, it registers the default
1415 * encoding supported.
1416 * NOTE: while public, this function usually doesn't need to be called
1417 * in normal processing.
1418 */
1419 void
xmlInitCharEncodingHandlers(void)1420 xmlInitCharEncodingHandlers(void) {
1421 unsigned short int tst = 0x1234;
1422 unsigned char *ptr = (unsigned char *) &tst;
1423
1424 if (handlers != NULL) return;
1425
1426 handlers = (xmlCharEncodingHandlerPtr *)
1427 xmlMalloc(MAX_ENCODING_HANDLERS * sizeof(xmlCharEncodingHandlerPtr));
1428
1429 if (*ptr == 0x12) xmlLittleEndian = 0;
1430 else if (*ptr == 0x34) xmlLittleEndian = 1;
1431 else {
1432 xmlEncodingErr(XML_ERR_INTERNAL_ERROR,
1433 "Odd problem at endianness detection\n", NULL);
1434 }
1435
1436 if (handlers == NULL) {
1437 xmlEncodingErrMemory("xmlInitCharEncodingHandlers : out of memory !\n");
1438 return;
1439 }
1440 xmlNewCharEncodingHandler("UTF-8", UTF8ToUTF8, UTF8ToUTF8);
1441 #ifdef LIBXML_OUTPUT_ENABLED
1442 xmlUTF16LEHandler =
1443 xmlNewCharEncodingHandler("UTF-16LE", UTF16LEToUTF8, UTF8ToUTF16LE);
1444 xmlUTF16BEHandler =
1445 xmlNewCharEncodingHandler("UTF-16BE", UTF16BEToUTF8, UTF8ToUTF16BE);
1446 xmlNewCharEncodingHandler("UTF-16", UTF16LEToUTF8, UTF8ToUTF16);
1447 xmlNewCharEncodingHandler("ISO-8859-1", isolat1ToUTF8, UTF8Toisolat1);
1448 xmlNewCharEncodingHandler("ASCII", asciiToUTF8, UTF8Toascii);
1449 xmlNewCharEncodingHandler("US-ASCII", asciiToUTF8, UTF8Toascii);
1450 #ifdef LIBXML_HTML_ENABLED
1451 xmlNewCharEncodingHandler("HTML", NULL, UTF8ToHtml);
1452 #endif
1453 #else
1454 xmlUTF16LEHandler =
1455 xmlNewCharEncodingHandler("UTF-16LE", UTF16LEToUTF8, NULL);
1456 xmlUTF16BEHandler =
1457 xmlNewCharEncodingHandler("UTF-16BE", UTF16BEToUTF8, NULL);
1458 xmlNewCharEncodingHandler("UTF-16", UTF16LEToUTF8, NULL);
1459 xmlNewCharEncodingHandler("ISO-8859-1", isolat1ToUTF8, NULL);
1460 xmlNewCharEncodingHandler("ASCII", asciiToUTF8, NULL);
1461 xmlNewCharEncodingHandler("US-ASCII", asciiToUTF8, NULL);
1462 #endif /* LIBXML_OUTPUT_ENABLED */
1463 #if !defined(LIBXML_ICONV_ENABLED) && !defined(LIBXML_ICU_ENABLED)
1464 #ifdef LIBXML_ISO8859X_ENABLED
1465 xmlRegisterCharEncodingHandlersISO8859x ();
1466 #endif
1467 #endif
1468
1469 }
1470
1471 /**
1472 * xmlCleanupCharEncodingHandlers:
1473 *
1474 * DEPRECATED: This function will be made private. Call xmlCleanupParser
1475 * to free global state but see the warnings there. xmlCleanupParser
1476 * should be only called once at program exit. In most cases, you don't
1477 * have call cleanup functions at all.
1478 *
1479 * Cleanup the memory allocated for the char encoding support, it
1480 * unregisters all the encoding handlers and the aliases.
1481 */
1482 void
xmlCleanupCharEncodingHandlers(void)1483 xmlCleanupCharEncodingHandlers(void) {
1484 xmlCleanupEncodingAliases();
1485
1486 if (handlers == NULL) return;
1487
1488 for (;nbCharEncodingHandler > 0;) {
1489 nbCharEncodingHandler--;
1490 if (handlers[nbCharEncodingHandler] != NULL) {
1491 if (handlers[nbCharEncodingHandler]->name != NULL)
1492 xmlFree(handlers[nbCharEncodingHandler]->name);
1493 xmlFree(handlers[nbCharEncodingHandler]);
1494 }
1495 }
1496 xmlFree(handlers);
1497 handlers = NULL;
1498 nbCharEncodingHandler = 0;
1499 xmlDefaultCharEncodingHandler = NULL;
1500 }
1501
1502 /**
1503 * xmlRegisterCharEncodingHandler:
1504 * @handler: the xmlCharEncodingHandlerPtr handler block
1505 *
1506 * Register the char encoding handler, surprising, isn't it ?
1507 */
1508 void
xmlRegisterCharEncodingHandler(xmlCharEncodingHandlerPtr handler)1509 xmlRegisterCharEncodingHandler(xmlCharEncodingHandlerPtr handler) {
1510 if (handlers == NULL) xmlInitCharEncodingHandlers();
1511 if ((handler == NULL) || (handlers == NULL)) {
1512 xmlEncodingErr(XML_I18N_NO_HANDLER,
1513 "xmlRegisterCharEncodingHandler: NULL handler !\n", NULL);
1514 goto free_handler;
1515 }
1516
1517 if (nbCharEncodingHandler >= MAX_ENCODING_HANDLERS) {
1518 xmlEncodingErr(XML_I18N_EXCESS_HANDLER,
1519 "xmlRegisterCharEncodingHandler: Too many handler registered, see %s\n",
1520 "MAX_ENCODING_HANDLERS");
1521 goto free_handler;
1522 }
1523 handlers[nbCharEncodingHandler++] = handler;
1524 return;
1525
1526 free_handler:
1527 if (handler != NULL) {
1528 if (handler->name != NULL) {
1529 xmlFree(handler->name);
1530 }
1531 xmlFree(handler);
1532 }
1533 }
1534
1535 /**
1536 * xmlGetCharEncodingHandler:
1537 * @enc: an xmlCharEncoding value.
1538 *
1539 * Search in the registered set the handler able to read/write that encoding.
1540 *
1541 * Returns the handler or NULL if not found
1542 */
1543 xmlCharEncodingHandlerPtr
xmlGetCharEncodingHandler(xmlCharEncoding enc)1544 xmlGetCharEncodingHandler(xmlCharEncoding enc) {
1545 xmlCharEncodingHandlerPtr handler;
1546
1547 if (handlers == NULL) xmlInitCharEncodingHandlers();
1548 switch (enc) {
1549 case XML_CHAR_ENCODING_ERROR:
1550 return(NULL);
1551 case XML_CHAR_ENCODING_NONE:
1552 return(NULL);
1553 case XML_CHAR_ENCODING_UTF8:
1554 return(NULL);
1555 case XML_CHAR_ENCODING_UTF16LE:
1556 return(xmlUTF16LEHandler);
1557 case XML_CHAR_ENCODING_UTF16BE:
1558 return(xmlUTF16BEHandler);
1559 case XML_CHAR_ENCODING_EBCDIC:
1560 handler = xmlFindCharEncodingHandler("EBCDIC");
1561 if (handler != NULL) return(handler);
1562 handler = xmlFindCharEncodingHandler("ebcdic");
1563 if (handler != NULL) return(handler);
1564 handler = xmlFindCharEncodingHandler("EBCDIC-US");
1565 if (handler != NULL) return(handler);
1566 handler = xmlFindCharEncodingHandler("IBM-037");
1567 if (handler != NULL) return(handler);
1568 break;
1569 case XML_CHAR_ENCODING_UCS4BE:
1570 handler = xmlFindCharEncodingHandler("ISO-10646-UCS-4");
1571 if (handler != NULL) return(handler);
1572 handler = xmlFindCharEncodingHandler("UCS-4");
1573 if (handler != NULL) return(handler);
1574 handler = xmlFindCharEncodingHandler("UCS4");
1575 if (handler != NULL) return(handler);
1576 break;
1577 case XML_CHAR_ENCODING_UCS4LE:
1578 handler = xmlFindCharEncodingHandler("ISO-10646-UCS-4");
1579 if (handler != NULL) return(handler);
1580 handler = xmlFindCharEncodingHandler("UCS-4");
1581 if (handler != NULL) return(handler);
1582 handler = xmlFindCharEncodingHandler("UCS4");
1583 if (handler != NULL) return(handler);
1584 break;
1585 case XML_CHAR_ENCODING_UCS4_2143:
1586 break;
1587 case XML_CHAR_ENCODING_UCS4_3412:
1588 break;
1589 case XML_CHAR_ENCODING_UCS2:
1590 handler = xmlFindCharEncodingHandler("ISO-10646-UCS-2");
1591 if (handler != NULL) return(handler);
1592 handler = xmlFindCharEncodingHandler("UCS-2");
1593 if (handler != NULL) return(handler);
1594 handler = xmlFindCharEncodingHandler("UCS2");
1595 if (handler != NULL) return(handler);
1596 break;
1597
1598 /*
1599 * We used to keep ISO Latin encodings native in the
1600 * generated data. This led to so many problems that
1601 * this has been removed. One can still change this
1602 * back by registering no-ops encoders for those
1603 */
1604 case XML_CHAR_ENCODING_8859_1:
1605 handler = xmlFindCharEncodingHandler("ISO-8859-1");
1606 if (handler != NULL) return(handler);
1607 break;
1608 case XML_CHAR_ENCODING_8859_2:
1609 handler = xmlFindCharEncodingHandler("ISO-8859-2");
1610 if (handler != NULL) return(handler);
1611 break;
1612 case XML_CHAR_ENCODING_8859_3:
1613 handler = xmlFindCharEncodingHandler("ISO-8859-3");
1614 if (handler != NULL) return(handler);
1615 break;
1616 case XML_CHAR_ENCODING_8859_4:
1617 handler = xmlFindCharEncodingHandler("ISO-8859-4");
1618 if (handler != NULL) return(handler);
1619 break;
1620 case XML_CHAR_ENCODING_8859_5:
1621 handler = xmlFindCharEncodingHandler("ISO-8859-5");
1622 if (handler != NULL) return(handler);
1623 break;
1624 case XML_CHAR_ENCODING_8859_6:
1625 handler = xmlFindCharEncodingHandler("ISO-8859-6");
1626 if (handler != NULL) return(handler);
1627 break;
1628 case XML_CHAR_ENCODING_8859_7:
1629 handler = xmlFindCharEncodingHandler("ISO-8859-7");
1630 if (handler != NULL) return(handler);
1631 break;
1632 case XML_CHAR_ENCODING_8859_8:
1633 handler = xmlFindCharEncodingHandler("ISO-8859-8");
1634 if (handler != NULL) return(handler);
1635 break;
1636 case XML_CHAR_ENCODING_8859_9:
1637 handler = xmlFindCharEncodingHandler("ISO-8859-9");
1638 if (handler != NULL) return(handler);
1639 break;
1640
1641
1642 case XML_CHAR_ENCODING_2022_JP:
1643 handler = xmlFindCharEncodingHandler("ISO-2022-JP");
1644 if (handler != NULL) return(handler);
1645 break;
1646 case XML_CHAR_ENCODING_SHIFT_JIS:
1647 handler = xmlFindCharEncodingHandler("SHIFT-JIS");
1648 if (handler != NULL) return(handler);
1649 handler = xmlFindCharEncodingHandler("SHIFT_JIS");
1650 if (handler != NULL) return(handler);
1651 handler = xmlFindCharEncodingHandler("Shift_JIS");
1652 if (handler != NULL) return(handler);
1653 break;
1654 case XML_CHAR_ENCODING_EUC_JP:
1655 handler = xmlFindCharEncodingHandler("EUC-JP");
1656 if (handler != NULL) return(handler);
1657 break;
1658 default:
1659 break;
1660 }
1661
1662 #ifdef DEBUG_ENCODING
1663 xmlGenericError(xmlGenericErrorContext,
1664 "No handler found for encoding %d\n", enc);
1665 #endif
1666 return(NULL);
1667 }
1668
1669 /**
1670 * xmlFindCharEncodingHandler:
1671 * @name: a string describing the char encoding.
1672 *
1673 * Search in the registered set the handler able to read/write that encoding.
1674 *
1675 * Returns the handler or NULL if not found
1676 */
1677 xmlCharEncodingHandlerPtr
xmlFindCharEncodingHandler(const char * name)1678 xmlFindCharEncodingHandler(const char *name) {
1679 const char *nalias;
1680 const char *norig;
1681 xmlCharEncoding alias;
1682 #ifdef LIBXML_ICONV_ENABLED
1683 xmlCharEncodingHandlerPtr enc;
1684 iconv_t icv_in, icv_out;
1685 #endif /* LIBXML_ICONV_ENABLED */
1686 #ifdef LIBXML_ICU_ENABLED
1687 xmlCharEncodingHandlerPtr encu;
1688 uconv_t *ucv_in, *ucv_out;
1689 #endif /* LIBXML_ICU_ENABLED */
1690 char upper[100];
1691 int i;
1692
1693 if (handlers == NULL) xmlInitCharEncodingHandlers();
1694 if (name == NULL) return(xmlDefaultCharEncodingHandler);
1695 if (name[0] == 0) return(xmlDefaultCharEncodingHandler);
1696
1697 /*
1698 * Do the alias resolution
1699 */
1700 norig = name;
1701 nalias = xmlGetEncodingAlias(name);
1702 if (nalias != NULL)
1703 name = nalias;
1704
1705 /*
1706 * Check first for directly registered encoding names
1707 */
1708 for (i = 0;i < 99;i++) {
1709 upper[i] = toupper(name[i]);
1710 if (upper[i] == 0) break;
1711 }
1712 upper[i] = 0;
1713
1714 if (handlers != NULL) {
1715 for (i = 0;i < nbCharEncodingHandler; i++) {
1716 if (!strcmp(upper, handlers[i]->name)) {
1717 #ifdef DEBUG_ENCODING
1718 xmlGenericError(xmlGenericErrorContext,
1719 "Found registered handler for encoding %s\n", name);
1720 #endif
1721 return(handlers[i]);
1722 }
1723 }
1724 }
1725
1726 #ifdef LIBXML_ICONV_ENABLED
1727 /* check whether iconv can handle this */
1728 icv_in = iconv_open("UTF-8", name);
1729 icv_out = iconv_open(name, "UTF-8");
1730 if (icv_in == (iconv_t) -1) {
1731 icv_in = iconv_open("UTF-8", upper);
1732 }
1733 if (icv_out == (iconv_t) -1) {
1734 icv_out = iconv_open(upper, "UTF-8");
1735 }
1736 if ((icv_in != (iconv_t) -1) && (icv_out != (iconv_t) -1)) {
1737 enc = (xmlCharEncodingHandlerPtr)
1738 xmlMalloc(sizeof(xmlCharEncodingHandler));
1739 if (enc == NULL) {
1740 iconv_close(icv_in);
1741 iconv_close(icv_out);
1742 return(NULL);
1743 }
1744 memset(enc, 0, sizeof(xmlCharEncodingHandler));
1745 enc->name = xmlMemStrdup(name);
1746 enc->input = NULL;
1747 enc->output = NULL;
1748 enc->iconv_in = icv_in;
1749 enc->iconv_out = icv_out;
1750 #ifdef DEBUG_ENCODING
1751 xmlGenericError(xmlGenericErrorContext,
1752 "Found iconv handler for encoding %s\n", name);
1753 #endif
1754 return enc;
1755 } else if ((icv_in != (iconv_t) -1) || icv_out != (iconv_t) -1) {
1756 xmlEncodingErr(XML_ERR_INTERNAL_ERROR,
1757 "iconv : problems with filters for '%s'\n", name);
1758 if (icv_in != (iconv_t) -1)
1759 iconv_close(icv_in);
1760 else
1761 iconv_close(icv_out);
1762 }
1763 #endif /* LIBXML_ICONV_ENABLED */
1764 #ifdef LIBXML_ICU_ENABLED
1765 /* check whether icu can handle this */
1766 ucv_in = openIcuConverter(name, 1);
1767 ucv_out = openIcuConverter(name, 0);
1768 if (ucv_in != NULL && ucv_out != NULL) {
1769 encu = (xmlCharEncodingHandlerPtr)
1770 xmlMalloc(sizeof(xmlCharEncodingHandler));
1771 if (encu == NULL) {
1772 closeIcuConverter(ucv_in);
1773 closeIcuConverter(ucv_out);
1774 return(NULL);
1775 }
1776 memset(encu, 0, sizeof(xmlCharEncodingHandler));
1777 encu->name = xmlMemStrdup(name);
1778 encu->input = NULL;
1779 encu->output = NULL;
1780 encu->uconv_in = ucv_in;
1781 encu->uconv_out = ucv_out;
1782 #ifdef DEBUG_ENCODING
1783 xmlGenericError(xmlGenericErrorContext,
1784 "Found ICU converter handler for encoding %s\n", name);
1785 #endif
1786 return encu;
1787 } else if (ucv_in != NULL || ucv_out != NULL) {
1788 closeIcuConverter(ucv_in);
1789 closeIcuConverter(ucv_out);
1790 xmlEncodingErr(XML_ERR_INTERNAL_ERROR,
1791 "ICU converter : problems with filters for '%s'\n", name);
1792 }
1793 #endif /* LIBXML_ICU_ENABLED */
1794
1795 #ifdef DEBUG_ENCODING
1796 xmlGenericError(xmlGenericErrorContext,
1797 "No handler found for encoding %s\n", name);
1798 #endif
1799
1800 /*
1801 * Fallback using the canonical names
1802 */
1803 alias = xmlParseCharEncoding(norig);
1804 if (alias != XML_CHAR_ENCODING_ERROR) {
1805 const char* canon;
1806 canon = xmlGetCharEncodingName(alias);
1807 if ((canon != NULL) && (strcmp(name, canon))) {
1808 return(xmlFindCharEncodingHandler(canon));
1809 }
1810 }
1811
1812 /* If "none of the above", give up */
1813 return(NULL);
1814 }
1815
1816 /************************************************************************
1817 * *
1818 * ICONV based generic conversion functions *
1819 * *
1820 ************************************************************************/
1821
1822 #ifdef LIBXML_ICONV_ENABLED
1823 /**
1824 * xmlIconvWrapper:
1825 * @cd: iconv converter data structure
1826 * @out: a pointer to an array of bytes to store the result
1827 * @outlen: the length of @out
1828 * @in: a pointer to an array of input bytes
1829 * @inlen: the length of @in
1830 *
1831 * Returns 0 if success, or
1832 * -1 by lack of space, or
1833 * -2 if the transcoding fails (for *in is not valid utf8 string or
1834 * the result of transformation can't fit into the encoding we want), or
1835 * -3 if there the last byte can't form a single output char.
1836 *
1837 * The value of @inlen after return is the number of octets consumed
1838 * as the return value is positive, else unpredictable.
1839 * The value of @outlen after return is the number of octets produced.
1840 */
1841 static int
xmlIconvWrapper(iconv_t cd,unsigned char * out,int * outlen,const unsigned char * in,int * inlen)1842 xmlIconvWrapper(iconv_t cd, unsigned char *out, int *outlen,
1843 const unsigned char *in, int *inlen) {
1844 size_t icv_inlen, icv_outlen;
1845 const char *icv_in = (const char *) in;
1846 char *icv_out = (char *) out;
1847 size_t ret;
1848
1849 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL)) {
1850 if (outlen != NULL) *outlen = 0;
1851 return(-1);
1852 }
1853 icv_inlen = *inlen;
1854 icv_outlen = *outlen;
1855 /*
1856 * Some versions take const, other versions take non-const input.
1857 */
1858 ret = iconv(cd, (void *) &icv_in, &icv_inlen, &icv_out, &icv_outlen);
1859 *inlen -= icv_inlen;
1860 *outlen -= icv_outlen;
1861 if ((icv_inlen != 0) || (ret == (size_t) -1)) {
1862 #ifdef EILSEQ
1863 if (errno == EILSEQ) {
1864 return -2;
1865 } else
1866 #endif
1867 #ifdef E2BIG
1868 if (errno == E2BIG) {
1869 return -1;
1870 } else
1871 #endif
1872 #ifdef EINVAL
1873 if (errno == EINVAL) {
1874 return -3;
1875 } else
1876 #endif
1877 {
1878 return -3;
1879 }
1880 }
1881 return 0;
1882 }
1883 #endif /* LIBXML_ICONV_ENABLED */
1884
1885 /************************************************************************
1886 * *
1887 * ICU based generic conversion functions *
1888 * *
1889 ************************************************************************/
1890
1891 #ifdef LIBXML_ICU_ENABLED
1892 /**
1893 * xmlUconvWrapper:
1894 * @cd: ICU uconverter data structure
1895 * @toUnicode : non-zero if toUnicode. 0 otherwise.
1896 * @out: a pointer to an array of bytes to store the result
1897 * @outlen: the length of @out
1898 * @in: a pointer to an array of input bytes
1899 * @inlen: the length of @in
1900 * @flush: if true, indicates end of input
1901 *
1902 * Returns 0 if success, or
1903 * -1 by lack of space, or
1904 * -2 if the transcoding fails (for *in is not valid utf8 string or
1905 * the result of transformation can't fit into the encoding we want), or
1906 * -3 if there the last byte can't form a single output char.
1907 *
1908 * The value of @inlen after return is the number of octets consumed
1909 * as the return value is positive, else unpredictable.
1910 * The value of @outlen after return is the number of octets produced.
1911 */
1912 static int
xmlUconvWrapper(uconv_t * cd,int toUnicode,unsigned char * out,int * outlen,const unsigned char * in,int * inlen,int flush)1913 xmlUconvWrapper(uconv_t *cd, int toUnicode, unsigned char *out, int *outlen,
1914 const unsigned char *in, int *inlen, int flush) {
1915 const char *ucv_in = (const char *) in;
1916 char *ucv_out = (char *) out;
1917 UErrorCode err = U_ZERO_ERROR;
1918
1919 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL)) {
1920 if (outlen != NULL) *outlen = 0;
1921 return(-1);
1922 }
1923
1924 if (toUnicode) {
1925 /* encoding => UTF-16 => UTF-8 */
1926 ucnv_convertEx(cd->utf8, cd->uconv, &ucv_out, ucv_out + *outlen,
1927 &ucv_in, ucv_in + *inlen, cd->pivot_buf,
1928 &cd->pivot_source, &cd->pivot_target,
1929 cd->pivot_buf + ICU_PIVOT_BUF_SIZE, 0, flush, &err);
1930 } else {
1931 /* UTF-8 => UTF-16 => encoding */
1932 ucnv_convertEx(cd->uconv, cd->utf8, &ucv_out, ucv_out + *outlen,
1933 &ucv_in, ucv_in + *inlen, cd->pivot_buf,
1934 &cd->pivot_source, &cd->pivot_target,
1935 cd->pivot_buf + ICU_PIVOT_BUF_SIZE, 0, flush, &err);
1936 }
1937 *inlen = ucv_in - (const char*) in;
1938 *outlen = ucv_out - (char *) out;
1939 if (U_SUCCESS(err)) {
1940 /* reset pivot buf if this is the last call for input (flush==TRUE) */
1941 if (flush)
1942 cd->pivot_source = cd->pivot_target = cd->pivot_buf;
1943 return 0;
1944 }
1945 if (err == U_BUFFER_OVERFLOW_ERROR)
1946 return -1;
1947 if (err == U_INVALID_CHAR_FOUND || err == U_ILLEGAL_CHAR_FOUND)
1948 return -2;
1949 return -3;
1950 }
1951 #endif /* LIBXML_ICU_ENABLED */
1952
1953 /************************************************************************
1954 * *
1955 * The real API used by libxml for on-the-fly conversion *
1956 * *
1957 ************************************************************************/
1958
1959 /**
1960 * xmlEncInputChunk:
1961 * @handler: encoding handler
1962 * @out: a pointer to an array of bytes to store the result
1963 * @outlen: the length of @out
1964 * @in: a pointer to an array of input bytes
1965 * @inlen: the length of @in
1966 * @flush: flush (ICU-related)
1967 *
1968 * Returns 0 if success, or
1969 * -1 by lack of space, or
1970 * -2 if the transcoding fails (for *in is not valid utf8 string or
1971 * the result of transformation can't fit into the encoding we want), or
1972 * -3 if there the last byte can't form a single output char.
1973 *
1974 * The value of @inlen after return is the number of octets consumed
1975 * as the return value is 0, else unpredictable.
1976 * The value of @outlen after return is the number of octets produced.
1977 */
1978 static int
xmlEncInputChunk(xmlCharEncodingHandler * handler,unsigned char * out,int * outlen,const unsigned char * in,int * inlen,int flush)1979 xmlEncInputChunk(xmlCharEncodingHandler *handler, unsigned char *out,
1980 int *outlen, const unsigned char *in, int *inlen, int flush) {
1981 int ret;
1982 (void)flush;
1983
1984 if (handler->input != NULL) {
1985 ret = handler->input(out, outlen, in, inlen);
1986 if (ret > 0)
1987 ret = 0;
1988 }
1989 #ifdef LIBXML_ICONV_ENABLED
1990 else if (handler->iconv_in != NULL) {
1991 ret = xmlIconvWrapper(handler->iconv_in, out, outlen, in, inlen);
1992 }
1993 #endif /* LIBXML_ICONV_ENABLED */
1994 #ifdef LIBXML_ICU_ENABLED
1995 else if (handler->uconv_in != NULL) {
1996 ret = xmlUconvWrapper(handler->uconv_in, 1, out, outlen, in, inlen,
1997 flush);
1998 }
1999 #endif /* LIBXML_ICU_ENABLED */
2000 else {
2001 *outlen = 0;
2002 *inlen = 0;
2003 ret = -2;
2004 }
2005
2006 return(ret);
2007 }
2008
2009 /**
2010 * xmlEncOutputChunk:
2011 * @handler: encoding handler
2012 * @out: a pointer to an array of bytes to store the result
2013 * @outlen: the length of @out
2014 * @in: a pointer to an array of input bytes
2015 * @inlen: the length of @in
2016 *
2017 * Returns 0 if success, or
2018 * -1 by lack of space, or
2019 * -2 if the transcoding fails (for *in is not valid utf8 string or
2020 * the result of transformation can't fit into the encoding we want), or
2021 * -3 if there the last byte can't form a single output char.
2022 * -4 if no output function was found.
2023 *
2024 * The value of @inlen after return is the number of octets consumed
2025 * as the return value is 0, else unpredictable.
2026 * The value of @outlen after return is the number of octets produced.
2027 */
2028 static int
xmlEncOutputChunk(xmlCharEncodingHandler * handler,unsigned char * out,int * outlen,const unsigned char * in,int * inlen)2029 xmlEncOutputChunk(xmlCharEncodingHandler *handler, unsigned char *out,
2030 int *outlen, const unsigned char *in, int *inlen) {
2031 int ret;
2032
2033 if (handler->output != NULL) {
2034 ret = handler->output(out, outlen, in, inlen);
2035 if (ret > 0)
2036 ret = 0;
2037 }
2038 #ifdef LIBXML_ICONV_ENABLED
2039 else if (handler->iconv_out != NULL) {
2040 ret = xmlIconvWrapper(handler->iconv_out, out, outlen, in, inlen);
2041 }
2042 #endif /* LIBXML_ICONV_ENABLED */
2043 #ifdef LIBXML_ICU_ENABLED
2044 else if (handler->uconv_out != NULL) {
2045 ret = xmlUconvWrapper(handler->uconv_out, 0, out, outlen, in, inlen,
2046 1);
2047 }
2048 #endif /* LIBXML_ICU_ENABLED */
2049 else {
2050 *outlen = 0;
2051 *inlen = 0;
2052 ret = -4;
2053 }
2054
2055 return(ret);
2056 }
2057
2058 /**
2059 * xmlCharEncFirstLineInt:
2060 * @handler: char encoding transformation data structure
2061 * @out: an xmlBuffer for the output.
2062 * @in: an xmlBuffer for the input
2063 * @len: number of bytes to convert for the first line, or -1
2064 *
2065 * Front-end for the encoding handler input function, but handle only
2066 * the very first line, i.e. limit itself to 45 chars.
2067 *
2068 * Returns the number of byte written if success, or
2069 * -1 general error
2070 * -2 if the transcoding fails (for *in is not valid utf8 string or
2071 * the result of transformation can't fit into the encoding we want), or
2072 */
2073 int
xmlCharEncFirstLineInt(xmlCharEncodingHandler * handler,xmlBufferPtr out,xmlBufferPtr in,int len)2074 xmlCharEncFirstLineInt(xmlCharEncodingHandler *handler, xmlBufferPtr out,
2075 xmlBufferPtr in, int len) {
2076 int ret;
2077 int written;
2078 int toconv;
2079
2080 if (handler == NULL) return(-1);
2081 if (out == NULL) return(-1);
2082 if (in == NULL) return(-1);
2083
2084 /* calculate space available */
2085 written = out->size - out->use - 1; /* count '\0' */
2086 toconv = in->use;
2087 /*
2088 * echo '<?xml version="1.0" encoding="UCS4"?>' | wc -c => 38
2089 * 45 chars should be sufficient to reach the end of the encoding
2090 * declaration without going too far inside the document content.
2091 * on UTF-16 this means 90bytes, on UCS4 this means 180
2092 * The actual value depending on guessed encoding is passed as @len
2093 * if provided
2094 */
2095 if (len >= 0) {
2096 if (toconv > len)
2097 toconv = len;
2098 } else {
2099 if (toconv > 180)
2100 toconv = 180;
2101 }
2102 if (toconv * 2 >= written) {
2103 xmlBufferGrow(out, toconv * 2);
2104 written = out->size - out->use - 1;
2105 }
2106
2107 ret = xmlEncInputChunk(handler, &out->content[out->use], &written,
2108 in->content, &toconv, 0);
2109 xmlBufferShrink(in, toconv);
2110 out->use += written;
2111 out->content[out->use] = 0;
2112 if (ret == -1) ret = -3;
2113
2114 #ifdef DEBUG_ENCODING
2115 switch (ret) {
2116 case 0:
2117 xmlGenericError(xmlGenericErrorContext,
2118 "converted %d bytes to %d bytes of input\n",
2119 toconv, written);
2120 break;
2121 case -1:
2122 xmlGenericError(xmlGenericErrorContext,"converted %d bytes to %d bytes of input, %d left\n",
2123 toconv, written, in->use);
2124 break;
2125 case -2:
2126 xmlGenericError(xmlGenericErrorContext,
2127 "input conversion failed due to input error\n");
2128 break;
2129 case -3:
2130 xmlGenericError(xmlGenericErrorContext,"converted %d bytes to %d bytes of input, %d left\n",
2131 toconv, written, in->use);
2132 break;
2133 default:
2134 xmlGenericError(xmlGenericErrorContext,"Unknown input conversion failed %d\n", ret);
2135 }
2136 #endif /* DEBUG_ENCODING */
2137 /*
2138 * Ignore when input buffer is not on a boundary
2139 */
2140 if (ret == -3) ret = 0;
2141 if (ret == -1) ret = 0;
2142 return(written ? written : ret);
2143 }
2144
2145 /**
2146 * xmlCharEncFirstLine:
2147 * @handler: char encoding transformation data structure
2148 * @out: an xmlBuffer for the output.
2149 * @in: an xmlBuffer for the input
2150 *
2151 * Front-end for the encoding handler input function, but handle only
2152 * the very first line, i.e. limit itself to 45 chars.
2153 *
2154 * Returns the number of byte written if success, or
2155 * -1 general error
2156 * -2 if the transcoding fails (for *in is not valid utf8 string or
2157 * the result of transformation can't fit into the encoding we want), or
2158 */
2159 int
xmlCharEncFirstLine(xmlCharEncodingHandler * handler,xmlBufferPtr out,xmlBufferPtr in)2160 xmlCharEncFirstLine(xmlCharEncodingHandler *handler, xmlBufferPtr out,
2161 xmlBufferPtr in) {
2162 return(xmlCharEncFirstLineInt(handler, out, in, -1));
2163 }
2164
2165 /**
2166 * xmlCharEncFirstLineInput:
2167 * @input: a parser input buffer
2168 * @len: number of bytes to convert for the first line, or -1
2169 *
2170 * Front-end for the encoding handler input function, but handle only
2171 * the very first line. Point is that this is based on autodetection
2172 * of the encoding and once that first line is converted we may find
2173 * out that a different decoder is needed to process the input.
2174 *
2175 * Returns the number of byte written if success, or
2176 * -1 general error
2177 * -2 if the transcoding fails (for *in is not valid utf8 string or
2178 * the result of transformation can't fit into the encoding we want), or
2179 */
2180 int
xmlCharEncFirstLineInput(xmlParserInputBufferPtr input,int len)2181 xmlCharEncFirstLineInput(xmlParserInputBufferPtr input, int len)
2182 {
2183 int ret;
2184 size_t written;
2185 size_t toconv;
2186 int c_in;
2187 int c_out;
2188 xmlBufPtr in;
2189 xmlBufPtr out;
2190
2191 if ((input == NULL) || (input->encoder == NULL) ||
2192 (input->buffer == NULL) || (input->raw == NULL))
2193 return (-1);
2194 out = input->buffer;
2195 in = input->raw;
2196
2197 toconv = xmlBufUse(in);
2198 if (toconv == 0)
2199 return (0);
2200 written = xmlBufAvail(out);
2201 /*
2202 * echo '<?xml version="1.0" encoding="UCS4"?>' | wc -c => 38
2203 * 45 chars should be sufficient to reach the end of the encoding
2204 * declaration without going too far inside the document content.
2205 * on UTF-16 this means 90bytes, on UCS4 this means 180
2206 * The actual value depending on guessed encoding is passed as @len
2207 * if provided
2208 */
2209 if (len >= 0) {
2210 if (toconv > (unsigned int) len)
2211 toconv = len;
2212 } else {
2213 if (toconv > 180)
2214 toconv = 180;
2215 }
2216 if (toconv * 2 >= written) {
2217 xmlBufGrow(out, toconv * 2);
2218 written = xmlBufAvail(out);
2219 }
2220 if (written > 360)
2221 written = 360;
2222
2223 c_in = toconv;
2224 c_out = written;
2225 ret = xmlEncInputChunk(input->encoder, xmlBufEnd(out), &c_out,
2226 xmlBufContent(in), &c_in, 0);
2227 xmlBufShrink(in, c_in);
2228 xmlBufAddLen(out, c_out);
2229 if (ret == -1)
2230 ret = -3;
2231
2232 switch (ret) {
2233 case 0:
2234 #ifdef DEBUG_ENCODING
2235 xmlGenericError(xmlGenericErrorContext,
2236 "converted %d bytes to %d bytes of input\n",
2237 c_in, c_out);
2238 #endif
2239 break;
2240 case -1:
2241 #ifdef DEBUG_ENCODING
2242 xmlGenericError(xmlGenericErrorContext,
2243 "converted %d bytes to %d bytes of input, %d left\n",
2244 c_in, c_out, (int)xmlBufUse(in));
2245 #endif
2246 break;
2247 case -3:
2248 #ifdef DEBUG_ENCODING
2249 xmlGenericError(xmlGenericErrorContext,
2250 "converted %d bytes to %d bytes of input, %d left\n",
2251 c_in, c_out, (int)xmlBufUse(in));
2252 #endif
2253 break;
2254 case -2: {
2255 char buf[50];
2256 const xmlChar *content = xmlBufContent(in);
2257
2258 snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
2259 content[0], content[1],
2260 content[2], content[3]);
2261 buf[49] = 0;
2262 xmlEncodingErr(XML_I18N_CONV_FAILED,
2263 "input conversion failed due to input error, bytes %s\n",
2264 buf);
2265 }
2266 }
2267 /*
2268 * Ignore when input buffer is not on a boundary
2269 */
2270 if (ret == -3) ret = 0;
2271 if (ret == -1) ret = 0;
2272 return(c_out ? c_out : ret);
2273 }
2274
2275 /**
2276 * xmlCharEncInput:
2277 * @input: a parser input buffer
2278 * @flush: try to flush all the raw buffer
2279 *
2280 * Generic front-end for the encoding handler on parser input
2281 *
2282 * Returns the number of byte written if success, or
2283 * -1 general error
2284 * -2 if the transcoding fails (for *in is not valid utf8 string or
2285 * the result of transformation can't fit into the encoding we want), or
2286 */
2287 int
xmlCharEncInput(xmlParserInputBufferPtr input,int flush)2288 xmlCharEncInput(xmlParserInputBufferPtr input, int flush)
2289 {
2290 int ret;
2291 size_t written;
2292 size_t toconv;
2293 int c_in;
2294 int c_out;
2295 xmlBufPtr in;
2296 xmlBufPtr out;
2297
2298 if ((input == NULL) || (input->encoder == NULL) ||
2299 (input->buffer == NULL) || (input->raw == NULL))
2300 return (-1);
2301 out = input->buffer;
2302 in = input->raw;
2303
2304 toconv = xmlBufUse(in);
2305 if (toconv == 0)
2306 return (0);
2307 if ((toconv > 64 * 1024) && (flush == 0))
2308 toconv = 64 * 1024;
2309 written = xmlBufAvail(out);
2310 if (toconv * 2 >= written) {
2311 xmlBufGrow(out, toconv * 2);
2312 written = xmlBufAvail(out);
2313 }
2314 if ((written > 128 * 1024) && (flush == 0))
2315 written = 128 * 1024;
2316
2317 c_in = toconv;
2318 c_out = written;
2319 ret = xmlEncInputChunk(input->encoder, xmlBufEnd(out), &c_out,
2320 xmlBufContent(in), &c_in, flush);
2321 xmlBufShrink(in, c_in);
2322 xmlBufAddLen(out, c_out);
2323 if (ret == -1)
2324 ret = -3;
2325
2326 switch (ret) {
2327 case 0:
2328 #ifdef DEBUG_ENCODING
2329 xmlGenericError(xmlGenericErrorContext,
2330 "converted %d bytes to %d bytes of input\n",
2331 c_in, c_out);
2332 #endif
2333 break;
2334 case -1:
2335 #ifdef DEBUG_ENCODING
2336 xmlGenericError(xmlGenericErrorContext,
2337 "converted %d bytes to %d bytes of input, %d left\n",
2338 c_in, c_out, (int)xmlBufUse(in));
2339 #endif
2340 break;
2341 case -3:
2342 #ifdef DEBUG_ENCODING
2343 xmlGenericError(xmlGenericErrorContext,
2344 "converted %d bytes to %d bytes of input, %d left\n",
2345 c_in, c_out, (int)xmlBufUse(in));
2346 #endif
2347 break;
2348 case -2: {
2349 char buf[50];
2350 const xmlChar *content = xmlBufContent(in);
2351
2352 snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
2353 content[0], content[1],
2354 content[2], content[3]);
2355 buf[49] = 0;
2356 xmlEncodingErr(XML_I18N_CONV_FAILED,
2357 "input conversion failed due to input error, bytes %s\n",
2358 buf);
2359 }
2360 }
2361 /*
2362 * Ignore when input buffer is not on a boundary
2363 */
2364 if (ret == -3)
2365 ret = 0;
2366 return (c_out? c_out : ret);
2367 }
2368
2369 /**
2370 * xmlCharEncInFunc:
2371 * @handler: char encoding transformation data structure
2372 * @out: an xmlBuffer for the output.
2373 * @in: an xmlBuffer for the input
2374 *
2375 * Generic front-end for the encoding handler input function
2376 *
2377 * Returns the number of byte written if success, or
2378 * -1 general error
2379 * -2 if the transcoding fails (for *in is not valid utf8 string or
2380 * the result of transformation can't fit into the encoding we want), or
2381 */
2382 int
xmlCharEncInFunc(xmlCharEncodingHandler * handler,xmlBufferPtr out,xmlBufferPtr in)2383 xmlCharEncInFunc(xmlCharEncodingHandler * handler, xmlBufferPtr out,
2384 xmlBufferPtr in)
2385 {
2386 int ret;
2387 int written;
2388 int toconv;
2389
2390 if (handler == NULL)
2391 return (-1);
2392 if (out == NULL)
2393 return (-1);
2394 if (in == NULL)
2395 return (-1);
2396
2397 toconv = in->use;
2398 if (toconv == 0)
2399 return (0);
2400 written = out->size - out->use -1; /* count '\0' */
2401 if (toconv * 2 >= written) {
2402 xmlBufferGrow(out, out->size + toconv * 2);
2403 written = out->size - out->use - 1;
2404 }
2405 ret = xmlEncInputChunk(handler, &out->content[out->use], &written,
2406 in->content, &toconv, 1);
2407 xmlBufferShrink(in, toconv);
2408 out->use += written;
2409 out->content[out->use] = 0;
2410 if (ret == -1)
2411 ret = -3;
2412
2413 switch (ret) {
2414 case 0:
2415 #ifdef DEBUG_ENCODING
2416 xmlGenericError(xmlGenericErrorContext,
2417 "converted %d bytes to %d bytes of input\n",
2418 toconv, written);
2419 #endif
2420 break;
2421 case -1:
2422 #ifdef DEBUG_ENCODING
2423 xmlGenericError(xmlGenericErrorContext,
2424 "converted %d bytes to %d bytes of input, %d left\n",
2425 toconv, written, in->use);
2426 #endif
2427 break;
2428 case -3:
2429 #ifdef DEBUG_ENCODING
2430 xmlGenericError(xmlGenericErrorContext,
2431 "converted %d bytes to %d bytes of input, %d left\n",
2432 toconv, written, in->use);
2433 #endif
2434 break;
2435 case -2: {
2436 char buf[50];
2437
2438 snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
2439 in->content[0], in->content[1],
2440 in->content[2], in->content[3]);
2441 buf[49] = 0;
2442 xmlEncodingErr(XML_I18N_CONV_FAILED,
2443 "input conversion failed due to input error, bytes %s\n",
2444 buf);
2445 }
2446 }
2447 /*
2448 * Ignore when input buffer is not on a boundary
2449 */
2450 if (ret == -3)
2451 ret = 0;
2452 return (written? written : ret);
2453 }
2454
2455 #ifdef LIBXML_OUTPUT_ENABLED
2456 /**
2457 * xmlCharEncOutput:
2458 * @output: a parser output buffer
2459 * @init: is this an initialization call without data
2460 *
2461 * Generic front-end for the encoding handler on parser output
2462 * a first call with @init == 1 has to be made first to initiate the
2463 * output in case of non-stateless encoding needing to initiate their
2464 * state or the output (like the BOM in UTF16).
2465 * In case of UTF8 sequence conversion errors for the given encoder,
2466 * the content will be automatically remapped to a CharRef sequence.
2467 *
2468 * Returns the number of byte written if success, or
2469 * -1 general error
2470 * -2 if the transcoding fails (for *in is not valid utf8 string or
2471 * the result of transformation can't fit into the encoding we want), or
2472 */
2473 int
xmlCharEncOutput(xmlOutputBufferPtr output,int init)2474 xmlCharEncOutput(xmlOutputBufferPtr output, int init)
2475 {
2476 int ret;
2477 size_t written;
2478 int writtentot = 0;
2479 size_t toconv;
2480 int c_in;
2481 int c_out;
2482 xmlBufPtr in;
2483 xmlBufPtr out;
2484
2485 if ((output == NULL) || (output->encoder == NULL) ||
2486 (output->buffer == NULL) || (output->conv == NULL))
2487 return (-1);
2488 out = output->conv;
2489 in = output->buffer;
2490
2491 retry:
2492
2493 written = xmlBufAvail(out);
2494
2495 /*
2496 * First specific handling of the initialization call
2497 */
2498 if (init) {
2499 c_in = 0;
2500 c_out = written;
2501 /* TODO: Check return value. */
2502 xmlEncOutputChunk(output->encoder, xmlBufEnd(out), &c_out,
2503 NULL, &c_in);
2504 xmlBufAddLen(out, c_out);
2505 #ifdef DEBUG_ENCODING
2506 xmlGenericError(xmlGenericErrorContext,
2507 "initialized encoder\n");
2508 #endif
2509 return(c_out);
2510 }
2511
2512 /*
2513 * Conversion itself.
2514 */
2515 toconv = xmlBufUse(in);
2516 if (toconv == 0)
2517 return (writtentot);
2518 if (toconv > 64 * 1024)
2519 toconv = 64 * 1024;
2520 if (toconv * 4 >= written) {
2521 xmlBufGrow(out, toconv * 4);
2522 written = xmlBufAvail(out);
2523 }
2524 if (written > 256 * 1024)
2525 written = 256 * 1024;
2526
2527 c_in = toconv;
2528 c_out = written;
2529 ret = xmlEncOutputChunk(output->encoder, xmlBufEnd(out), &c_out,
2530 xmlBufContent(in), &c_in);
2531 xmlBufShrink(in, c_in);
2532 xmlBufAddLen(out, c_out);
2533 writtentot += c_out;
2534 if (ret == -1) {
2535 if (c_out > 0) {
2536 /* Can be a limitation of iconv or uconv */
2537 goto retry;
2538 }
2539 ret = -3;
2540 }
2541
2542 /*
2543 * Attempt to handle error cases
2544 */
2545 switch (ret) {
2546 case 0:
2547 #ifdef DEBUG_ENCODING
2548 xmlGenericError(xmlGenericErrorContext,
2549 "converted %d bytes to %d bytes of output\n",
2550 c_in, c_out);
2551 #endif
2552 break;
2553 case -1:
2554 #ifdef DEBUG_ENCODING
2555 xmlGenericError(xmlGenericErrorContext,
2556 "output conversion failed by lack of space\n");
2557 #endif
2558 break;
2559 case -3:
2560 #ifdef DEBUG_ENCODING
2561 xmlGenericError(xmlGenericErrorContext,"converted %d bytes to %d bytes of output %d left\n",
2562 c_in, c_out, (int) xmlBufUse(in));
2563 #endif
2564 break;
2565 case -4:
2566 xmlEncodingErr(XML_I18N_NO_OUTPUT,
2567 "xmlCharEncOutFunc: no output function !\n", NULL);
2568 ret = -1;
2569 break;
2570 case -2: {
2571 xmlChar charref[20];
2572 int len = (int) xmlBufUse(in);
2573 xmlChar *content = xmlBufContent(in);
2574 int cur, charrefLen;
2575
2576 cur = xmlGetUTF8Char(content, &len);
2577 if (cur <= 0)
2578 break;
2579
2580 #ifdef DEBUG_ENCODING
2581 xmlGenericError(xmlGenericErrorContext,
2582 "handling output conversion error\n");
2583 xmlGenericError(xmlGenericErrorContext,
2584 "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
2585 content[0], content[1],
2586 content[2], content[3]);
2587 #endif
2588 /*
2589 * Removes the UTF8 sequence, and replace it by a charref
2590 * and continue the transcoding phase, hoping the error
2591 * did not mangle the encoder state.
2592 */
2593 charrefLen = snprintf((char *) &charref[0], sizeof(charref),
2594 "&#%d;", cur);
2595 xmlBufShrink(in, len);
2596 xmlBufGrow(out, charrefLen * 4);
2597 c_out = xmlBufAvail(out);
2598 c_in = charrefLen;
2599 ret = xmlEncOutputChunk(output->encoder, xmlBufEnd(out), &c_out,
2600 charref, &c_in);
2601
2602 if ((ret < 0) || (c_in != charrefLen)) {
2603 char buf[50];
2604
2605 snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
2606 content[0], content[1],
2607 content[2], content[3]);
2608 buf[49] = 0;
2609 xmlEncodingErr(XML_I18N_CONV_FAILED,
2610 "output conversion failed due to conv error, bytes %s\n",
2611 buf);
2612 if (xmlBufGetAllocationScheme(in) != XML_BUFFER_ALLOC_IMMUTABLE)
2613 content[0] = ' ';
2614 break;
2615 }
2616
2617 xmlBufAddLen(out, c_out);
2618 writtentot += c_out;
2619 goto retry;
2620 }
2621 }
2622 return(writtentot ? writtentot : ret);
2623 }
2624 #endif
2625
2626 /**
2627 * xmlCharEncOutFunc:
2628 * @handler: char encoding transformation data structure
2629 * @out: an xmlBuffer for the output.
2630 * @in: an xmlBuffer for the input
2631 *
2632 * Generic front-end for the encoding handler output function
2633 * a first call with @in == NULL has to be made firs to initiate the
2634 * output in case of non-stateless encoding needing to initiate their
2635 * state or the output (like the BOM in UTF16).
2636 * In case of UTF8 sequence conversion errors for the given encoder,
2637 * the content will be automatically remapped to a CharRef sequence.
2638 *
2639 * Returns the number of byte written if success, or
2640 * -1 general error
2641 * -2 if the transcoding fails (for *in is not valid utf8 string or
2642 * the result of transformation can't fit into the encoding we want), or
2643 */
2644 int
xmlCharEncOutFunc(xmlCharEncodingHandler * handler,xmlBufferPtr out,xmlBufferPtr in)2645 xmlCharEncOutFunc(xmlCharEncodingHandler *handler, xmlBufferPtr out,
2646 xmlBufferPtr in) {
2647 int ret;
2648 int written;
2649 int writtentot = 0;
2650 int toconv;
2651
2652 if (handler == NULL) return(-1);
2653 if (out == NULL) return(-1);
2654
2655 retry:
2656
2657 written = out->size - out->use;
2658
2659 if (written > 0)
2660 written--; /* Gennady: count '/0' */
2661
2662 /*
2663 * First specific handling of in = NULL, i.e. the initialization call
2664 */
2665 if (in == NULL) {
2666 toconv = 0;
2667 /* TODO: Check return value. */
2668 xmlEncOutputChunk(handler, &out->content[out->use], &written,
2669 NULL, &toconv);
2670 out->use += written;
2671 out->content[out->use] = 0;
2672 #ifdef DEBUG_ENCODING
2673 xmlGenericError(xmlGenericErrorContext,
2674 "initialized encoder\n");
2675 #endif
2676 return(0);
2677 }
2678
2679 /*
2680 * Conversion itself.
2681 */
2682 toconv = in->use;
2683 if (toconv == 0)
2684 return(0);
2685 if (toconv * 4 >= written) {
2686 xmlBufferGrow(out, toconv * 4);
2687 written = out->size - out->use - 1;
2688 }
2689 ret = xmlEncOutputChunk(handler, &out->content[out->use], &written,
2690 in->content, &toconv);
2691 xmlBufferShrink(in, toconv);
2692 out->use += written;
2693 writtentot += written;
2694 out->content[out->use] = 0;
2695 if (ret == -1) {
2696 if (written > 0) {
2697 /* Can be a limitation of iconv or uconv */
2698 goto retry;
2699 }
2700 ret = -3;
2701 }
2702
2703 /*
2704 * Attempt to handle error cases
2705 */
2706 switch (ret) {
2707 case 0:
2708 #ifdef DEBUG_ENCODING
2709 xmlGenericError(xmlGenericErrorContext,
2710 "converted %d bytes to %d bytes of output\n",
2711 toconv, written);
2712 #endif
2713 break;
2714 case -1:
2715 #ifdef DEBUG_ENCODING
2716 xmlGenericError(xmlGenericErrorContext,
2717 "output conversion failed by lack of space\n");
2718 #endif
2719 break;
2720 case -3:
2721 #ifdef DEBUG_ENCODING
2722 xmlGenericError(xmlGenericErrorContext,"converted %d bytes to %d bytes of output %d left\n",
2723 toconv, written, in->use);
2724 #endif
2725 break;
2726 case -4:
2727 xmlEncodingErr(XML_I18N_NO_OUTPUT,
2728 "xmlCharEncOutFunc: no output function !\n", NULL);
2729 ret = -1;
2730 break;
2731 case -2: {
2732 xmlChar charref[20];
2733 int len = in->use;
2734 const xmlChar *utf = (const xmlChar *) in->content;
2735 int cur, charrefLen;
2736
2737 cur = xmlGetUTF8Char(utf, &len);
2738 if (cur <= 0)
2739 break;
2740
2741 #ifdef DEBUG_ENCODING
2742 xmlGenericError(xmlGenericErrorContext,
2743 "handling output conversion error\n");
2744 xmlGenericError(xmlGenericErrorContext,
2745 "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
2746 in->content[0], in->content[1],
2747 in->content[2], in->content[3]);
2748 #endif
2749 /*
2750 * Removes the UTF8 sequence, and replace it by a charref
2751 * and continue the transcoding phase, hoping the error
2752 * did not mangle the encoder state.
2753 */
2754 charrefLen = snprintf((char *) &charref[0], sizeof(charref),
2755 "&#%d;", cur);
2756 xmlBufferShrink(in, len);
2757 xmlBufferGrow(out, charrefLen * 4);
2758 written = out->size - out->use - 1;
2759 toconv = charrefLen;
2760 ret = xmlEncOutputChunk(handler, &out->content[out->use], &written,
2761 charref, &toconv);
2762
2763 if ((ret < 0) || (toconv != charrefLen)) {
2764 char buf[50];
2765
2766 snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
2767 in->content[0], in->content[1],
2768 in->content[2], in->content[3]);
2769 buf[49] = 0;
2770 xmlEncodingErr(XML_I18N_CONV_FAILED,
2771 "output conversion failed due to conv error, bytes %s\n",
2772 buf);
2773 if (in->alloc != XML_BUFFER_ALLOC_IMMUTABLE)
2774 in->content[0] = ' ';
2775 break;
2776 }
2777
2778 out->use += written;
2779 writtentot += written;
2780 out->content[out->use] = 0;
2781 goto retry;
2782 }
2783 }
2784 return(writtentot ? writtentot : ret);
2785 }
2786
2787 /**
2788 * xmlCharEncCloseFunc:
2789 * @handler: char encoding transformation data structure
2790 *
2791 * Generic front-end for encoding handler close function
2792 *
2793 * Returns 0 if success, or -1 in case of error
2794 */
2795 int
xmlCharEncCloseFunc(xmlCharEncodingHandler * handler)2796 xmlCharEncCloseFunc(xmlCharEncodingHandler *handler) {
2797 int ret = 0;
2798 int tofree = 0;
2799 int i, handler_in_list = 0;
2800
2801 /* Avoid unused variable warning if features are disabled. */
2802 (void) handler_in_list;
2803
2804 if (handler == NULL) return(-1);
2805 if (handler->name == NULL) return(-1);
2806 if (handlers != NULL) {
2807 for (i = 0;i < nbCharEncodingHandler; i++) {
2808 if (handler == handlers[i]) {
2809 handler_in_list = 1;
2810 break;
2811 }
2812 }
2813 }
2814 #ifdef LIBXML_ICONV_ENABLED
2815 /*
2816 * Iconv handlers can be used only once, free the whole block.
2817 * and the associated icon resources.
2818 */
2819 if ((handler_in_list == 0) &&
2820 ((handler->iconv_out != NULL) || (handler->iconv_in != NULL))) {
2821 tofree = 1;
2822 if (handler->iconv_out != NULL) {
2823 if (iconv_close(handler->iconv_out))
2824 ret = -1;
2825 handler->iconv_out = NULL;
2826 }
2827 if (handler->iconv_in != NULL) {
2828 if (iconv_close(handler->iconv_in))
2829 ret = -1;
2830 handler->iconv_in = NULL;
2831 }
2832 }
2833 #endif /* LIBXML_ICONV_ENABLED */
2834 #ifdef LIBXML_ICU_ENABLED
2835 if ((handler_in_list == 0) &&
2836 ((handler->uconv_out != NULL) || (handler->uconv_in != NULL))) {
2837 tofree = 1;
2838 if (handler->uconv_out != NULL) {
2839 closeIcuConverter(handler->uconv_out);
2840 handler->uconv_out = NULL;
2841 }
2842 if (handler->uconv_in != NULL) {
2843 closeIcuConverter(handler->uconv_in);
2844 handler->uconv_in = NULL;
2845 }
2846 }
2847 #endif
2848 if (tofree) {
2849 /* free up only dynamic handlers iconv/uconv */
2850 if (handler->name != NULL)
2851 xmlFree(handler->name);
2852 handler->name = NULL;
2853 xmlFree(handler);
2854 }
2855 #ifdef DEBUG_ENCODING
2856 if (ret)
2857 xmlGenericError(xmlGenericErrorContext,
2858 "failed to close the encoding handler\n");
2859 else
2860 xmlGenericError(xmlGenericErrorContext,
2861 "closed the encoding handler\n");
2862 #endif
2863
2864 return(ret);
2865 }
2866
2867 /**
2868 * xmlByteConsumed:
2869 * @ctxt: an XML parser context
2870 *
2871 * This function provides the current index of the parser relative
2872 * to the start of the current entity. This function is computed in
2873 * bytes from the beginning starting at zero and finishing at the
2874 * size in byte of the file if parsing a file. The function is
2875 * of constant cost if the input is UTF-8 but can be costly if run
2876 * on non-UTF-8 input.
2877 *
2878 * Returns the index in bytes from the beginning of the entity or -1
2879 * in case the index could not be computed.
2880 */
2881 long
xmlByteConsumed(xmlParserCtxtPtr ctxt)2882 xmlByteConsumed(xmlParserCtxtPtr ctxt) {
2883 xmlParserInputPtr in;
2884
2885 if (ctxt == NULL) return(-1);
2886 in = ctxt->input;
2887 if (in == NULL) return(-1);
2888 if ((in->buf != NULL) && (in->buf->encoder != NULL)) {
2889 unsigned int unused = 0;
2890 xmlCharEncodingHandler * handler = in->buf->encoder;
2891 /*
2892 * Encoding conversion, compute the number of unused original
2893 * bytes from the input not consumed and subtract that from
2894 * the raw consumed value, this is not a cheap operation
2895 */
2896 if (in->end - in->cur > 0) {
2897 unsigned char convbuf[32000];
2898 const unsigned char *cur = (const unsigned char *)in->cur;
2899 int toconv = in->end - in->cur, written = 32000;
2900
2901 int ret;
2902
2903 do {
2904 toconv = in->end - cur;
2905 written = 32000;
2906 ret = xmlEncOutputChunk(handler, &convbuf[0], &written,
2907 cur, &toconv);
2908 if (ret < 0) {
2909 if (written > 0)
2910 ret = -2;
2911 else
2912 return(-1);
2913 }
2914 unused += written;
2915 cur += toconv;
2916 } while (ret == -2);
2917 }
2918 if (in->buf->rawconsumed < unused)
2919 return(-1);
2920 return(in->buf->rawconsumed - unused);
2921 }
2922 return(in->consumed + (in->cur - in->base));
2923 }
2924
2925 #if !defined(LIBXML_ICONV_ENABLED) && !defined(LIBXML_ICU_ENABLED)
2926 #ifdef LIBXML_ISO8859X_ENABLED
2927
2928 /**
2929 * UTF8ToISO8859x:
2930 * @out: a pointer to an array of bytes to store the result
2931 * @outlen: the length of @out
2932 * @in: a pointer to an array of UTF-8 chars
2933 * @inlen: the length of @in
2934 * @xlattable: the 2-level transcoding table
2935 *
2936 * Take a block of UTF-8 chars in and try to convert it to an ISO 8859-*
2937 * block of chars out.
2938 *
2939 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2940 * The value of @inlen after return is the number of octets consumed
2941 * as the return value is positive, else unpredictable.
2942 * The value of @outlen after return is the number of octets consumed.
2943 */
2944 static int
UTF8ToISO8859x(unsigned char * out,int * outlen,const unsigned char * in,int * inlen,const unsigned char * const xlattable)2945 UTF8ToISO8859x(unsigned char* out, int *outlen,
2946 const unsigned char* in, int *inlen,
2947 const unsigned char* const xlattable) {
2948 const unsigned char* outstart = out;
2949 const unsigned char* inend;
2950 const unsigned char* instart = in;
2951 const unsigned char* processed = in;
2952
2953 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) ||
2954 (xlattable == NULL))
2955 return(-1);
2956 if (in == NULL) {
2957 /*
2958 * initialization nothing to do
2959 */
2960 *outlen = 0;
2961 *inlen = 0;
2962 return(0);
2963 }
2964 inend = in + (*inlen);
2965 while (in < inend) {
2966 unsigned char d = *in++;
2967 if (d < 0x80) {
2968 *out++ = d;
2969 } else if (d < 0xC0) {
2970 /* trailing byte in leading position */
2971 *outlen = out - outstart;
2972 *inlen = processed - instart;
2973 return(-2);
2974 } else if (d < 0xE0) {
2975 unsigned char c;
2976 if (!(in < inend)) {
2977 /* trailing byte not in input buffer */
2978 *outlen = out - outstart;
2979 *inlen = processed - instart;
2980 return(-3);
2981 }
2982 c = *in++;
2983 if ((c & 0xC0) != 0x80) {
2984 /* not a trailing byte */
2985 *outlen = out - outstart;
2986 *inlen = processed - instart;
2987 return(-2);
2988 }
2989 c = c & 0x3F;
2990 d = d & 0x1F;
2991 d = xlattable [48 + c + xlattable [d] * 64];
2992 if (d == 0) {
2993 /* not in character set */
2994 *outlen = out - outstart;
2995 *inlen = processed - instart;
2996 return(-2);
2997 }
2998 *out++ = d;
2999 } else if (d < 0xF0) {
3000 unsigned char c1;
3001 unsigned char c2;
3002 if (!(in < inend - 1)) {
3003 /* trailing bytes not in input buffer */
3004 *outlen = out - outstart;
3005 *inlen = processed - instart;
3006 return(-3);
3007 }
3008 c1 = *in++;
3009 if ((c1 & 0xC0) != 0x80) {
3010 /* not a trailing byte (c1) */
3011 *outlen = out - outstart;
3012 *inlen = processed - instart;
3013 return(-2);
3014 }
3015 c2 = *in++;
3016 if ((c2 & 0xC0) != 0x80) {
3017 /* not a trailing byte (c2) */
3018 *outlen = out - outstart;
3019 *inlen = processed - instart;
3020 return(-2);
3021 }
3022 c1 = c1 & 0x3F;
3023 c2 = c2 & 0x3F;
3024 d = d & 0x0F;
3025 d = xlattable [48 + c2 + xlattable [48 + c1 +
3026 xlattable [32 + d] * 64] * 64];
3027 if (d == 0) {
3028 /* not in character set */
3029 *outlen = out - outstart;
3030 *inlen = processed - instart;
3031 return(-2);
3032 }
3033 *out++ = d;
3034 } else {
3035 /* cannot transcode >= U+010000 */
3036 *outlen = out - outstart;
3037 *inlen = processed - instart;
3038 return(-2);
3039 }
3040 processed = in;
3041 }
3042 *outlen = out - outstart;
3043 *inlen = processed - instart;
3044 return(*outlen);
3045 }
3046
3047 /**
3048 * ISO8859xToUTF8
3049 * @out: a pointer to an array of bytes to store the result
3050 * @outlen: the length of @out
3051 * @in: a pointer to an array of ISO Latin 1 chars
3052 * @inlen: the length of @in
3053 *
3054 * Take a block of ISO 8859-* chars in and try to convert it to an UTF-8
3055 * block of chars out.
3056 * Returns 0 if success, or -1 otherwise
3057 * The value of @inlen after return is the number of octets consumed
3058 * The value of @outlen after return is the number of octets produced.
3059 */
3060 static int
ISO8859xToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen,unsigned short const * unicodetable)3061 ISO8859xToUTF8(unsigned char* out, int *outlen,
3062 const unsigned char* in, int *inlen,
3063 unsigned short const *unicodetable) {
3064 unsigned char* outstart = out;
3065 unsigned char* outend;
3066 const unsigned char* instart = in;
3067 const unsigned char* inend;
3068 const unsigned char* instop;
3069 unsigned int c;
3070
3071 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) ||
3072 (in == NULL) || (unicodetable == NULL))
3073 return(-1);
3074 outend = out + *outlen;
3075 inend = in + *inlen;
3076 instop = inend;
3077
3078 while ((in < inend) && (out < outend - 2)) {
3079 if (*in >= 0x80) {
3080 c = unicodetable [*in - 0x80];
3081 if (c == 0) {
3082 /* undefined code point */
3083 *outlen = out - outstart;
3084 *inlen = in - instart;
3085 return (-1);
3086 }
3087 if (c < 0x800) {
3088 *out++ = ((c >> 6) & 0x1F) | 0xC0;
3089 *out++ = (c & 0x3F) | 0x80;
3090 } else {
3091 *out++ = ((c >> 12) & 0x0F) | 0xE0;
3092 *out++ = ((c >> 6) & 0x3F) | 0x80;
3093 *out++ = (c & 0x3F) | 0x80;
3094 }
3095 ++in;
3096 }
3097 if (instop - in > outend - out) instop = in + (outend - out);
3098 while ((*in < 0x80) && (in < instop)) {
3099 *out++ = *in++;
3100 }
3101 }
3102 if ((in < inend) && (out < outend) && (*in < 0x80)) {
3103 *out++ = *in++;
3104 }
3105 if ((in < inend) && (out < outend) && (*in < 0x80)) {
3106 *out++ = *in++;
3107 }
3108 *outlen = out - outstart;
3109 *inlen = in - instart;
3110 return (*outlen);
3111 }
3112
3113
3114 /************************************************************************
3115 * Lookup tables for ISO-8859-2..ISO-8859-16 transcoding *
3116 ************************************************************************/
3117
3118 static unsigned short const xmlunicodetable_ISO8859_2 [128] = {
3119 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3120 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3121 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3122 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3123 0x00a0, 0x0104, 0x02d8, 0x0141, 0x00a4, 0x013d, 0x015a, 0x00a7,
3124 0x00a8, 0x0160, 0x015e, 0x0164, 0x0179, 0x00ad, 0x017d, 0x017b,
3125 0x00b0, 0x0105, 0x02db, 0x0142, 0x00b4, 0x013e, 0x015b, 0x02c7,
3126 0x00b8, 0x0161, 0x015f, 0x0165, 0x017a, 0x02dd, 0x017e, 0x017c,
3127 0x0154, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0139, 0x0106, 0x00c7,
3128 0x010c, 0x00c9, 0x0118, 0x00cb, 0x011a, 0x00cd, 0x00ce, 0x010e,
3129 0x0110, 0x0143, 0x0147, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x00d7,
3130 0x0158, 0x016e, 0x00da, 0x0170, 0x00dc, 0x00dd, 0x0162, 0x00df,
3131 0x0155, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x013a, 0x0107, 0x00e7,
3132 0x010d, 0x00e9, 0x0119, 0x00eb, 0x011b, 0x00ed, 0x00ee, 0x010f,
3133 0x0111, 0x0144, 0x0148, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x00f7,
3134 0x0159, 0x016f, 0x00fa, 0x0171, 0x00fc, 0x00fd, 0x0163, 0x02d9,
3135 };
3136
3137 static const unsigned char xmltranscodetable_ISO8859_2 [48 + 6 * 64] = {
3138 "\x00\x00\x01\x05\x02\x04\x00\x00\x00\x00\x00\x03\x00\x00\x00\x00"
3139 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3140 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3141 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3142 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3143 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3144 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3145 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3146 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3147 "\xa0\x00\x00\x00\xa4\x00\x00\xa7\xa8\x00\x00\x00\x00\xad\x00\x00"
3148 "\xb0\x00\x00\x00\xb4\x00\x00\x00\xb8\x00\x00\x00\x00\x00\x00\x00"
3149 "\x00\x00\xc3\xe3\xa1\xb1\xc6\xe6\x00\x00\x00\x00\xc8\xe8\xcf\xef"
3150 "\xd0\xf0\x00\x00\x00\x00\x00\x00\xca\xea\xcc\xec\x00\x00\x00\x00"
3151 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3152 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\xc5\xe5\x00\x00\xa5\xb5\x00"
3153 "\x00\x00\x00\x00\x00\x00\x00\xb7\x00\x00\x00\x00\x00\x00\x00\x00"
3154 "\x00\x00\x00\x00\x00\x00\x00\x00\xa2\xff\x00\xb2\x00\xbd\x00\x00"
3155 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3156 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3157 "\x00\xa3\xb3\xd1\xf1\x00\x00\xd2\xf2\x00\x00\x00\x00\x00\x00\x00"
3158 "\xd5\xf5\x00\x00\xc0\xe0\x00\x00\xd8\xf8\xa6\xb6\x00\x00\xaa\xba"
3159 "\xa9\xb9\xde\xfe\xab\xbb\x00\x00\x00\x00\x00\x00\x00\x00\xd9\xf9"
3160 "\xdb\xfb\x00\x00\x00\x00\x00\x00\x00\xac\xbc\xaf\xbf\xae\xbe\x00"
3161 "\x00\xc1\xc2\x00\xc4\x00\x00\xc7\x00\xc9\x00\xcb\x00\xcd\xce\x00"
3162 "\x00\x00\x00\xd3\xd4\x00\xd6\xd7\x00\x00\xda\x00\xdc\xdd\x00\xdf"
3163 "\x00\xe1\xe2\x00\xe4\x00\x00\xe7\x00\xe9\x00\xeb\x00\xed\xee\x00"
3164 "\x00\x00\x00\xf3\xf4\x00\xf6\xf7\x00\x00\xfa\x00\xfc\xfd\x00\x00"
3165 };
3166
3167 static unsigned short const xmlunicodetable_ISO8859_3 [128] = {
3168 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3169 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3170 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3171 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3172 0x00a0, 0x0126, 0x02d8, 0x00a3, 0x00a4, 0x0000, 0x0124, 0x00a7,
3173 0x00a8, 0x0130, 0x015e, 0x011e, 0x0134, 0x00ad, 0x0000, 0x017b,
3174 0x00b0, 0x0127, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x0125, 0x00b7,
3175 0x00b8, 0x0131, 0x015f, 0x011f, 0x0135, 0x00bd, 0x0000, 0x017c,
3176 0x00c0, 0x00c1, 0x00c2, 0x0000, 0x00c4, 0x010a, 0x0108, 0x00c7,
3177 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
3178 0x0000, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x0120, 0x00d6, 0x00d7,
3179 0x011c, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x016c, 0x015c, 0x00df,
3180 0x00e0, 0x00e1, 0x00e2, 0x0000, 0x00e4, 0x010b, 0x0109, 0x00e7,
3181 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
3182 0x0000, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x0121, 0x00f6, 0x00f7,
3183 0x011d, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x016d, 0x015d, 0x02d9,
3184 };
3185
3186 static const unsigned char xmltranscodetable_ISO8859_3 [48 + 7 * 64] = {
3187 "\x04\x00\x01\x06\x02\x05\x00\x00\x00\x00\x00\x03\x00\x00\x00\x00"
3188 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3189 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3190 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3191 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3192 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3193 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3194 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3195 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3196 "\xa0\x00\x00\xa3\xa4\x00\x00\xa7\xa8\x00\x00\x00\x00\xad\x00\x00"
3197 "\xb0\x00\xb2\xb3\xb4\xb5\x00\xb7\xb8\x00\x00\x00\x00\xbd\x00\x00"
3198 "\x00\x00\x00\x00\x00\x00\x00\x00\xc6\xe6\xc5\xe5\x00\x00\x00\x00"
3199 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xd8\xf8\xab\xbb"
3200 "\xd5\xf5\x00\x00\xa6\xb6\xa1\xb1\x00\x00\x00\x00\x00\x00\x00\x00"
3201 "\xa9\xb9\x00\x00\xac\xbc\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3202 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3203 "\x00\x00\x00\x00\x00\x00\x00\x00\xa2\xff\x00\x00\x00\x00\x00\x00"
3204 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3205 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3206 "\xf0\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3207 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3208 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3209 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3210 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3211 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xde\xfe\xaa\xba"
3212 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xdd\xfd\x00\x00"
3213 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xaf\xbf\x00\x00\x00"
3214 "\xc0\xc1\xc2\x00\xc4\x00\x00\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3215 "\x00\xd1\xd2\xd3\xd4\x00\xd6\xd7\x00\xd9\xda\xdb\xdc\x00\x00\xdf"
3216 "\xe0\xe1\xe2\x00\xe4\x00\x00\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3217 "\x00\xf1\xf2\xf3\xf4\x00\xf6\xf7\x00\xf9\xfa\xfb\xfc\x00\x00\x00"
3218 };
3219
3220 static unsigned short const xmlunicodetable_ISO8859_4 [128] = {
3221 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3222 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3223 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3224 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3225 0x00a0, 0x0104, 0x0138, 0x0156, 0x00a4, 0x0128, 0x013b, 0x00a7,
3226 0x00a8, 0x0160, 0x0112, 0x0122, 0x0166, 0x00ad, 0x017d, 0x00af,
3227 0x00b0, 0x0105, 0x02db, 0x0157, 0x00b4, 0x0129, 0x013c, 0x02c7,
3228 0x00b8, 0x0161, 0x0113, 0x0123, 0x0167, 0x014a, 0x017e, 0x014b,
3229 0x0100, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x012e,
3230 0x010c, 0x00c9, 0x0118, 0x00cb, 0x0116, 0x00cd, 0x00ce, 0x012a,
3231 0x0110, 0x0145, 0x014c, 0x0136, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
3232 0x00d8, 0x0172, 0x00da, 0x00db, 0x00dc, 0x0168, 0x016a, 0x00df,
3233 0x0101, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x012f,
3234 0x010d, 0x00e9, 0x0119, 0x00eb, 0x0117, 0x00ed, 0x00ee, 0x012b,
3235 0x0111, 0x0146, 0x014d, 0x0137, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
3236 0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x0169, 0x016b, 0x02d9,
3237 };
3238
3239 static const unsigned char xmltranscodetable_ISO8859_4 [48 + 6 * 64] = {
3240 "\x00\x00\x01\x05\x02\x03\x00\x00\x00\x00\x00\x04\x00\x00\x00\x00"
3241 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3242 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3243 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3244 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3245 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3246 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3247 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3248 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3249 "\xa0\x00\x00\x00\xa4\x00\x00\xa7\xa8\x00\x00\x00\x00\xad\x00\xaf"
3250 "\xb0\x00\x00\x00\xb4\x00\x00\x00\xb8\x00\x00\x00\x00\x00\x00\x00"
3251 "\xc0\xe0\x00\x00\xa1\xb1\x00\x00\x00\x00\x00\x00\xc8\xe8\x00\x00"
3252 "\xd0\xf0\xaa\xba\x00\x00\xcc\xec\xca\xea\x00\x00\x00\x00\x00\x00"
3253 "\x00\x00\xab\xbb\x00\x00\x00\x00\xa5\xb5\xcf\xef\x00\x00\xc7\xe7"
3254 "\x00\x00\x00\x00\x00\x00\xd3\xf3\xa2\x00\x00\xa6\xb6\x00\x00\x00"
3255 "\x00\x00\x00\x00\x00\xd1\xf1\x00\x00\x00\xbd\xbf\xd2\xf2\x00\x00"
3256 "\x00\x00\x00\x00\x00\x00\xa3\xb3\x00\x00\x00\x00\x00\x00\x00\x00"
3257 "\xa9\xb9\x00\x00\x00\x00\xac\xbc\xdd\xfd\xde\xfe\x00\x00\x00\x00"
3258 "\x00\x00\xd9\xf9\x00\x00\x00\x00\x00\x00\x00\x00\x00\xae\xbe\x00"
3259 "\x00\x00\x00\x00\x00\x00\x00\xb7\x00\x00\x00\x00\x00\x00\x00\x00"
3260 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\xff\x00\xb2\x00\x00\x00\x00"
3261 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3262 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3263 "\x00\xc1\xc2\xc3\xc4\xc5\xc6\x00\x00\xc9\x00\xcb\x00\xcd\xce\x00"
3264 "\x00\x00\x00\x00\xd4\xd5\xd6\xd7\xd8\x00\xda\xdb\xdc\x00\x00\xdf"
3265 "\x00\xe1\xe2\xe3\xe4\xe5\xe6\x00\x00\xe9\x00\xeb\x00\xed\xee\x00"
3266 "\x00\x00\x00\x00\xf4\xf5\xf6\xf7\xf8\x00\xfa\xfb\xfc\x00\x00\x00"
3267 };
3268
3269 static unsigned short const xmlunicodetable_ISO8859_5 [128] = {
3270 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3271 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3272 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3273 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3274 0x00a0, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0406, 0x0407,
3275 0x0408, 0x0409, 0x040a, 0x040b, 0x040c, 0x00ad, 0x040e, 0x040f,
3276 0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
3277 0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
3278 0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
3279 0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
3280 0x0430, 0x0431, 0x0432, 0x0433, 0x0434, 0x0435, 0x0436, 0x0437,
3281 0x0438, 0x0439, 0x043a, 0x043b, 0x043c, 0x043d, 0x043e, 0x043f,
3282 0x0440, 0x0441, 0x0442, 0x0443, 0x0444, 0x0445, 0x0446, 0x0447,
3283 0x0448, 0x0449, 0x044a, 0x044b, 0x044c, 0x044d, 0x044e, 0x044f,
3284 0x2116, 0x0451, 0x0452, 0x0453, 0x0454, 0x0455, 0x0456, 0x0457,
3285 0x0458, 0x0459, 0x045a, 0x045b, 0x045c, 0x00a7, 0x045e, 0x045f,
3286 };
3287
3288 static const unsigned char xmltranscodetable_ISO8859_5 [48 + 6 * 64] = {
3289 "\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3290 "\x02\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3291 "\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3292 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3293 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3294 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3295 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3296 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3297 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3298 "\xa0\x00\x00\x00\x00\x00\x00\xfd\x00\x00\x00\x00\x00\xad\x00\x00"
3299 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3300 "\x00\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\x00\xae\xaf"
3301 "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
3302 "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3303 "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
3304 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3305 "\x00\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\x00\xfe\xff"
3306 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3307 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3308 "\x00\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3309 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3310 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3311 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3312 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3313 "\x00\x00\x00\x00\x00\x00\xf0\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3314 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3315 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3316 };
3317
3318 static unsigned short const xmlunicodetable_ISO8859_6 [128] = {
3319 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3320 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3321 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3322 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3323 0x00a0, 0x0000, 0x0000, 0x0000, 0x00a4, 0x0000, 0x0000, 0x0000,
3324 0x0000, 0x0000, 0x0000, 0x0000, 0x060c, 0x00ad, 0x0000, 0x0000,
3325 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3326 0x0000, 0x0000, 0x0000, 0x061b, 0x0000, 0x0000, 0x0000, 0x061f,
3327 0x0000, 0x0621, 0x0622, 0x0623, 0x0624, 0x0625, 0x0626, 0x0627,
3328 0x0628, 0x0629, 0x062a, 0x062b, 0x062c, 0x062d, 0x062e, 0x062f,
3329 0x0630, 0x0631, 0x0632, 0x0633, 0x0634, 0x0635, 0x0636, 0x0637,
3330 0x0638, 0x0639, 0x063a, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3331 0x0640, 0x0641, 0x0642, 0x0643, 0x0644, 0x0645, 0x0646, 0x0647,
3332 0x0648, 0x0649, 0x064a, 0x064b, 0x064c, 0x064d, 0x064e, 0x064f,
3333 0x0650, 0x0651, 0x0652, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3334 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3335 };
3336
3337 static const unsigned char xmltranscodetable_ISO8859_6 [48 + 5 * 64] = {
3338 "\x02\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3339 "\x00\x00\x00\x00\x00\x00\x00\x00\x03\x04\x00\x00\x00\x00\x00\x00"
3340 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3341 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3342 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3343 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3344 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3345 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3346 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3347 "\xa0\x00\x00\x00\xa4\x00\x00\x00\x00\x00\x00\x00\x00\xad\x00\x00"
3348 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3349 "\xff\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3350 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3351 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3352 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3353 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xac\x00\x00\x00"
3354 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xbb\x00\x00\x00\xbf"
3355 "\x00\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3356 "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\x00\x00\x00\x00\x00"
3357 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3358 "\xf0\xf1\xf2\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3359 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3360 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3361 };
3362
3363 static unsigned short const xmlunicodetable_ISO8859_7 [128] = {
3364 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3365 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3366 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3367 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3368 0x00a0, 0x2018, 0x2019, 0x00a3, 0x0000, 0x0000, 0x00a6, 0x00a7,
3369 0x00a8, 0x00a9, 0x0000, 0x00ab, 0x00ac, 0x00ad, 0x0000, 0x2015,
3370 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x0384, 0x0385, 0x0386, 0x00b7,
3371 0x0388, 0x0389, 0x038a, 0x00bb, 0x038c, 0x00bd, 0x038e, 0x038f,
3372 0x0390, 0x0391, 0x0392, 0x0393, 0x0394, 0x0395, 0x0396, 0x0397,
3373 0x0398, 0x0399, 0x039a, 0x039b, 0x039c, 0x039d, 0x039e, 0x039f,
3374 0x03a0, 0x03a1, 0x0000, 0x03a3, 0x03a4, 0x03a5, 0x03a6, 0x03a7,
3375 0x03a8, 0x03a9, 0x03aa, 0x03ab, 0x03ac, 0x03ad, 0x03ae, 0x03af,
3376 0x03b0, 0x03b1, 0x03b2, 0x03b3, 0x03b4, 0x03b5, 0x03b6, 0x03b7,
3377 0x03b8, 0x03b9, 0x03ba, 0x03bb, 0x03bc, 0x03bd, 0x03be, 0x03bf,
3378 0x03c0, 0x03c1, 0x03c2, 0x03c3, 0x03c4, 0x03c5, 0x03c6, 0x03c7,
3379 0x03c8, 0x03c9, 0x03ca, 0x03cb, 0x03cc, 0x03cd, 0x03ce, 0x0000,
3380 };
3381
3382 static const unsigned char xmltranscodetable_ISO8859_7 [48 + 7 * 64] = {
3383 "\x04\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x05\x06"
3384 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3385 "\x00\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3386 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3387 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3388 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3389 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3390 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3391 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3392 "\xa0\x00\x00\xa3\x00\x00\xa6\xa7\xa8\xa9\x00\xab\xac\xad\x00\x00"
3393 "\xb0\xb1\xb2\xb3\x00\x00\x00\xb7\x00\x00\x00\xbb\x00\xbd\x00\x00"
3394 "\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3395 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3396 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3397 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3398 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3399 "\x00\x00\x00\x00\x00\xaf\x00\x00\xa1\xa2\x00\x00\x00\x00\x00\x00"
3400 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3401 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3402 "\xff\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3403 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3404 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3405 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3406 "\x00\x00\x00\x00\xb4\xb5\xb6\x00\xb8\xb9\xba\x00\xbc\x00\xbe\xbf"
3407 "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3408 "\xd0\xd1\x00\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
3409 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3410 "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\x00"
3411 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3412 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3413 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3414 };
3415
3416 static unsigned short const xmlunicodetable_ISO8859_8 [128] = {
3417 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3418 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3419 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3420 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3421 0x00a0, 0x0000, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
3422 0x00a8, 0x00a9, 0x00d7, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
3423 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
3424 0x00b8, 0x00b9, 0x00f7, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x0000,
3425 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3426 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3427 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
3428 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x2017,
3429 0x05d0, 0x05d1, 0x05d2, 0x05d3, 0x05d4, 0x05d5, 0x05d6, 0x05d7,
3430 0x05d8, 0x05d9, 0x05da, 0x05db, 0x05dc, 0x05dd, 0x05de, 0x05df,
3431 0x05e0, 0x05e1, 0x05e2, 0x05e3, 0x05e4, 0x05e5, 0x05e6, 0x05e7,
3432 0x05e8, 0x05e9, 0x05ea, 0x0000, 0x0000, 0x200e, 0x200f, 0x0000,
3433 };
3434
3435 static const unsigned char xmltranscodetable_ISO8859_8 [48 + 7 * 64] = {
3436 "\x02\x00\x01\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3437 "\x00\x00\x00\x00\x00\x00\x00\x06\x00\x00\x00\x00\x00\x00\x00\x00"
3438 "\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3439 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3440 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3441 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3442 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3443 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3444 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3445 "\xa0\x00\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\x00\xab\xac\xad\xae\xaf"
3446 "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\x00\xbb\xbc\xbd\xbe\x00"
3447 "\xff\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3448 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3449 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3450 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3451 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3452 "\x00\x00\x00\x00\x00\x00\x00\xaa\x00\x00\x00\x00\x00\x00\x00\x00"
3453 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3454 "\x00\x00\x00\x00\x00\x00\x00\xba\x00\x00\x00\x00\x00\x00\x00\x00"
3455 "\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3456 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3457 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3458 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3459 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xfd\xfe"
3460 "\x00\x00\x00\x00\x00\x00\x00\xdf\x00\x00\x00\x00\x00\x00\x00\x00"
3461 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3462 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3463 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3464 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3465 "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\x00\x00\x00\x00\x00"
3466 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3467 };
3468
3469 static unsigned short const xmlunicodetable_ISO8859_9 [128] = {
3470 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3471 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3472 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3473 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3474 0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
3475 0x00a8, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
3476 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
3477 0x00b8, 0x00b9, 0x00ba, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00bf,
3478 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
3479 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
3480 0x011e, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
3481 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0130, 0x015e, 0x00df,
3482 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
3483 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
3484 0x011f, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
3485 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0131, 0x015f, 0x00ff,
3486 };
3487
3488 static const unsigned char xmltranscodetable_ISO8859_9 [48 + 5 * 64] = {
3489 "\x00\x00\x01\x02\x03\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3490 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3491 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3492 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3493 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3494 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3495 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3496 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3497 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3498 "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
3499 "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
3500 "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3501 "\x00\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\x00\x00\xdf"
3502 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3503 "\x00\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\x00\x00\xff"
3504 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3505 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xd0\xf0"
3506 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3507 "\xdd\xfd\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3508 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3509 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xde\xfe"
3510 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3511 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3512 };
3513
3514 static unsigned short const xmlunicodetable_ISO8859_10 [128] = {
3515 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3516 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3517 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3518 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3519 0x00a0, 0x0104, 0x0112, 0x0122, 0x012a, 0x0128, 0x0136, 0x00a7,
3520 0x013b, 0x0110, 0x0160, 0x0166, 0x017d, 0x00ad, 0x016a, 0x014a,
3521 0x00b0, 0x0105, 0x0113, 0x0123, 0x012b, 0x0129, 0x0137, 0x00b7,
3522 0x013c, 0x0111, 0x0161, 0x0167, 0x017e, 0x2015, 0x016b, 0x014b,
3523 0x0100, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x012e,
3524 0x010c, 0x00c9, 0x0118, 0x00cb, 0x0116, 0x00cd, 0x00ce, 0x00cf,
3525 0x00d0, 0x0145, 0x014c, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x0168,
3526 0x00d8, 0x0172, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
3527 0x0101, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x012f,
3528 0x010d, 0x00e9, 0x0119, 0x00eb, 0x0117, 0x00ed, 0x00ee, 0x00ef,
3529 0x00f0, 0x0146, 0x014d, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x0169,
3530 0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x0138,
3531 };
3532
3533 static const unsigned char xmltranscodetable_ISO8859_10 [48 + 7 * 64] = {
3534 "\x00\x00\x01\x06\x02\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3535 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3536 "\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3537 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3538 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3539 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3540 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3541 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3542 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3543 "\xa0\x00\x00\x00\x00\x00\x00\xa7\x00\x00\x00\x00\x00\xad\x00\x00"
3544 "\xb0\x00\x00\x00\x00\x00\x00\xb7\x00\x00\x00\x00\x00\x00\x00\x00"
3545 "\xc0\xe0\x00\x00\xa1\xb1\x00\x00\x00\x00\x00\x00\xc8\xe8\x00\x00"
3546 "\xa9\xb9\xa2\xb2\x00\x00\xcc\xec\xca\xea\x00\x00\x00\x00\x00\x00"
3547 "\x00\x00\xa3\xb3\x00\x00\x00\x00\xa5\xb5\xa4\xb4\x00\x00\xc7\xe7"
3548 "\x00\x00\x00\x00\x00\x00\xa6\xb6\xff\x00\x00\xa8\xb8\x00\x00\x00"
3549 "\x00\x00\x00\x00\x00\xd1\xf1\x00\x00\x00\xaf\xbf\xd2\xf2\x00\x00"
3550 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3551 "\xaa\xba\x00\x00\x00\x00\xab\xbb\xd7\xf7\xae\xbe\x00\x00\x00\x00"
3552 "\x00\x00\xd9\xf9\x00\x00\x00\x00\x00\x00\x00\x00\x00\xac\xbc\x00"
3553 "\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3554 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3555 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3556 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3557 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3558 "\x00\x00\x00\x00\x00\xbd\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3559 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3560 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3561 "\x00\xc1\xc2\xc3\xc4\xc5\xc6\x00\x00\xc9\x00\xcb\x00\xcd\xce\xcf"
3562 "\xd0\x00\x00\xd3\xd4\xd5\xd6\x00\xd8\x00\xda\xdb\xdc\xdd\xde\xdf"
3563 "\x00\xe1\xe2\xe3\xe4\xe5\xe6\x00\x00\xe9\x00\xeb\x00\xed\xee\xef"
3564 "\xf0\x00\x00\xf3\xf4\xf5\xf6\x00\xf8\x00\xfa\xfb\xfc\xfd\xfe\x00"
3565 };
3566
3567 static unsigned short const xmlunicodetable_ISO8859_11 [128] = {
3568 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3569 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3570 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3571 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3572 0x00a0, 0x0e01, 0x0e02, 0x0e03, 0x0e04, 0x0e05, 0x0e06, 0x0e07,
3573 0x0e08, 0x0e09, 0x0e0a, 0x0e0b, 0x0e0c, 0x0e0d, 0x0e0e, 0x0e0f,
3574 0x0e10, 0x0e11, 0x0e12, 0x0e13, 0x0e14, 0x0e15, 0x0e16, 0x0e17,
3575 0x0e18, 0x0e19, 0x0e1a, 0x0e1b, 0x0e1c, 0x0e1d, 0x0e1e, 0x0e1f,
3576 0x0e20, 0x0e21, 0x0e22, 0x0e23, 0x0e24, 0x0e25, 0x0e26, 0x0e27,
3577 0x0e28, 0x0e29, 0x0e2a, 0x0e2b, 0x0e2c, 0x0e2d, 0x0e2e, 0x0e2f,
3578 0x0e30, 0x0e31, 0x0e32, 0x0e33, 0x0e34, 0x0e35, 0x0e36, 0x0e37,
3579 0x0e38, 0x0e39, 0x0e3a, 0x0000, 0x0000, 0x0000, 0x0000, 0x0e3f,
3580 0x0e40, 0x0e41, 0x0e42, 0x0e43, 0x0e44, 0x0e45, 0x0e46, 0x0e47,
3581 0x0e48, 0x0e49, 0x0e4a, 0x0e4b, 0x0e4c, 0x0e4d, 0x0e4e, 0x0e4f,
3582 0x0e50, 0x0e51, 0x0e52, 0x0e53, 0x0e54, 0x0e55, 0x0e56, 0x0e57,
3583 0x0e58, 0x0e59, 0x0e5a, 0x0e5b, 0x0000, 0x0000, 0x0000, 0x0000,
3584 };
3585
3586 static const unsigned char xmltranscodetable_ISO8859_11 [48 + 6 * 64] = {
3587 "\x04\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3588 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3589 "\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3590 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3591 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3592 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3593 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3594 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3595 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3596 "\xa0\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3597 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3598 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3599 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3600 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3601 "\x00\x00\x00\x00\x00\x00\x00\x00\x03\x05\x00\x00\x00\x00\x00\x00"
3602 "\x00\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
3603 "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
3604 "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3605 "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\x00\x00\x00\x00\xdf"
3606 "\xff\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3607 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3608 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3609 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3610 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3611 "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\x00\x00\x00\x00"
3612 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3613 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3614 };
3615
3616 static unsigned short const xmlunicodetable_ISO8859_13 [128] = {
3617 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3618 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3619 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3620 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3621 0x00a0, 0x201d, 0x00a2, 0x00a3, 0x00a4, 0x201e, 0x00a6, 0x00a7,
3622 0x00d8, 0x00a9, 0x0156, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00c6,
3623 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x201c, 0x00b5, 0x00b6, 0x00b7,
3624 0x00f8, 0x00b9, 0x0157, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00e6,
3625 0x0104, 0x012e, 0x0100, 0x0106, 0x00c4, 0x00c5, 0x0118, 0x0112,
3626 0x010c, 0x00c9, 0x0179, 0x0116, 0x0122, 0x0136, 0x012a, 0x013b,
3627 0x0160, 0x0143, 0x0145, 0x00d3, 0x014c, 0x00d5, 0x00d6, 0x00d7,
3628 0x0172, 0x0141, 0x015a, 0x016a, 0x00dc, 0x017b, 0x017d, 0x00df,
3629 0x0105, 0x012f, 0x0101, 0x0107, 0x00e4, 0x00e5, 0x0119, 0x0113,
3630 0x010d, 0x00e9, 0x017a, 0x0117, 0x0123, 0x0137, 0x012b, 0x013c,
3631 0x0161, 0x0144, 0x0146, 0x00f3, 0x014d, 0x00f5, 0x00f6, 0x00f7,
3632 0x0173, 0x0142, 0x015b, 0x016b, 0x00fc, 0x017c, 0x017e, 0x2019,
3633 };
3634
3635 static const unsigned char xmltranscodetable_ISO8859_13 [48 + 7 * 64] = {
3636 "\x00\x00\x01\x04\x06\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3637 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3638 "\x00\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3639 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3640 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3641 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3642 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3643 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3644 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3645 "\xa0\x00\xa2\xa3\xa4\x00\xa6\xa7\x00\xa9\x00\xab\xac\xad\xae\x00"
3646 "\xb0\xb1\xb2\xb3\x00\xb5\xb6\xb7\x00\xb9\x00\xbb\xbc\xbd\xbe\x00"
3647 "\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3648 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3649 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3650 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3651 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3652 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\xff\x00\x00\xb4\xa1\xa5\x00"
3653 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3654 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3655 "\x00\x00\x00\x00\xc4\xc5\xaf\x00\x00\xc9\x00\x00\x00\x00\x00\x00"
3656 "\x00\x00\x00\xd3\x00\xd5\xd6\xd7\xa8\x00\x00\x00\xdc\x00\x00\xdf"
3657 "\x00\x00\x00\x00\xe4\xe5\xbf\x00\x00\xe9\x00\x00\x00\x00\x00\x00"
3658 "\x00\x00\x00\xf3\x00\xf5\xf6\xf7\xb8\x00\x00\x00\xfc\x00\x00\x00"
3659 "\x00\xd9\xf9\xd1\xf1\xd2\xf2\x00\x00\x00\x00\x00\xd4\xf4\x00\x00"
3660 "\x00\x00\x00\x00\x00\x00\xaa\xba\x00\x00\xda\xfa\x00\x00\x00\x00"
3661 "\xd0\xf0\x00\x00\x00\x00\x00\x00\x00\x00\xdb\xfb\x00\x00\x00\x00"
3662 "\x00\x00\xd8\xf8\x00\x00\x00\x00\x00\xca\xea\xdd\xfd\xde\xfe\x00"
3663 "\xc2\xe2\x00\x00\xc0\xe0\xc3\xe3\x00\x00\x00\x00\xc8\xe8\x00\x00"
3664 "\x00\x00\xc7\xe7\x00\x00\xcb\xeb\xc6\xe6\x00\x00\x00\x00\x00\x00"
3665 "\x00\x00\xcc\xec\x00\x00\x00\x00\x00\x00\xce\xee\x00\x00\xc1\xe1"
3666 "\x00\x00\x00\x00\x00\x00\xcd\xed\x00\x00\x00\xcf\xef\x00\x00\x00"
3667 };
3668
3669 static unsigned short const xmlunicodetable_ISO8859_14 [128] = {
3670 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3671 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3672 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3673 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3674 0x00a0, 0x1e02, 0x1e03, 0x00a3, 0x010a, 0x010b, 0x1e0a, 0x00a7,
3675 0x1e80, 0x00a9, 0x1e82, 0x1e0b, 0x1ef2, 0x00ad, 0x00ae, 0x0178,
3676 0x1e1e, 0x1e1f, 0x0120, 0x0121, 0x1e40, 0x1e41, 0x00b6, 0x1e56,
3677 0x1e81, 0x1e57, 0x1e83, 0x1e60, 0x1ef3, 0x1e84, 0x1e85, 0x1e61,
3678 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
3679 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
3680 0x0174, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x1e6a,
3681 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x0176, 0x00df,
3682 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
3683 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
3684 0x0175, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x1e6b,
3685 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x0177, 0x00ff,
3686 };
3687
3688 static const unsigned char xmltranscodetable_ISO8859_14 [48 + 10 * 64] = {
3689 "\x00\x00\x01\x09\x04\x07\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3690 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3691 "\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3692 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3693 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3694 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3695 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3696 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3697 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3698 "\xa0\x00\x00\xa3\x00\x00\x00\xa7\x00\xa9\x00\x00\x00\xad\xae\x00"
3699 "\x00\x00\x00\x00\x00\x00\xb6\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3700 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3701 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3702 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3703 "\x00\x00\x00\x00\x00\x00\x00\x00\x03\x08\x05\x06\x00\x00\x00\x00"
3704 "\x00\x00\xa1\xa2\x00\x00\x00\x00\x00\x00\xa6\xab\x00\x00\x00\x00"
3705 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xb0\xb1"
3706 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3707 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3708 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\xa5\x00\x00\x00\x00"
3709 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3710 "\xb2\xb3\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3711 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3712 "\xa8\xb8\xaa\xba\xbd\xbe\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3713 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3714 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3715 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3716 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3717 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3718 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3719 "\x00\x00\xac\xbc\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3720 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3721 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3722 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3723 "\x00\x00\x00\x00\xd0\xf0\xde\xfe\xaf\x00\x00\x00\x00\x00\x00\x00"
3724 "\xb4\xb5\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3725 "\x00\x00\x00\x00\x00\x00\xb7\xb9\x00\x00\x00\x00\x00\x00\x00\x00"
3726 "\xbb\xbf\x00\x00\x00\x00\x00\x00\x00\x00\xd7\xf7\x00\x00\x00\x00"
3727 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3728 "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3729 "\x00\xd1\xd2\xd3\xd4\xd5\xd6\x00\xd8\xd9\xda\xdb\xdc\xdd\x00\xdf"
3730 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3731 "\x00\xf1\xf2\xf3\xf4\xf5\xf6\x00\xf8\xf9\xfa\xfb\xfc\xfd\x00\xff"
3732 };
3733
3734 static unsigned short const xmlunicodetable_ISO8859_15 [128] = {
3735 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3736 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3737 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3738 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3739 0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x20ac, 0x00a5, 0x0160, 0x00a7,
3740 0x0161, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
3741 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x017d, 0x00b5, 0x00b6, 0x00b7,
3742 0x017e, 0x00b9, 0x00ba, 0x00bb, 0x0152, 0x0153, 0x0178, 0x00bf,
3743 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
3744 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
3745 0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
3746 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
3747 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
3748 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
3749 0x00f0, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
3750 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x00ff,
3751 };
3752
3753 static const unsigned char xmltranscodetable_ISO8859_15 [48 + 6 * 64] = {
3754 "\x00\x00\x01\x05\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3755 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3756 "\x00\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3757 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3758 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3759 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3760 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3761 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3762 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3763 "\xa0\xa1\xa2\xa3\x00\xa5\x00\xa7\x00\xa9\xaa\xab\xac\xad\xae\xaf"
3764 "\xb0\xb1\xb2\xb3\x00\xb5\xb6\xb7\x00\xb9\xba\xbb\x00\x00\x00\xbf"
3765 "\x00\x00\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3766 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3767 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3768 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3769 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3770 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3771 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x00\x00\x00"
3772 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3773 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3774 "\x00\x00\xbc\xbd\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3775 "\xa6\xa8\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3776 "\x00\x00\x00\x00\x00\x00\x00\x00\xbe\x00\x00\x00\x00\xb4\xb8\x00"
3777 "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3778 "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
3779 "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3780 "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
3781 };
3782
3783 static unsigned short const xmlunicodetable_ISO8859_16 [128] = {
3784 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
3785 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
3786 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
3787 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
3788 0x00a0, 0x0104, 0x0105, 0x0141, 0x20ac, 0x201e, 0x0160, 0x00a7,
3789 0x0161, 0x00a9, 0x0218, 0x00ab, 0x0179, 0x00ad, 0x017a, 0x017b,
3790 0x00b0, 0x00b1, 0x010c, 0x0142, 0x017d, 0x201d, 0x00b6, 0x00b7,
3791 0x017e, 0x010d, 0x0219, 0x00bb, 0x0152, 0x0153, 0x0178, 0x017c,
3792 0x00c0, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0106, 0x00c6, 0x00c7,
3793 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
3794 0x0110, 0x0143, 0x00d2, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x015a,
3795 0x0170, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0118, 0x021a, 0x00df,
3796 0x00e0, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x0107, 0x00e6, 0x00e7,
3797 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
3798 0x0111, 0x0144, 0x00f2, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x015b,
3799 0x0171, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0119, 0x021b, 0x00ff,
3800 };
3801
3802 static const unsigned char xmltranscodetable_ISO8859_16 [48 + 9 * 64] = {
3803 "\x00\x00\x01\x08\x02\x03\x00\x00\x07\x00\x00\x00\x00\x00\x00\x00"
3804 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3805 "\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3806 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3807 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3808 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3809 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3810 "\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
3811 "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
3812 "\xa0\x00\x00\x00\x00\x00\x00\xa7\x00\xa9\x00\xab\x00\xad\x00\x00"
3813 "\xb0\xb1\x00\x00\x00\x00\xb6\xb7\x00\x00\x00\xbb\x00\x00\x00\x00"
3814 "\x00\x00\xc3\xe3\xa1\xa2\xc5\xe5\x00\x00\x00\x00\xb2\xb9\x00\x00"
3815 "\xd0\xf0\x00\x00\x00\x00\x00\x00\xdd\xfd\x00\x00\x00\x00\x00\x00"
3816 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3817 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3818 "\x00\xa3\xb3\xd1\xf1\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3819 "\xd5\xf5\xbc\xbd\x00\x00\x00\x00\x00\x00\xd7\xf7\x00\x00\x00\x00"
3820 "\xa6\xa8\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3821 "\xd8\xf8\x00\x00\x00\x00\x00\x00\xbe\xac\xae\xaf\xbf\xb4\xb8\x00"
3822 "\x06\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3823 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3824 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3825 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3826 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3827 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3828 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x00\x00\x00"
3829 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3830 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3831 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xb5\xa5\x00"
3832 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3833 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3834 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3835 "\x00\x00\x00\x00\x00\x00\x00\x00\xaa\xba\xde\xfe\x00\x00\x00\x00"
3836 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3837 "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
3838 "\xc0\xc1\xc2\x00\xc4\x00\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
3839 "\x00\x00\xd2\xd3\xd4\x00\xd6\x00\x00\xd9\xda\xdb\xdc\x00\x00\xdf"
3840 "\xe0\xe1\xe2\x00\xe4\x00\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef"
3841 "\x00\x00\xf2\xf3\xf4\x00\xf6\x00\x00\xf9\xfa\xfb\xfc\x00\x00\xff"
3842 };
3843
3844
3845 /*
3846 * auto-generated functions for ISO-8859-2 .. ISO-8859-16
3847 */
3848
ISO8859_2ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3849 static int ISO8859_2ToUTF8 (unsigned char* out, int *outlen,
3850 const unsigned char* in, int *inlen) {
3851 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_2);
3852 }
UTF8ToISO8859_2(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3853 static int UTF8ToISO8859_2 (unsigned char* out, int *outlen,
3854 const unsigned char* in, int *inlen) {
3855 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_2);
3856 }
3857
ISO8859_3ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3858 static int ISO8859_3ToUTF8 (unsigned char* out, int *outlen,
3859 const unsigned char* in, int *inlen) {
3860 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_3);
3861 }
UTF8ToISO8859_3(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3862 static int UTF8ToISO8859_3 (unsigned char* out, int *outlen,
3863 const unsigned char* in, int *inlen) {
3864 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_3);
3865 }
3866
ISO8859_4ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3867 static int ISO8859_4ToUTF8 (unsigned char* out, int *outlen,
3868 const unsigned char* in, int *inlen) {
3869 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_4);
3870 }
UTF8ToISO8859_4(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3871 static int UTF8ToISO8859_4 (unsigned char* out, int *outlen,
3872 const unsigned char* in, int *inlen) {
3873 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_4);
3874 }
3875
ISO8859_5ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3876 static int ISO8859_5ToUTF8 (unsigned char* out, int *outlen,
3877 const unsigned char* in, int *inlen) {
3878 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_5);
3879 }
UTF8ToISO8859_5(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3880 static int UTF8ToISO8859_5 (unsigned char* out, int *outlen,
3881 const unsigned char* in, int *inlen) {
3882 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_5);
3883 }
3884
ISO8859_6ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3885 static int ISO8859_6ToUTF8 (unsigned char* out, int *outlen,
3886 const unsigned char* in, int *inlen) {
3887 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_6);
3888 }
UTF8ToISO8859_6(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3889 static int UTF8ToISO8859_6 (unsigned char* out, int *outlen,
3890 const unsigned char* in, int *inlen) {
3891 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_6);
3892 }
3893
ISO8859_7ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3894 static int ISO8859_7ToUTF8 (unsigned char* out, int *outlen,
3895 const unsigned char* in, int *inlen) {
3896 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_7);
3897 }
UTF8ToISO8859_7(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3898 static int UTF8ToISO8859_7 (unsigned char* out, int *outlen,
3899 const unsigned char* in, int *inlen) {
3900 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_7);
3901 }
3902
ISO8859_8ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3903 static int ISO8859_8ToUTF8 (unsigned char* out, int *outlen,
3904 const unsigned char* in, int *inlen) {
3905 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_8);
3906 }
UTF8ToISO8859_8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3907 static int UTF8ToISO8859_8 (unsigned char* out, int *outlen,
3908 const unsigned char* in, int *inlen) {
3909 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_8);
3910 }
3911
ISO8859_9ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3912 static int ISO8859_9ToUTF8 (unsigned char* out, int *outlen,
3913 const unsigned char* in, int *inlen) {
3914 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_9);
3915 }
UTF8ToISO8859_9(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3916 static int UTF8ToISO8859_9 (unsigned char* out, int *outlen,
3917 const unsigned char* in, int *inlen) {
3918 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_9);
3919 }
3920
ISO8859_10ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3921 static int ISO8859_10ToUTF8 (unsigned char* out, int *outlen,
3922 const unsigned char* in, int *inlen) {
3923 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_10);
3924 }
UTF8ToISO8859_10(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3925 static int UTF8ToISO8859_10 (unsigned char* out, int *outlen,
3926 const unsigned char* in, int *inlen) {
3927 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_10);
3928 }
3929
ISO8859_11ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3930 static int ISO8859_11ToUTF8 (unsigned char* out, int *outlen,
3931 const unsigned char* in, int *inlen) {
3932 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_11);
3933 }
UTF8ToISO8859_11(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3934 static int UTF8ToISO8859_11 (unsigned char* out, int *outlen,
3935 const unsigned char* in, int *inlen) {
3936 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_11);
3937 }
3938
ISO8859_13ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3939 static int ISO8859_13ToUTF8 (unsigned char* out, int *outlen,
3940 const unsigned char* in, int *inlen) {
3941 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_13);
3942 }
UTF8ToISO8859_13(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3943 static int UTF8ToISO8859_13 (unsigned char* out, int *outlen,
3944 const unsigned char* in, int *inlen) {
3945 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_13);
3946 }
3947
ISO8859_14ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3948 static int ISO8859_14ToUTF8 (unsigned char* out, int *outlen,
3949 const unsigned char* in, int *inlen) {
3950 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_14);
3951 }
UTF8ToISO8859_14(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3952 static int UTF8ToISO8859_14 (unsigned char* out, int *outlen,
3953 const unsigned char* in, int *inlen) {
3954 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_14);
3955 }
3956
ISO8859_15ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3957 static int ISO8859_15ToUTF8 (unsigned char* out, int *outlen,
3958 const unsigned char* in, int *inlen) {
3959 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_15);
3960 }
UTF8ToISO8859_15(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3961 static int UTF8ToISO8859_15 (unsigned char* out, int *outlen,
3962 const unsigned char* in, int *inlen) {
3963 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_15);
3964 }
3965
ISO8859_16ToUTF8(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3966 static int ISO8859_16ToUTF8 (unsigned char* out, int *outlen,
3967 const unsigned char* in, int *inlen) {
3968 return ISO8859xToUTF8 (out, outlen, in, inlen, xmlunicodetable_ISO8859_16);
3969 }
UTF8ToISO8859_16(unsigned char * out,int * outlen,const unsigned char * in,int * inlen)3970 static int UTF8ToISO8859_16 (unsigned char* out, int *outlen,
3971 const unsigned char* in, int *inlen) {
3972 return UTF8ToISO8859x (out, outlen, in, inlen, xmltranscodetable_ISO8859_16);
3973 }
3974
3975 static void
xmlRegisterCharEncodingHandlersISO8859x(void)3976 xmlRegisterCharEncodingHandlersISO8859x (void) {
3977 xmlNewCharEncodingHandler ("ISO-8859-2", ISO8859_2ToUTF8, UTF8ToISO8859_2);
3978 xmlNewCharEncodingHandler ("ISO-8859-3", ISO8859_3ToUTF8, UTF8ToISO8859_3);
3979 xmlNewCharEncodingHandler ("ISO-8859-4", ISO8859_4ToUTF8, UTF8ToISO8859_4);
3980 xmlNewCharEncodingHandler ("ISO-8859-5", ISO8859_5ToUTF8, UTF8ToISO8859_5);
3981 xmlNewCharEncodingHandler ("ISO-8859-6", ISO8859_6ToUTF8, UTF8ToISO8859_6);
3982 xmlNewCharEncodingHandler ("ISO-8859-7", ISO8859_7ToUTF8, UTF8ToISO8859_7);
3983 xmlNewCharEncodingHandler ("ISO-8859-8", ISO8859_8ToUTF8, UTF8ToISO8859_8);
3984 xmlNewCharEncodingHandler ("ISO-8859-9", ISO8859_9ToUTF8, UTF8ToISO8859_9);
3985 xmlNewCharEncodingHandler ("ISO-8859-10", ISO8859_10ToUTF8, UTF8ToISO8859_10);
3986 xmlNewCharEncodingHandler ("ISO-8859-11", ISO8859_11ToUTF8, UTF8ToISO8859_11);
3987 xmlNewCharEncodingHandler ("ISO-8859-13", ISO8859_13ToUTF8, UTF8ToISO8859_13);
3988 xmlNewCharEncodingHandler ("ISO-8859-14", ISO8859_14ToUTF8, UTF8ToISO8859_14);
3989 xmlNewCharEncodingHandler ("ISO-8859-15", ISO8859_15ToUTF8, UTF8ToISO8859_15);
3990 xmlNewCharEncodingHandler ("ISO-8859-16", ISO8859_16ToUTF8, UTF8ToISO8859_16);
3991 }
3992
3993 #endif
3994 #endif
3995
3996