2 * "$Id: transcode.c 6649 2007-07-11 21:46:42Z mike $"
4 * Transcoding support for the Common UNIX Printing System (CUPS).
6 * Copyright 2007-2008 by Apple Inc.
7 * Copyright 1997-2007 by Easy Software Products.
9 * These coded instructions, statements, and computer programs are the
10 * property of Apple Inc. and are protected by Federal copyright
11 * law. Distribution and use rights are outlined in the file "LICENSE.txt"
12 * which should have been included with this file. If this file is
13 * file is missing or damaged, see the license at "http://www.cups.org/".
15 * This file is subject to the Apple OS-Developed Software exception.
19 * _cupsCharmapFlush() - Flush all character set maps out of cache.
20 * _cupsCharmapFree() - Free a character set map.
21 * _cupsCharmapGet() - Get a character set map.
22 * cupsCharsetToUTF8() - Convert legacy character set to UTF-8.
23 * cupsUTF8ToCharset() - Convert UTF-8 to legacy character set.
24 * cupsUTF8ToUTF32() - Convert UTF-8 to UTF-32.
25 * cupsUTF32ToUTF8() - Convert UTF-32 to UTF-8.
26 * compare_wide() - Compare key for wide (VBCS) match.
27 * conv_sbcs_to_utf8() - Convert legacy SBCS to UTF-8.
28 * conv_utf8_to_sbcs() - Convert UTF-8 to legacy SBCS.
29 * conv_utf8_to_vbcs() - Convert UTF-8 to legacy DBCS/VBCS.
30 * conv_vbcs_to_utf8() - Convert legacy DBCS/VBCS to UTF-8.
31 * free_sbcs_charmap() - Free memory used by a single byte character set.
32 * free_vbcs_charmap() - Free memory used by a variable byte character set.
33 * get_charmap() - Lookup or get a character set map (private).
34 * get_charmap_count() - Count lines in a charmap file.
35 * get_sbcs_charmap() - Get SBCS Charmap.
36 * get_vbcs_charmap() - Get DBCS/VBCS Charmap.
40 * Include necessary headers...
56 static pthread_mutex_t map_mutex
= PTHREAD_MUTEX_INITIALIZER
;
57 /* Mutex to control access to maps */
58 #endif /* HAVE_PTHREAD_H */
59 static _cups_cmap_t
*cmap_cache
= NULL
;
60 /* SBCS Charmap Cache */
61 static _cups_vmap_t
*vmap_cache
= NULL
;
62 /* VBCS Charmap Cache */
69 static int compare_wide(const void *k1
, const void *k2
);
70 static int conv_sbcs_to_utf8(cups_utf8_t
*dest
,
71 const cups_sbcs_t
*src
,
73 const cups_encoding_t encoding
);
74 static int conv_utf8_to_sbcs(cups_sbcs_t
*dest
,
75 const cups_utf8_t
*src
,
77 const cups_encoding_t encoding
);
78 static int conv_utf8_to_vbcs(cups_sbcs_t
*dest
,
79 const cups_utf8_t
*src
,
81 const cups_encoding_t encoding
);
82 static int conv_vbcs_to_utf8(cups_utf8_t
*dest
,
83 const cups_sbcs_t
*src
,
85 const cups_encoding_t encoding
);
86 static void free_sbcs_charmap(_cups_cmap_t
*sbcs
);
87 static void free_vbcs_charmap(_cups_vmap_t
*vbcs
);
88 static void *get_charmap(const cups_encoding_t encoding
);
89 static int get_charmap_count(cups_file_t
*fp
);
90 static _cups_cmap_t
*get_sbcs_charmap(const cups_encoding_t encoding
,
91 const char *filename
);
92 static _cups_vmap_t
*get_vbcs_charmap(const cups_encoding_t encoding
,
93 const char *filename
);
97 * '_cupsCharmapFlush()' - Flush all character set maps out of cache.
101 _cupsCharmapFlush(void)
103 _cups_cmap_t
*cmap
, /* Legacy SBCS / Unicode Charset Map */
104 *cnext
; /* Next Legacy SBCS Charset Map */
105 _cups_vmap_t
*vmap
, /* Legacy VBCS / Unicode Charset Map */
106 *vnext
; /* Next Legacy VBCS Charset Map */
109 #ifdef HAVE_PTHREAD_H
110 pthread_mutex_lock(&map_mutex
);
111 #endif /* HAVE_PTHREAD_H */
114 * Loop through SBCS charset map cache, free all memory...
117 for (cmap
= cmap_cache
; cmap
; cmap
= cnext
)
121 free_sbcs_charmap(cmap
);
127 * Loop through DBCS/VBCS charset map cache, free all memory...
130 for (vmap
= vmap_cache
; vmap
; vmap
= vnext
)
134 free_vbcs_charmap(vmap
);
139 #ifdef HAVE_PTHREAD_H
140 pthread_mutex_unlock(&map_mutex
);
141 #endif /* HAVE_PTHREAD_H */
146 * '_cupsCharmapFree()' - Free a character set map.
148 * This does not actually free; use '_cupsCharmapFlush()' for that.
153 const cups_encoding_t encoding
) /* I - Encoding */
155 _cups_cmap_t
*cmap
; /* Legacy SBCS / Unicode Charset Map */
156 _cups_vmap_t
*vmap
; /* Legacy VBCS / Unicode Charset Map */
160 * See if we already have this SBCS charset map loaded...
163 #ifdef HAVE_PTHREAD_H
164 pthread_mutex_lock(&map_mutex
);
165 #endif /* HAVE_PTHREAD_H */
167 for (cmap
= cmap_cache
; cmap
; cmap
= cmap
->next
)
169 if (cmap
->encoding
== encoding
)
178 * See if we already have this DBCS/VBCS charset map loaded...
181 for (vmap
= vmap_cache
; vmap
; vmap
= vmap
->next
)
183 if (vmap
->encoding
== encoding
)
191 #ifdef HAVE_PTHREAD_H
192 pthread_mutex_unlock(&map_mutex
);
193 #endif /* HAVE_PTHREAD_H */
198 * '_cupsCharmapGet()' - Get a character set map.
200 * This code handles single-byte (SBCS), double-byte (DBCS), and
201 * variable-byte (VBCS) character sets _without_ charset escapes...
202 * This code does not handle multiple-byte character sets (MBCS)
203 * (such as ISO-2022-JP) with charset switching via escapes...
206 void * /* O - Charset map pointer */
208 const cups_encoding_t encoding
) /* I - Encoding */
210 void *charmap
; /* Charset map pointer */
213 DEBUG_printf(("_cupsCharmapGet(encoding=%d)\n", encoding
));
216 * Check for valid arguments...
219 if (encoding
< 0 || encoding
>= CUPS_ENCODING_VBCS_END
)
221 DEBUG_puts(" Bad encoding, returning NULL!");
226 * Lookup or get the charset map pointer and return...
229 #ifdef HAVE_PTHREAD_H
230 pthread_mutex_lock(&map_mutex
);
231 #endif /* HAVE_PTHREAD_H */
233 charmap
= get_charmap(encoding
);
235 #ifdef HAVE_PTHREAD_H
236 pthread_mutex_unlock(&map_mutex
);
237 #endif /* HAVE_PTHREAD_H */
244 * 'cupsCharsetToUTF8()' - Convert legacy character set to UTF-8.
246 * This code handles single-byte (SBCS), double-byte (DBCS), and
247 * variable-byte (VBCS) character sets _without_ charset escapes...
248 * This code does not handle multiple-byte character sets (MBCS)
249 * (such as ISO-2022-JP) with charset switching via escapes...
252 int /* O - Count or -1 on error */
254 cups_utf8_t
*dest
, /* O - Target string */
255 const char *src
, /* I - Source string */
256 const int maxout
, /* I - Max output */
257 const cups_encoding_t encoding
) /* I - Encoding */
259 int bytes
; /* Number of bytes converted */
263 * Check for valid arguments...
266 DEBUG_printf(("cupsCharsetToUTF8(dest=%p, src=\"%s\", maxout=%d, encoding=%d)\n",
267 dest
, src
, maxout
, encoding
));
272 if (!dest
|| !src
|| maxout
< 1 || maxout
> CUPS_MAX_USTRING
)
274 DEBUG_puts(" Bad arguments, returning -1");
279 * Handle identity conversions...
282 if (encoding
== CUPS_UTF8
||
283 encoding
< 0 || encoding
>= CUPS_ENCODING_VBCS_END
)
285 strlcpy((char *)dest
, src
, maxout
);
286 return ((int)strlen((char *)dest
));
290 * Handle ISO-8859-1 to UTF-8 directly...
293 if (encoding
== CUPS_ISO8859_1
)
295 int ch
; /* Character from string */
296 cups_utf8_t
*destptr
, /* Pointer into UTF-8 buffer */
297 *destend
; /* End of UTF-8 buffer */
301 destend
= dest
+ maxout
- 2;
303 while (*src
&& destptr
< destend
)
309 *destptr
++ = 0xc0 | (ch
>> 6);
310 *destptr
++ = 0x80 | (ch
& 0x3f);
318 return ((int)(destptr
- dest
));
322 * Convert input legacy charset to UTF-8...
325 #ifdef HAVE_PTHREAD_H
326 pthread_mutex_lock(&map_mutex
);
327 #endif /* HAVE_PTHREAD_H */
329 if (encoding
< CUPS_ENCODING_SBCS_END
)
330 bytes
= conv_sbcs_to_utf8(dest
, (cups_sbcs_t
*)src
, maxout
, encoding
);
332 bytes
= conv_vbcs_to_utf8(dest
, (cups_sbcs_t
*)src
, maxout
, encoding
);
334 #ifdef HAVE_PTHREAD_H
335 pthread_mutex_unlock(&map_mutex
);
336 #endif /* HAVE_PTHREAD_H */
343 * 'cupsUTF8ToCharset()' - Convert UTF-8 to legacy character set.
345 * This code handles single-byte (SBCS), double-byte (DBCS), and
346 * variable-byte (VBCS) character sets _without_ charset escapes...
347 * This code does not handle multiple-byte character sets (MBCS)
348 * (such as ISO-2022-JP) with charset switching via escapes...
351 int /* O - Count or -1 on error */
353 char *dest
, /* O - Target string */
354 const cups_utf8_t
*src
, /* I - Source string */
355 const int maxout
, /* I - Max output */
356 const cups_encoding_t encoding
) /* I - Encoding */
358 int bytes
; /* Number of bytes converted */
362 * Check for valid arguments...
365 if (!dest
|| !src
|| maxout
< 1 || maxout
> CUPS_MAX_USTRING
)
374 * Handle identity conversions...
377 if (encoding
== CUPS_UTF8
||
378 encoding
< 0 || encoding
>= CUPS_ENCODING_VBCS_END
)
380 strlcpy(dest
, (char *)src
, maxout
);
381 return ((int)strlen(dest
));
385 * Handle UTF-8 to ISO-8859-1 directly...
388 if (encoding
== CUPS_ISO8859_1
)
390 int ch
; /* Character from string */
391 char *destptr
, /* Pointer into ISO-8859-1 buffer */
392 *destend
; /* End of ISO-8859-1 buffer */
396 destend
= dest
+ maxout
- 1;
398 while (*src
&& destptr
< destend
)
402 if ((ch
& 0xe0) == 0xc0)
404 ch
= ((ch
& 0x1f) << 6) | (*src
++ & 0x3f);
411 else if ((ch
& 0xf0) == 0xe0 ||
414 else if (!(ch
& 0x80))
420 return ((int)(destptr
- dest
));
424 * Convert input UTF-8 to legacy charset...
427 #ifdef HAVE_PTHREAD_H
428 pthread_mutex_lock(&map_mutex
);
429 #endif /* HAVE_PTHREAD_H */
431 if (encoding
< CUPS_ENCODING_SBCS_END
)
432 bytes
= conv_utf8_to_sbcs((cups_sbcs_t
*)dest
, src
, maxout
, encoding
);
434 bytes
= conv_utf8_to_vbcs((cups_sbcs_t
*)dest
, src
, maxout
, encoding
);
436 #ifdef HAVE_PTHREAD_H
437 pthread_mutex_unlock(&map_mutex
);
438 #endif /* HAVE_PTHREAD_H */
445 * 'cupsUTF8ToUTF32()' - Convert UTF-8 to UTF-32.
447 * 32-bit UTF-32 (actually 21-bit) maps to UTF-8 as follows...
449 * UTF-32 char UTF-8 char(s)
450 * --------------------------------------------------
451 * 0 to 127 = 0xxxxxxx (US-ASCII)
452 * 128 to 2047 = 110xxxxx 10yyyyyy
453 * 2048 to 65535 = 1110xxxx 10yyyyyy 10zzzzzz
454 * > 65535 = 11110xxx 10yyyyyy 10zzzzzz 10xxxxxx
456 * UTF-32 prohibits chars beyond Plane 16 (> 0x10ffff) in UCS-4,
457 * which would convert to five- or six-octet UTF-8 sequences...
460 int /* O - Count or -1 on error */
462 cups_utf32_t
*dest
, /* O - Target string */
463 const cups_utf8_t
*src
, /* I - Source string */
464 const int maxout
) /* I - Max output */
466 int i
; /* Looping variable */
467 cups_utf8_t ch
; /* Character value */
468 cups_utf8_t next
; /* Next character value */
469 cups_utf32_t ch32
; /* UTF-32 character value */
473 * Check for valid arguments and clear output...
476 DEBUG_printf(("cupsUTF8ToUTF32(dest=%p, src=\"%s\", maxout=%d)\n", dest
,
477 src
? (const char *)src
: "(null)", maxout
));
482 if (!dest
|| !src
|| maxout
< 1 || maxout
> CUPS_MAX_USTRING
)
484 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad arguments)");
490 * Convert input UTF-8 to output UTF-32 (and insert BOM)...
495 for (i
= maxout
- 1; *src
&& i
> 0; i
--)
500 * Convert UTF-8 character(s) to UTF-32 character...
506 * One-octet UTF-8 <= 127 (US-ASCII)...
511 DEBUG_printf(("cupsUTF8ToUTF32: %02x => %08X\n", src
[-1], ch
));
514 else if ((ch
& 0xe0) == 0xc0)
517 * Two-octet UTF-8 <= 2047 (Latin-x)...
521 if ((next
& 0xc0) != 0x80)
523 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
528 ch32
= ((ch
& 0x1f) << 6) | (next
& 0x3f);
531 * Check for non-shortest form (invalid UTF-8)...
536 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
543 DEBUG_printf(("cupsUTF8ToUTF32: %02x %02x => %08X\n",
544 src
[-2], src
[-1], (unsigned)ch32
));
546 else if ((ch
& 0xf0) == 0xe0)
549 * Three-octet UTF-8 <= 65535 (Plane 0 - BMP)...
553 if ((next
& 0xc0) != 0x80)
555 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
560 ch32
= ((ch
& 0x0f) << 6) | (next
& 0x3f);
563 if ((next
& 0xc0) != 0x80)
565 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
570 ch32
= (ch32
<< 6) | (next
& 0x3f);
573 * Check for non-shortest form (invalid UTF-8)...
578 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
585 DEBUG_printf(("cupsUTF8ToUTF32: %02x %02x %02x => %08X\n",
586 src
[-3], src
[-2], src
[-1], (unsigned)ch32
));
588 else if ((ch
& 0xf8) == 0xf0)
591 * Four-octet UTF-8...
595 if ((next
& 0xc0) != 0x80)
597 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
602 ch32
= ((ch
& 0x07) << 6) | (next
& 0x3f);
605 if ((next
& 0xc0) != 0x80)
607 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
612 ch32
= (ch32
<< 6) | (next
& 0x3f);
615 if ((next
& 0xc0) != 0x80)
617 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
622 ch32
= (ch32
<< 6) | (next
& 0x3f);
625 * Check for non-shortest form (invalid UTF-8)...
630 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
637 DEBUG_printf(("cupsUTF8ToUTF32: %02x %02x %02x %02x => %08X\n",
638 src
[-4], src
[-3], src
[-2], src
[-1], (unsigned)ch32
));
643 * More than 4-octet (invalid UTF-8 sequence)...
646 DEBUG_puts("cupsUTF8ToUTF32: Returning -1 (bad UTF-8 sequence)");
652 * Check for UTF-16 surrogate (illegal UTF-8)...
655 if (ch32
>= 0xd800 && ch32
<= 0xdfff)
661 DEBUG_printf(("cupsUTF8ToUTF32: Returning %d characters\n", maxout
- 1 - i
));
663 return (maxout
- 1 - i
);
668 * 'cupsUTF32ToUTF8()' - Convert UTF-32 to UTF-8.
670 * 32-bit UTF-32 (actually 21-bit) maps to UTF-8 as follows...
672 * UTF-32 char UTF-8 char(s)
673 * --------------------------------------------------
674 * 0 to 127 = 0xxxxxxx (US-ASCII)
675 * 128 to 2047 = 110xxxxx 10yyyyyy
676 * 2048 to 65535 = 1110xxxx 10yyyyyy 10zzzzzz
677 * > 65535 = 11110xxx 10yyyyyy 10zzzzzz 10xxxxxx
679 * UTF-32 prohibits chars beyond Plane 16 (> 0x10ffff) in UCS-4,
680 * which would convert to five- or six-octet UTF-8 sequences...
683 int /* O - Count or -1 on error */
685 cups_utf8_t
*dest
, /* O - Target string */
686 const cups_utf32_t
*src
, /* I - Source string */
687 const int maxout
) /* I - Max output */
689 cups_utf8_t
*start
; /* Start of destination string */
690 int i
; /* Looping variable */
691 int swap
; /* Byte-swap input to output */
692 cups_utf32_t ch
; /* Character value */
696 * Check for valid arguments and clear output...
699 DEBUG_printf(("cupsUTF32ToUTF8(dest=%p, src=%p, maxout=%d)\n", dest
, src
,
705 if (!dest
|| !src
|| maxout
< 1)
707 DEBUG_puts("cupsUTF32ToUTF8: Returning -1 (bad args)");
713 * Check for leading BOM in UTF-32 and inverted BOM...
717 swap
= *src
== 0xfffe0000;
719 DEBUG_printf(("cupsUTF32ToUTF8: swap=%d\n", swap
));
721 if (*src
== 0xfffe0000 || *src
== 0xfeff)
725 * Convert input UTF-32 to output UTF-8...
728 for (i
= maxout
- 1; *src
&& i
> 0;)
733 * Byte swap input UTF-32, if necessary...
734 * (only byte-swapping 24 of 32 bits)
738 ch
= ((ch
>> 24) | ((ch
>> 8) & 0xff00) | ((ch
<< 8) & 0xff0000));
741 * Check for beyond Plane 16 (invalid UTF-32)...
746 DEBUG_puts("cupsUTF32ToUTF8: Returning -1 (character out of range)");
752 * Convert UTF-32 character to UTF-8 character(s)...
758 * One-octet UTF-8 <= 127 (US-ASCII)...
761 *dest
++ = (cups_utf8_t
)ch
;
764 DEBUG_printf(("cupsUTF32ToUTF8: %08x => %02x\n", (unsigned)ch
, dest
[-1]));
769 * Two-octet UTF-8 <= 2047 (Latin-x)...
774 DEBUG_puts("cupsUTF32ToUTF8: Returning -1 (too long 2)");
779 *dest
++ = (cups_utf8_t
)(0xc0 | ((ch
>> 6) & 0x1f));
780 *dest
++ = (cups_utf8_t
)(0x80 | (ch
& 0x3f));
783 DEBUG_printf(("cupsUTF32ToUTF8: %08x => %02x %02x\n", (unsigned)ch
,
784 dest
[-2], dest
[-1]));
786 else if (ch
< 0x10000)
789 * Three-octet UTF-8 <= 65535 (Plane 0 - BMP)...
794 DEBUG_puts("cupsUTF32ToUTF8: Returning -1 (too long 3)");
799 *dest
++ = (cups_utf8_t
)(0xe0 | ((ch
>> 12) & 0x0f));
800 *dest
++ = (cups_utf8_t
)(0x80 | ((ch
>> 6) & 0x3f));
801 *dest
++ = (cups_utf8_t
)(0x80 | (ch
& 0x3f));
804 DEBUG_printf(("cupsUTF32ToUTF8: %08x => %02x %02x %02x\n", (unsigned)ch
,
805 dest
[-3], dest
[-2], dest
[-1]));
810 * Four-octet UTF-8...
816 *dest
++ = (cups_utf8_t
)(0xf0 | ((ch
>> 18) & 0x07));
817 *dest
++ = (cups_utf8_t
)(0x80 | ((ch
>> 12) & 0x3f));
818 *dest
++ = (cups_utf8_t
)(0x80 | ((ch
>> 6) & 0x3f));
819 *dest
++ = (cups_utf8_t
)(0x80 | (ch
& 0x3f));
822 DEBUG_printf(("cupsUTF32ToUTF8: %08x => %02x %02x %02x %02x\n",
823 (unsigned)ch
, dest
[-4], dest
[-3], dest
[-2], dest
[-1]));
829 DEBUG_printf(("cupsUTF32ToUTF8: Returning %d\n", (int)(dest
- start
)));
831 return ((int)(dest
- start
));
836 * 'compare_wide()' - Compare key for wide (VBCS) match.
840 compare_wide(const void *k1
, /* I - Key char */
841 const void *k2
) /* I - Map char */
843 cups_vbcs_t key
; /* Legacy key character */
844 cups_vbcs_t map
; /* Legacy map character */
847 key
= *((cups_vbcs_t
*)k1
);
848 map
= ((_cups_wide2uni_t
*)k2
)->widechar
;
850 return ((int)(key
- map
));
855 * 'conv_sbcs_to_utf8()' - Convert legacy SBCS to UTF-8.
858 static int /* O - Count or -1 on error */
860 cups_utf8_t
*dest
, /* O - Target string */
861 const cups_sbcs_t
*src
, /* I - Source string */
862 int maxout
, /* I - Max output */
863 const cups_encoding_t encoding
) /* I - Encoding */
865 _cups_cmap_t
*cmap
; /* Legacy SBCS / Unicode Charset Map */
866 cups_ucs2_t
*crow
; /* Pointer to UCS-2 row in 'char2uni' */
867 cups_sbcs_t legchar
; /* Legacy character value */
868 cups_utf32_t work
[CUPS_MAX_USTRING
], /* Internal UCS-4 string */
869 *workptr
; /* Pointer into string */
873 * Find legacy charset map in cache...
876 if ((cmap
= (_cups_cmap_t
*)get_charmap(encoding
)) == NULL
)
880 * Convert input legacy charset to internal UCS-4 (and insert BOM)...
884 for (workptr
= work
+ 1; *src
&& workptr
< (work
+ CUPS_MAX_USTRING
- 1);)
889 * Convert ASCII verbatim (optimization)...
893 *workptr
++ = (cups_utf32_t
)legchar
;
897 * Convert unknown character to Replacement Character...
900 crow
= cmap
->char2uni
+ legchar
;
905 *workptr
++ = (cups_utf32_t
)*crow
;
912 * Convert internal UCS-4 to output UTF-8 (and delete BOM)...
917 return (cupsUTF32ToUTF8(dest
, work
, maxout
));
922 * 'conv_utf8_to_sbcs()' - Convert UTF-8 to legacy SBCS.
925 static int /* O - Count or -1 on error */
927 cups_sbcs_t
*dest
, /* O - Target string */
928 const cups_utf8_t
*src
, /* I - Source string */
929 int maxout
, /* I - Max output */
930 const cups_encoding_t encoding
) /* I - Encoding */
932 cups_sbcs_t
*start
; /* Start of destination string */
933 _cups_cmap_t
*cmap
; /* Legacy SBCS / Unicode Charset Map */
934 cups_sbcs_t
*srow
; /* Pointer to SBCS row in 'uni2char' */
935 cups_utf32_t unichar
; /* Character value */
936 cups_utf32_t work
[CUPS_MAX_USTRING
], /* Internal UCS-4 string */
937 *workptr
; /* Pointer into string */
941 * Find legacy charset map in cache...
944 if ((cmap
= (_cups_cmap_t
*)get_charmap(encoding
)) == NULL
)
948 * Convert input UTF-8 to internal UCS-4 (and insert BOM)...
951 if (cupsUTF8ToUTF32(work
, src
, CUPS_MAX_USTRING
) < 0)
955 * Convert internal UCS-4 to SBCS legacy charset (and delete BOM)...
958 for (workptr
= work
+ 1, start
= dest
; *workptr
&& maxout
> 1; maxout
--)
960 unichar
= *workptr
++;
965 * Convert ASCII verbatim (optimization)...
970 *dest
++ = (cups_sbcs_t
)unichar
;
975 * Convert unknown character to visible replacement...
978 srow
= cmap
->uni2char
[(int)((unichar
>> 8) & 0xff)];
981 srow
+= (int)(unichar
& 0xff);
993 return ((int)(dest
- start
));
998 * 'conv_utf8_to_vbcs()' - Convert UTF-8 to legacy DBCS/VBCS.
1001 static int /* O - Count or -1 on error */
1003 cups_sbcs_t
*dest
, /* O - Target string */
1004 const cups_utf8_t
*src
, /* I - Source string */
1005 int maxout
, /* I - Max output */
1006 const cups_encoding_t encoding
) /* I - Encoding */
1008 cups_sbcs_t
*start
; /* Start of destination string */
1009 _cups_vmap_t
*vmap
; /* Legacy DBCS / Unicode Charset Map */
1010 cups_vbcs_t
*vrow
; /* Pointer to VBCS row in 'uni2char' */
1011 cups_utf32_t unichar
; /* Character value */
1012 cups_vbcs_t legchar
; /* Legacy character value */
1013 cups_utf32_t work
[CUPS_MAX_USTRING
], /* Internal UCS-4 string */
1014 *workptr
; /* Pointer into string */
1017 DEBUG_printf(("conv_utf8_to_vbcs(dest=%p, src=\"%s\", maxout=%d, "
1018 "encoding=%d)\n", dest
, src
, maxout
, encoding
));
1021 * Find legacy charset map in cache...
1024 if ((vmap
= (_cups_vmap_t
*)get_charmap(encoding
)) == NULL
)
1026 DEBUG_puts("conv_utf8_to_vbcs: Returning -1 (no charmap)");
1032 * Convert input UTF-8 to internal UCS-4 (and insert BOM)...
1035 if (cupsUTF8ToUTF32(work
, src
, CUPS_MAX_USTRING
) < 0)
1037 DEBUG_puts("conv_utf8_to_vbcs: Returning -1 (Unable to convert to UTF-32)");
1043 * Convert internal UCS-4 to VBCS legacy charset (and delete BOM)...
1046 for (start
= dest
, workptr
= work
+ 1; *workptr
&& maxout
> 1; maxout
--)
1048 unichar
= *workptr
++;
1051 * Convert ASCII verbatim (optimization)...
1056 *dest
++ = (cups_sbcs_t
)unichar
;
1058 DEBUG_printf(("conv_utf8_to_vbcs: %08x => %02X\n", (unsigned)unichar
,
1065 * Convert unknown character to visible replacement...
1068 vrow
= vmap
->uni2char
[(int)((unichar
>> 8) & 0xff)];
1071 vrow
+= (int)(unichar
& 0xff);
1073 if (!vrow
|| !*vrow
)
1074 legchar
= (cups_vbcs_t
)'?';
1076 legchar
= (cups_vbcs_t
)*vrow
;
1079 * Save n-byte legacy character...
1082 if (legchar
> 0xffffff)
1086 DEBUG_puts("conv_utf8_to_vbcs: Returning -1 (out of space)");
1091 *dest
++ = (cups_sbcs_t
)(legchar
>> 24);
1092 *dest
++ = (cups_sbcs_t
)(legchar
>> 16);
1093 *dest
++ = (cups_sbcs_t
)(legchar
>> 8);
1094 *dest
++ = (cups_sbcs_t
)legchar
;
1098 DEBUG_printf(("conv_utf8_to_vbcs: %08x => %02X %02X %02X %02X\n",
1099 (unsigned)unichar
, dest
[-4], dest
[-3], dest
[-2], dest
[-1]));
1101 else if (legchar
> 0xffff)
1105 DEBUG_puts("conv_utf8_to_vbcs: Returning -1 (out of space)");
1110 *dest
++ = (cups_sbcs_t
)(legchar
>> 16);
1111 *dest
++ = (cups_sbcs_t
)(legchar
>> 8);
1112 *dest
++ = (cups_sbcs_t
)legchar
;
1116 DEBUG_printf(("conv_utf8_to_vbcs: %08x => %02X %02X %02X\n",
1117 (unsigned)unichar
, dest
[-3], dest
[-2], dest
[-1]));
1119 else if (legchar
> 0xff)
1121 *dest
++ = (cups_sbcs_t
)(legchar
>> 8);
1122 *dest
++ = (cups_sbcs_t
)legchar
;
1126 DEBUG_printf(("conv_utf8_to_vbcs: %08x => %02X %02X\n",
1127 (unsigned)unichar
, dest
[-2], dest
[-1]));
1133 DEBUG_printf(("conv_utf8_to_vbcs: %08x => %02X\n",
1134 (unsigned)unichar
, dest
[-1]));
1142 DEBUG_printf(("conv_utf8_to_vbcs: Returning %d characters\n",
1143 (int)(dest
- start
)));
1145 return ((int)(dest
- start
));
1150 * 'conv_vbcs_to_utf8()' - Convert legacy DBCS/VBCS to UTF-8.
1153 static int /* O - Count or -1 on error */
1155 cups_utf8_t
*dest
, /* O - Target string */
1156 const cups_sbcs_t
*src
, /* I - Source string */
1157 int maxout
, /* I - Max output */
1158 const cups_encoding_t encoding
) /* I - Encoding */
1160 _cups_vmap_t
*vmap
; /* Legacy VBCS / Unicode Charset Map */
1161 cups_ucs2_t
*crow
; /* Pointer to UCS-2 row in 'char2uni' */
1162 _cups_wide2uni_t
*wide2uni
; /* Pointer to row in 'wide2uni' */
1163 cups_sbcs_t leadchar
; /* Lead char of n-byte legacy char */
1164 cups_vbcs_t legchar
; /* Legacy character value */
1165 cups_utf32_t work
[CUPS_MAX_USTRING
], /* Internal UCS-4 string */
1166 *workptr
; /* Pointer into string */
1170 * Find legacy charset map in cache...
1173 DEBUG_printf(("conv_vbcs_to_utf8(dest=%p, src=%p, maxout=%d, encoding=%d)\n",
1174 dest
, src
, maxout
, encoding
));
1176 if ((vmap
= (_cups_vmap_t
*)get_charmap(encoding
)) == NULL
)
1178 DEBUG_puts("conv_vbcs_to_utf8: Returning -1 (NULL vmap)");
1184 * Convert input legacy charset to internal UCS-4 (and insert BOM)...
1188 for (workptr
= work
+ 1; *src
&& workptr
< (work
+ CUPS_MAX_USTRING
- 1);)
1191 leadchar
= (cups_sbcs_t
)legchar
;
1194 * Convert ASCII verbatim (optimization)...
1199 *workptr
++ = (cups_utf32_t
)legchar
;
1201 DEBUG_printf(("conv_vbcs_to_utf8: %02X => %08X\n", src
[-1],
1202 (unsigned)legchar
));
1207 * Convert 2-byte legacy character...
1210 if (vmap
->lead2char
[(int)leadchar
] == leadchar
)
1214 DEBUG_puts("conv_vbcs_to_utf8: Returning -1 (short string)");
1219 legchar
= (legchar
<< 8) | *src
++;
1222 * Convert unknown character to Replacement Character...
1225 crow
= vmap
->char2uni
[(int)((legchar
>> 8) & 0xff)];
1227 crow
+= (int) (legchar
& 0xff);
1229 if (!crow
|| !*crow
)
1230 *workptr
++ = 0xfffd;
1232 *workptr
++ = (cups_utf32_t
)*crow
;
1234 DEBUG_printf(("conv_vbcs_to_utf8: %02X %02X => %08X\n",
1235 src
[-2], src
[-1], (unsigned)workptr
[-1]));
1240 * Fetch 3-byte or 4-byte legacy character...
1243 if (vmap
->lead3char
[(int)leadchar
] == leadchar
)
1245 if (!*src
|| !src
[1])
1247 DEBUG_puts("conv_vbcs_to_utf8: Returning -1 (short string 2)");
1252 legchar
= (legchar
<< 8) | *src
++;
1253 legchar
= (legchar
<< 8) | *src
++;
1255 else if (vmap
->lead4char
[(int)leadchar
] == leadchar
)
1257 if (!*src
|| !src
[1] || !src
[2])
1259 DEBUG_puts("conv_vbcs_to_utf8: Returning -1 (short string 3)");
1264 legchar
= (legchar
<< 8) | *src
++;
1265 legchar
= (legchar
<< 8) | *src
++;
1266 legchar
= (legchar
<< 8) | *src
++;
1270 DEBUG_puts("conv_vbcs_to_utf8: Returning -1 (bad character)");
1276 * Find 3-byte or 4-byte legacy character...
1279 wide2uni
= (_cups_wide2uni_t
*)bsearch(&legchar
,
1282 sizeof(_cups_wide2uni_t
),
1286 * Convert unknown character to Replacement Character...
1289 if (!wide2uni
|| !wide2uni
->unichar
)
1290 *workptr
++ = 0xfffd;
1292 *workptr
++ = wide2uni
->unichar
;
1294 if (vmap
->lead3char
[(int)leadchar
] == leadchar
)
1295 DEBUG_printf(("conv_vbcs_to_utf8: %02X %02X %02X => %08X\n",
1296 src
[-3], src
[-2], src
[-1], (unsigned)workptr
[-1]));
1298 DEBUG_printf(("conv_vbcs_to_utf8: %02X %02X %02X %02X => %08X\n",
1299 src
[-4], src
[-3], src
[-2], src
[-1], (unsigned)workptr
[-1]));
1306 DEBUG_printf(("conv_vbcs_to_utf8: Converting %d UTF-32 characters to UTF-8\n",
1307 (int)(workptr
- work
)));
1310 * Convert internal UCS-4 to output UTF-8 (and delete BOM)...
1313 return (cupsUTF32ToUTF8(dest
, work
, maxout
));
1318 * 'free_sbcs_charmap()' - Free memory used by a single byte character set.
1322 free_sbcs_charmap(_cups_cmap_t
*cmap
) /* I - Character set */
1324 int i
; /* Looping variable */
1327 for (i
= 0; i
< 256; i
++)
1328 if (cmap
->uni2char
[i
])
1329 free(cmap
->uni2char
[i
]);
1336 * 'free_vbcs_charmap()' - Free memory used by a variable byte character set.
1340 free_vbcs_charmap(_cups_vmap_t
*vmap
) /* I - Character set */
1342 int i
; /* Looping variable */
1345 for (i
= 0; i
< 256; i
++)
1346 if (vmap
->char2uni
[i
])
1347 free(vmap
->char2uni
[i
]);
1349 for (i
= 0; i
< 256; i
++)
1350 if (vmap
->uni2char
[i
])
1351 free(vmap
->uni2char
[i
]);
1354 free(vmap
->wide2uni
);
1361 * 'get_charmap()' - Lookup or get a character set map (private).
1363 * This code handles single-byte (SBCS), double-byte (DBCS), and
1364 * variable-byte (VBCS) character sets _without_ charset escapes...
1365 * This code does not handle multiple-byte character sets (MBCS)
1366 * (such as ISO-2022-JP) with charset switching via escapes...
1370 static void * /* O - Charset map pointer */
1372 const cups_encoding_t encoding
) /* I - Encoding */
1374 char filename
[1024]; /* Filename for charset map file */
1375 _cups_globals_t
*cg
= _cupsGlobals(); /* Global data */
1378 DEBUG_printf(("get_charmap(encoding=%d)\n", encoding
));
1381 * Get the data directory and charset map name...
1384 snprintf(filename
, sizeof(filename
), "%s/charmaps/%s.txt",
1385 cg
->cups_datadir
, _cupsEncodingName(encoding
));
1387 DEBUG_printf(("get_charmap: filename=\"%s\"\n", filename
));
1390 * Read charset map input file into cache...
1393 if (encoding
< CUPS_ENCODING_SBCS_END
)
1394 return (get_sbcs_charmap(encoding
, filename
));
1395 else if (encoding
< CUPS_ENCODING_VBCS_END
)
1396 return (get_vbcs_charmap(encoding
, filename
));
1403 * 'get_charmap_count()' - Count lines in a charmap file.
1406 static int /* O - Count or -1 on error */
1407 get_charmap_count(cups_file_t
*fp
) /* I - File to read from */
1409 int count
; /* Number of lines */
1410 char line
[256]; /* Line from input map file */
1414 * Count lines in map input file...
1419 while (cupsFileGets(fp
, line
, sizeof(line
)))
1424 * Return the number of lines...
1435 * 'get_sbcs_charmap()' - Get SBCS Charmap.
1438 static _cups_cmap_t
* /* O - Charmap or 0 on error */
1440 const cups_encoding_t encoding
, /* I - Charmap Encoding */
1441 const char *filename
) /* I - Charmap Filename */
1443 unsigned long legchar
; /* Legacy character value */
1444 cups_utf32_t unichar
; /* Unicode character value */
1445 _cups_cmap_t
*cmap
; /* Legacy SBCS / Unicode Charset Map */
1446 cups_file_t
*fp
; /* Charset map file pointer */
1447 char *s
; /* Line parsing pointer */
1448 cups_ucs2_t
*crow
; /* Pointer to UCS-2 row in 'char2uni' */
1449 cups_sbcs_t
*srow
; /* Pointer to SBCS row in 'uni2char' */
1450 char line
[256]; /* Line from charset map file */
1454 * See if we already have this SBCS charset map loaded...
1457 DEBUG_printf(("get_sbcs_charmap(encoding=%d, filename=\"%s\")\n", encoding
,
1460 for (cmap
= cmap_cache
; cmap
; cmap
= cmap
->next
)
1462 if (cmap
->encoding
== encoding
)
1465 DEBUG_printf(("get_sbcs_charmap: Returning existing cmap=%p\n", cmap
));
1467 return ((void *)cmap
);
1472 * Open SBCS charset map input file...
1475 if ((fp
= cupsFileOpen(filename
, "r")) == NULL
)
1477 DEBUG_printf(("get_sbcs_charmap: Returning NULL (%s)\n", strerror(errno
)));
1483 * Allocate memory for SBCS charset map...
1486 if ((cmap
= (_cups_cmap_t
*)calloc(1, sizeof(_cups_cmap_t
))) == NULL
)
1489 DEBUG_puts("get_sbcs_charmap: Returning NULL (Unable to allocate memory)");
1495 cmap
->encoding
= encoding
;
1498 * Save SBCS charset map into memory for transcoding...
1501 while (cupsFileGets(fp
, line
, sizeof(line
)))
1506 legchar
= strtol(line
, &s
, 16);
1507 if (legchar
< 0 || legchar
> 0xff)
1510 unichar
= strtol(s
, NULL
, 16);
1511 if (unichar
< 0 || unichar
> 0xffff)
1515 * Save legacy to Unicode mapping in direct lookup table...
1518 crow
= cmap
->char2uni
+ legchar
;
1519 *crow
= (cups_ucs2_t
)(unichar
& 0xffff);
1522 * Save Unicode to legacy mapping in indirect lookup table...
1525 srow
= cmap
->uni2char
[(unichar
>> 8) & 0xff];
1528 srow
= (cups_sbcs_t
*)calloc(256, sizeof(cups_sbcs_t
));
1532 cmap
->uni2char
[(unichar
>> 8) & 0xff] = srow
;
1535 srow
+= unichar
& 0xff;
1538 * Convert Replacement Character to visible replacement...
1541 if (unichar
== 0xfffd)
1542 legchar
= (unsigned long)'?';
1545 * First (oldest) legacy character uses Unicode mapping cell...
1549 *srow
= (cups_sbcs_t
)legchar
;
1555 * Add it to the cache and return...
1558 cmap
->next
= cmap_cache
;
1561 DEBUG_printf(("get_sbcs_charmap: Returning new cmap=%p\n", cmap
));
1566 * If we get here, there was an error in the cmap file...
1571 free_sbcs_charmap(cmap
);
1575 DEBUG_puts("get_sbcs_charmap: Returning NULL (Read/format error)");
1582 * 'get_vbcs_charmap()' - Get DBCS/VBCS Charmap.
1585 static _cups_vmap_t
* /* O - Charmap or 0 on error */
1587 const cups_encoding_t encoding
, /* I - Charmap Encoding */
1588 const char *filename
) /* I - Charmap Filename */
1590 _cups_vmap_t
*vmap
; /* Legacy VBCS / Unicode Charset Map */
1591 cups_ucs2_t
*crow
; /* Pointer to UCS-2 row in 'char2uni' */
1592 cups_vbcs_t
*vrow
; /* Pointer to VBCS row in 'uni2char' */
1593 _cups_wide2uni_t
*wide2uni
; /* Pointer to row in 'wide2uni' */
1594 cups_sbcs_t leadchar
; /* Lead char of 2-byte legacy char */
1595 unsigned long legchar
; /* Legacy character value */
1596 cups_utf32_t unichar
; /* Unicode character value */
1597 int mapcount
; /* Count of lines in charmap file */
1598 cups_file_t
*fp
; /* Charset map file pointer */
1599 char *s
; /* Line parsing pointer */
1600 char line
[256]; /* Line from charset map file */
1601 int i
; /* Loop variable */
1602 int legacy
; /* 32-bit legacy char */
1605 DEBUG_printf(("get_vbcs_charmap(encoding=%d, filename=\"%s\")\n",
1606 encoding
, filename
));
1609 * See if we already have this DBCS/VBCS charset map loaded...
1612 for (vmap
= vmap_cache
; vmap
; vmap
= vmap
->next
)
1614 if (vmap
->encoding
== encoding
)
1617 DEBUG_printf(("get_vbcs_charmap: Returning existing vmap=%p\n", vmap
));
1619 return ((void *)vmap
);
1624 * Open VBCS charset map input file...
1627 if ((fp
= cupsFileOpen(filename
, "r")) == NULL
)
1629 DEBUG_printf(("get_vbcs_charmap: Returning NULL (%s)\n", strerror(errno
)));
1635 * Count lines in charmap file...
1638 if ((mapcount
= get_charmap_count(fp
)) <= 0)
1640 DEBUG_puts("get_vbcs_charmap: Unable to get charmap count!");
1647 DEBUG_printf(("get_vbcs_charmap: mapcount=%d\n", mapcount
));
1650 * Allocate memory for DBCS/VBCS charset map...
1653 if ((vmap
= (_cups_vmap_t
*)calloc(1, sizeof(_cups_vmap_t
))) == NULL
)
1655 DEBUG_puts("get_vbcs_charmap: Unable to allocate memory!");
1663 vmap
->encoding
= encoding
;
1666 * Save DBCS/VBCS charset map into memory for transcoding...
1677 while (cupsFileGets(fp
, line
, sizeof(line
)))
1682 legchar
= strtoul(line
, &s
, 16);
1683 if (legchar
== ULONG_MAX
)
1686 unichar
= strtol(s
, NULL
, 16);
1687 if (unichar
< 0 || unichar
> 0xffff)
1692 /* DEBUG_printf((" i=%d, legchar=0x%08lx, unichar=0x%04x\n", i,
1693 legchar, (unsigned)unichar)); */
1696 * Save lead char of 2/3/4-byte legacy char...
1699 if (legchar
> 0xffffff)
1701 leadchar
= (cups_sbcs_t
)(legchar
>> 24);
1702 vmap
->lead4char
[leadchar
] = leadchar
;
1704 else if (legchar
> 0xffff)
1706 leadchar
= (cups_sbcs_t
)(legchar
>> 16);
1707 vmap
->lead3char
[leadchar
] = leadchar
;
1709 else if (legchar
> 0xff)
1711 leadchar
= (cups_sbcs_t
)(legchar
>> 8);
1712 vmap
->lead2char
[leadchar
] = leadchar
;
1716 * Save Legacy to Unicode mapping...
1719 if (legchar
<= 0xffff)
1722 * Save DBCS 16-bit to Unicode mapping in indirect lookup table...
1725 crow
= vmap
->char2uni
[(int)leadchar
];
1728 crow
= (cups_ucs2_t
*)calloc(256, sizeof(cups_ucs2_t
));
1732 vmap
->char2uni
[(int)leadchar
] = crow
;
1735 crow
[(int)(legchar
& 0xff)] = (cups_ucs2_t
)unichar
;
1740 * Save VBCS 32-bit to Unicode mapping in sorted list table...
1746 vmap
->widecount
= (mapcount
- i
+ 1);
1747 wide2uni
= (_cups_wide2uni_t
*)calloc(vmap
->widecount
,
1748 sizeof(_cups_wide2uni_t
));
1752 vmap
->wide2uni
= wide2uni
;
1755 wide2uni
->widechar
= (cups_vbcs_t
)legchar
;
1756 wide2uni
->unichar
= (cups_ucs2_t
)unichar
;
1761 * Save Unicode to legacy mapping in indirect lookup table...
1764 vrow
= vmap
->uni2char
[(int)((unichar
>> 8) & 0xff)];
1767 vrow
= (cups_vbcs_t
*)calloc(256, sizeof(cups_vbcs_t
));
1771 vmap
->uni2char
[(int) ((unichar
>> 8) & 0xff)] = vrow
;
1774 vrow
+= (int)(unichar
& 0xff);
1777 * Convert Replacement Character to visible replacement...
1780 if (unichar
== 0xfffd)
1781 legchar
= (unsigned long)'?';
1784 * First (oldest) legacy character uses Unicode mapping cell...
1788 *vrow
= (cups_vbcs_t
)legchar
;
1791 vmap
->charcount
= (i
- vmap
->widecount
);
1796 * Add it to the cache and return...
1799 vmap
->next
= vmap_cache
;
1802 DEBUG_printf(("get_vbcs_charmap: Returning new vmap=%p\n", vmap
));
1807 * If we get here, the file contains errors...
1812 free_vbcs_charmap(vmap
);
1816 DEBUG_puts("get_vbcs_charmap: Returning NULL (Read/format error)");
1823 * End of "$Id: transcode.c 6649 2007-07-11 21:46:42Z mike $"