PostgreSQL Source Code  git master
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros
pg_wchar.h
Go to the documentation of this file.
1 /*-------------------------------------------------------------------------
2  *
3  * pg_wchar.h
4  * multibyte-character support
5  *
6  * Portions Copyright (c) 1996-2017, PostgreSQL Global Development Group
7  * Portions Copyright (c) 1994, Regents of the University of California
8  *
9  * src/include/mb/pg_wchar.h
10  *
11  * NOTES
12  * This is used both by the backend and by libpq, but should not be
13  * included by libpq client programs. In particular, a libpq client
14  * should not assume that the encoding IDs used by the version of libpq
15  * it's linked to match up with the IDs declared here.
16  *
17  *-------------------------------------------------------------------------
18  */
19 #ifndef PG_WCHAR_H
20 #define PG_WCHAR_H
21 
22 /*
23  * The pg_wchar type
24  */
25 typedef unsigned int pg_wchar;
26 
27 /*
28  * Maximum byte length of multibyte characters in any backend encoding
29  */
30 #define MAX_MULTIBYTE_CHAR_LEN 4
31 
32 /*
33  * various definitions for EUC
34  */
35 #define SS2 0x8e /* single shift 2 (JIS0201) */
36 #define SS3 0x8f /* single shift 3 (JIS0212) */
37 
38 /*
39  * SJIS validation macros
40  */
41 #define ISSJISHEAD(c) (((c) >= 0x81 && (c) <= 0x9f) || ((c) >= 0xe0 && (c) <= 0xfc))
42 #define ISSJISTAIL(c) (((c) >= 0x40 && (c) <= 0x7e) || ((c) >= 0x80 && (c) <= 0xfc))
43 
44 /*----------------------------------------------------
45  * MULE Internal Encoding (MIC)
46  *
47  * This encoding follows the design used within XEmacs; it is meant to
48  * subsume many externally-defined character sets. Each character includes
49  * identification of the character set it belongs to, so the encoding is
50  * general but somewhat bulky.
51  *
52  * Currently PostgreSQL supports 5 types of MULE character sets:
53  *
54  * 1) 1-byte ASCII characters. Each byte is below 0x80.
55  *
56  * 2) "Official" single byte charsets such as ISO-8859-1 (Latin1).
57  * Each MULE character consists of 2 bytes: LC1 + C1, where LC1 is
58  * an identifier for the charset (in the range 0x81 to 0x8d) and C1
59  * is the character code (in the range 0xa0 to 0xff).
60  *
61  * 3) "Private" single byte charsets such as SISHENG. Each MULE
62  * character consists of 3 bytes: LCPRV1 + LC12 + C1, where LCPRV1
63  * is a private-charset flag, LC12 is an identifier for the charset,
64  * and C1 is the character code (in the range 0xa0 to 0xff).
65  * LCPRV1 is either 0x9a (if LC12 is in the range 0xa0 to 0xdf)
66  * or 0x9b (if LC12 is in the range 0xe0 to 0xef).
67  *
68  * 4) "Official" multibyte charsets such as JIS X0208. Each MULE
69  * character consists of 3 bytes: LC2 + C1 + C2, where LC2 is
70  * an identifier for the charset (in the range 0x90 to 0x99) and C1
71  * and C2 form the character code (each in the range 0xa0 to 0xff).
72  *
73  * 5) "Private" multibyte charsets such as CNS 11643-1992 Plane 3.
74  * Each MULE character consists of 4 bytes: LCPRV2 + LC22 + C1 + C2,
75  * where LCPRV2 is a private-charset flag, LC22 is an identifier for
76  * the charset, and C1 and C2 form the character code (each in the range
77  * 0xa0 to 0xff). LCPRV2 is either 0x9c (if LC22 is in the range 0xf0
78  * to 0xf4) or 0x9d (if LC22 is in the range 0xf5 to 0xfe).
79  *
80  * "Official" encodings are those that have been assigned code numbers by
81  * the XEmacs project; "private" encodings have Postgres-specific charset
82  * identifiers.
83  *
84  * See the "XEmacs Internals Manual", available at http://www.xemacs.org,
85  * for more details. Note that for historical reasons, Postgres'
86  * private-charset flag values do not match what XEmacs says they should be,
87  * so this isn't really exactly MULE (not that private charsets would be
88  * interoperable anyway).
89  *
90  * Note that XEmacs's implementation is different from what emacs does.
91  * We follow emacs's implementation, rather than XEmacs's.
92  *----------------------------------------------------
93  */
94 
95 /*
96  * Charset identifiers (also called "leading bytes" in the MULE documentation)
97  */
98 
99 /*
100  * Charset IDs for official single byte encodings (0x81-0x8e)
101  */
102 #define LC_ISO8859_1 0x81 /* ISO8859 Latin 1 */
103 #define LC_ISO8859_2 0x82 /* ISO8859 Latin 2 */
104 #define LC_ISO8859_3 0x83 /* ISO8859 Latin 3 */
105 #define LC_ISO8859_4 0x84 /* ISO8859 Latin 4 */
106 #define LC_TIS620 0x85 /* Thai (not supported yet) */
107 #define LC_ISO8859_7 0x86 /* Greek (not supported yet) */
108 #define LC_ISO8859_6 0x87 /* Arabic (not supported yet) */
109 #define LC_ISO8859_8 0x88 /* Hebrew (not supported yet) */
110 #define LC_JISX0201K 0x89 /* Japanese 1 byte kana */
111 #define LC_JISX0201R 0x8a /* Japanese 1 byte Roman */
112 /* Note that 0x8b seems to be unused as of Emacs 20.7.
113  * However, there might be a chance that 0x8b could be used
114  * in later versions of Emacs.
115  */
116 #define LC_KOI8_R 0x8b /* Cyrillic KOI8-R */
117 #define LC_ISO8859_5 0x8c /* ISO8859 Cyrillic */
118 #define LC_ISO8859_9 0x8d /* ISO8859 Latin 5 (not supported yet) */
119 #define LC_ISO8859_15 0x8e /* ISO8859 Latin 15 (not supported yet) */
120 /* #define CONTROL_1 0x8f control characters (unused) */
121 
122 /* Is a leading byte for "official" single byte encodings? */
123 #define IS_LC1(c) ((unsigned char)(c) >= 0x81 && (unsigned char)(c) <= 0x8d)
124 
125 /*
126  * Charset IDs for official multibyte encodings (0x90-0x99)
127  * 0x9a-0x9d are free. 0x9e and 0x9f are reserved.
128  */
129 #define LC_JISX0208_1978 0x90 /* Japanese Kanji, old JIS (not supported) */
130 #define LC_GB2312_80 0x91 /* Chinese */
131 #define LC_JISX0208 0x92 /* Japanese Kanji (JIS X 0208) */
132 #define LC_KS5601 0x93 /* Korean */
133 #define LC_JISX0212 0x94 /* Japanese Kanji (JIS X 0212) */
134 #define LC_CNS11643_1 0x95 /* CNS 11643-1992 Plane 1 */
135 #define LC_CNS11643_2 0x96 /* CNS 11643-1992 Plane 2 */
136 #define LC_JISX0213_1 0x97/* Japanese Kanji (JIS X 0213 Plane 1) (not
137  * supported) */
138 #define LC_BIG5_1 0x98 /* Plane 1 Chinese traditional (not supported) */
139 #define LC_BIG5_2 0x99 /* Plane 1 Chinese traditional (not supported) */
140 
141 /* Is a leading byte for "official" multibyte encodings? */
142 #define IS_LC2(c) ((unsigned char)(c) >= 0x90 && (unsigned char)(c) <= 0x99)
143 
144 /*
145  * Postgres-specific prefix bytes for "private" single byte encodings
146  * (According to the MULE docs, we should be using 0x9e for this)
147  */
148 #define LCPRV1_A 0x9a
149 #define LCPRV1_B 0x9b
150 #define IS_LCPRV1(c) ((unsigned char)(c) == LCPRV1_A || (unsigned char)(c) == LCPRV1_B)
151 #define IS_LCPRV1_A_RANGE(c) \
152  ((unsigned char)(c) >= 0xa0 && (unsigned char)(c) <= 0xdf)
153 #define IS_LCPRV1_B_RANGE(c) \
154  ((unsigned char)(c) >= 0xe0 && (unsigned char)(c) <= 0xef)
155 
156 /*
157  * Postgres-specific prefix bytes for "private" multibyte encodings
158  * (According to the MULE docs, we should be using 0x9f for this)
159  */
160 #define LCPRV2_A 0x9c
161 #define LCPRV2_B 0x9d
162 #define IS_LCPRV2(c) ((unsigned char)(c) == LCPRV2_A || (unsigned char)(c) == LCPRV2_B)
163 #define IS_LCPRV2_A_RANGE(c) \
164  ((unsigned char)(c) >= 0xf0 && (unsigned char)(c) <= 0xf4)
165 #define IS_LCPRV2_B_RANGE(c) \
166  ((unsigned char)(c) >= 0xf5 && (unsigned char)(c) <= 0xfe)
167 
168 /*
169  * Charset IDs for private single byte encodings (0xa0-0xef)
170  */
171 #define LC_SISHENG 0xa0/* Chinese SiSheng characters for
172  * PinYin/ZhuYin (not supported) */
173 #define LC_IPA 0xa1/* IPA (International Phonetic Association)
174  * (not supported) */
175 #define LC_VISCII_LOWER 0xa2/* Vietnamese VISCII1.1 lower-case (not
176  * supported) */
177 #define LC_VISCII_UPPER 0xa3/* Vietnamese VISCII1.1 upper-case (not
178  * supported) */
179 #define LC_ARABIC_DIGIT 0xa4 /* Arabic digit (not supported) */
180 #define LC_ARABIC_1_COLUMN 0xa5 /* Arabic 1-column (not supported) */
181 #define LC_ASCII_RIGHT_TO_LEFT 0xa6 /* ASCII (left half of ISO8859-1) with
182  * right-to-left direction (not
183  * supported) */
184 #define LC_LAO 0xa7/* Lao characters (ISO10646 0E80..0EDF) (not
185  * supported) */
186 #define LC_ARABIC_2_COLUMN 0xa8 /* Arabic 1-column (not supported) */
187 
188 /*
189  * Charset IDs for private multibyte encodings (0xf0-0xff)
190  */
191 #define LC_INDIAN_1_COLUMN 0xf0/* Indian charset for 1-column width glyphs
192  * (not supported) */
193 #define LC_TIBETAN_1_COLUMN 0xf1/* Tibetan 1-column width glyphs (not
194  * supported) */
195 #define LC_UNICODE_SUBSET_2 0xf2/* Unicode characters of the range
196  * U+2500..U+33FF. (not supported) */
197 #define LC_UNICODE_SUBSET_3 0xf3/* Unicode characters of the range
198  * U+E000..U+FFFF. (not supported) */
199 #define LC_UNICODE_SUBSET 0xf4/* Unicode characters of the range
200  * U+0100..U+24FF. (not supported) */
201 #define LC_ETHIOPIC 0xf5 /* Ethiopic characters (not supported) */
202 #define LC_CNS11643_3 0xf6 /* CNS 11643-1992 Plane 3 */
203 #define LC_CNS11643_4 0xf7 /* CNS 11643-1992 Plane 4 */
204 #define LC_CNS11643_5 0xf8 /* CNS 11643-1992 Plane 5 */
205 #define LC_CNS11643_6 0xf9 /* CNS 11643-1992 Plane 6 */
206 #define LC_CNS11643_7 0xfa /* CNS 11643-1992 Plane 7 */
207 #define LC_INDIAN_2_COLUMN 0xfb/* Indian charset for 2-column width glyphs
208  * (not supported) */
209 #define LC_TIBETAN 0xfc /* Tibetan (not supported) */
210 /* #define FREE 0xfd free (unused) */
211 /* #define FREE 0xfe free (unused) */
212 /* #define FREE 0xff free (unused) */
213 
214 /*----------------------------------------------------
215  * end of MULE stuff
216  *----------------------------------------------------
217  */
218 
219 /*
220  * PostgreSQL encoding identifiers
221  *
222  * WARNING: the order of this enum must be same as order of entries
223  * in the pg_enc2name_tbl[] array (in mb/encnames.c), and
224  * in the pg_wchar_table[] array (in mb/wchar.c)!
225  *
226  * If you add some encoding don't forget to check
227  * PG_ENCODING_BE_LAST macro.
228  *
229  * PG_SQL_ASCII is default encoding and must be = 0.
230  *
231  * XXX We must avoid renumbering any backend encoding until libpq's major
232  * version number is increased beyond 5; it turns out that the backend
233  * encoding IDs are effectively part of libpq's ABI as far as 8.2 initdb and
234  * psql are concerned.
235  */
236 typedef enum pg_enc
237 {
238  PG_SQL_ASCII = 0, /* SQL/ASCII */
239  PG_EUC_JP, /* EUC for Japanese */
240  PG_EUC_CN, /* EUC for Chinese */
241  PG_EUC_KR, /* EUC for Korean */
242  PG_EUC_TW, /* EUC for Taiwan */
243  PG_EUC_JIS_2004, /* EUC-JIS-2004 */
244  PG_UTF8, /* Unicode UTF8 */
245  PG_MULE_INTERNAL, /* Mule internal code */
246  PG_LATIN1, /* ISO-8859-1 Latin 1 */
247  PG_LATIN2, /* ISO-8859-2 Latin 2 */
248  PG_LATIN3, /* ISO-8859-3 Latin 3 */
249  PG_LATIN4, /* ISO-8859-4 Latin 4 */
250  PG_LATIN5, /* ISO-8859-9 Latin 5 */
251  PG_LATIN6, /* ISO-8859-10 Latin6 */
252  PG_LATIN7, /* ISO-8859-13 Latin7 */
253  PG_LATIN8, /* ISO-8859-14 Latin8 */
254  PG_LATIN9, /* ISO-8859-15 Latin9 */
255  PG_LATIN10, /* ISO-8859-16 Latin10 */
256  PG_WIN1256, /* windows-1256 */
257  PG_WIN1258, /* Windows-1258 */
258  PG_WIN866, /* (MS-DOS CP866) */
259  PG_WIN874, /* windows-874 */
260  PG_KOI8R, /* KOI8-R */
261  PG_WIN1251, /* windows-1251 */
262  PG_WIN1252, /* windows-1252 */
263  PG_ISO_8859_5, /* ISO-8859-5 */
264  PG_ISO_8859_6, /* ISO-8859-6 */
265  PG_ISO_8859_7, /* ISO-8859-7 */
266  PG_ISO_8859_8, /* ISO-8859-8 */
267  PG_WIN1250, /* windows-1250 */
268  PG_WIN1253, /* windows-1253 */
269  PG_WIN1254, /* windows-1254 */
270  PG_WIN1255, /* windows-1255 */
271  PG_WIN1257, /* windows-1257 */
272  PG_KOI8U, /* KOI8-U */
273  /* PG_ENCODING_BE_LAST points to the above entry */
274 
275  /* followings are for client encoding only */
276  PG_SJIS, /* Shift JIS (Windows-932) */
277  PG_BIG5, /* Big5 (Windows-950) */
278  PG_GBK, /* GBK (Windows-936) */
279  PG_UHC, /* UHC (Windows-949) */
280  PG_GB18030, /* GB18030 */
281  PG_JOHAB, /* EUC for Korean JOHAB */
282  PG_SHIFT_JIS_2004, /* Shift-JIS-2004 */
283  _PG_LAST_ENCODING_ /* mark only */
284 
285 } pg_enc;
286 
287 #define PG_ENCODING_BE_LAST PG_KOI8U
288 
289 /*
290  * Please use these tests before access to pg_encconv_tbl[]
291  * or to other places...
292  */
293 #define PG_VALID_BE_ENCODING(_enc) \
294  ((_enc) >= 0 && (_enc) <= PG_ENCODING_BE_LAST)
295 
296 #define PG_ENCODING_IS_CLIENT_ONLY(_enc) \
297  ((_enc) > PG_ENCODING_BE_LAST && (_enc) < _PG_LAST_ENCODING_)
298 
299 #define PG_VALID_ENCODING(_enc) \
300  ((_enc) >= 0 && (_enc) < _PG_LAST_ENCODING_)
301 
302 /* On FE are possible all encodings */
303 #define PG_VALID_FE_ENCODING(_enc) PG_VALID_ENCODING(_enc)
304 
305 /*
306  * Table for mapping an encoding number to official encoding name and
307  * possibly other subsidiary data. Be careful to check encoding number
308  * before accessing a table entry!
309  *
310  * if (PG_VALID_ENCODING(encoding))
311  * pg_enc2name_tbl[ encoding ];
312  */
313 typedef struct pg_enc2name
314 {
315  const char *name;
317 #ifdef WIN32
318  unsigned codepage; /* codepage for WIN32 */
319 #endif
320 } pg_enc2name;
321 
322 extern const pg_enc2name pg_enc2name_tbl[];
323 
324 /*
325  * Encoding names for gettext
326  */
327 typedef struct pg_enc2gettext
328 {
330  const char *name;
332 
333 extern const pg_enc2gettext pg_enc2gettext_tbl[];
334 
335 /*
336  * Encoding names for ICU
337  */
338 extern bool is_encoding_supported_by_icu(int encoding);
339 extern const char *get_encoding_name_for_icu(int encoding);
340 
341 /*
342  * pg_wchar stuff
343  */
344 typedef int (*mb2wchar_with_len_converter) (const unsigned char *from,
345  pg_wchar *to,
346  int len);
347 
348 typedef int (*wchar2mb_with_len_converter) (const pg_wchar *from,
349  unsigned char *to,
350  int len);
351 
352 typedef int (*mblen_converter) (const unsigned char *mbstr);
353 
354 typedef int (*mbdisplaylen_converter) (const unsigned char *mbstr);
355 
356 typedef bool (*mbcharacter_incrementer) (unsigned char *mbstr, int len);
357 
358 typedef int (*mbverifier) (const unsigned char *mbstr, int len);
359 
360 typedef struct
361 {
363  * string to a wchar */
365  * string to a multibyte */
366  mblen_converter mblen; /* get byte length of a char */
367  mbdisplaylen_converter dsplen; /* get display width of a char */
368  mbverifier mbverify; /* verify multibyte sequence */
369  int maxmblen; /* max bytes for a char in this encoding */
370 } pg_wchar_tbl;
371 
372 extern const pg_wchar_tbl pg_wchar_table[];
373 
374 /*
375  * Data structures for conversions between UTF-8 and other encodings
376  * (UtfToLocal() and LocalToUtf()). In these data structures, characters of
377  * either encoding are represented by uint32 words; hence we can only support
378  * characters up to 4 bytes long. For example, the byte sequence 0xC2 0x89
379  * would be represented by 0x0000C289, and 0xE8 0xA2 0xB4 by 0x00E8A2B4.
380  *
381  * There are three possible ways a character can be mapped:
382  *
383  * 1. Using a radix tree, from source to destination code.
384  * 2. Using a sorted array of source -> destination code pairs. This
385  * method is used for "combining" characters. There are so few of
386  * them that building a radix tree would be wasteful.
387  * 3. Using a conversion function.
388  */
389 
390 /*
391  * Radix tree for character conversion.
392  *
393  * Logically, this is actually four different radix trees, for 1-byte,
394  * 2-byte, 3-byte and 4-byte inputs. The 1-byte tree is a simple lookup
395  * table from source to target code. The 2-byte tree consists of two levels:
396  * one lookup table for the first byte, where the value in the lookup table
397  * points to a lookup table for the second byte. And so on.
398  *
399  * Physically, all the trees are stored in one big array, in 'chars16' or
400  * 'chars32', depending on the maximum value that needs to be reprented. For
401  * each level in each tree, we also store lower and upper bound of allowed
402  * values - values outside those bounds are considered invalid, and are left
403  * out of the tables.
404  *
405  * In the intermediate levels of the trees, the values stored are offsets
406  * into the chars[16|32] array.
407  *
408  * In the beginning of the chars[16|32] array, there is always a number of
409  * zeros, so that you safely follow an index from an intermediate table
410  * without explicitly checking for a zero. Following a zero any number of
411  * times will always bring you to the dummy, all-zeros table in the
412  * beginning. This helps to shave some cycles when looking up values.
413  */
414 typedef struct
415 {
416  /*
417  * Array containing all the values. Only one of chars16 or chars32 is
418  * used, depending on how wide the values we need to represent are.
419  */
420  const uint16 *chars16;
421  const uint32 *chars32;
422 
423  /* Radix tree for 1-byte inputs */
424  uint32 b1root; /* offset of table in the chars[16|32] array */
425  uint8 b1_lower; /* min allowed value for a single byte input */
426  uint8 b1_upper; /* max allowed value for a single byte input */
427 
428  /* Radix tree for 2-byte inputs */
429  uint32 b2root; /* offset of 1st byte's table */
430  uint8 b2_1_lower; /* min/max allowed value for 1st input byte */
432  uint8 b2_2_lower; /* min/max allowed value for 2nd input byte */
434 
435  /* Radix tree for 3-byte inputs */
436  uint32 b3root; /* offset of 1st byte's table */
437  uint8 b3_1_lower; /* min/max allowed value for 1st input byte */
439  uint8 b3_2_lower; /* min/max allowed value for 2nd input byte */
441  uint8 b3_3_lower; /* min/max allowed value for 3rd input byte */
443 
444  /* Radix tree for 4-byte inputs */
445  uint32 b4root; /* offset of 1st byte's table */
446  uint8 b4_1_lower; /* min/max allowed value for 1st input byte */
448  uint8 b4_2_lower; /* min/max allowed value for 2nd input byte */
450  uint8 b4_3_lower; /* min/max allowed value for 3rd input byte */
452  uint8 b4_4_lower; /* min/max allowed value for 4th input byte */
454 
456 
457 /*
458  * UTF-8 to local code conversion map (for combined characters)
459  */
460 typedef struct
461 {
462  uint32 utf1; /* UTF-8 code 1 */
463  uint32 utf2; /* UTF-8 code 2 */
464  uint32 code; /* local code */
466 
467 /*
468  * local code to UTF-8 conversion map (for combined characters)
469  */
470 typedef struct
471 {
472  uint32 code; /* local code */
473  uint32 utf1; /* UTF-8 code 1 */
474  uint32 utf2; /* UTF-8 code 2 */
476 
477 /*
478  * callback function for algorithmic encoding conversions (in either direction)
479  *
480  * if function returns zero, it does not know how to convert the code
481  */
483 
484 /*
485  * Support macro for encoding conversion functions to validate their
486  * arguments. (This could be made more compact if we included fmgr.h
487  * here, but we don't want to do that because this header file is also
488  * used by frontends.)
489  */
490 #define CHECK_ENCODING_CONVERSION_ARGS(srcencoding,destencoding) \
491  check_encoding_conversion_args(PG_GETARG_INT32(0), \
492  PG_GETARG_INT32(1), \
493  PG_GETARG_INT32(4), \
494  (srcencoding), \
495  (destencoding))
496 
497 
498 /*
499  * These functions are considered part of libpq's exported API and
500  * are also declared in libpq-fe.h.
501  */
502 extern int pg_char_to_encoding(const char *name);
503 extern const char *pg_encoding_to_char(int encoding);
504 extern int pg_valid_server_encoding_id(int encoding);
505 
506 /*
507  * Remaining functions are not considered part of libpq's API, though many
508  * of them do exist inside libpq.
509  */
510 extern int pg_mb2wchar(const char *from, pg_wchar *to);
511 extern int pg_mb2wchar_with_len(const char *from, pg_wchar *to, int len);
513  const char *from, pg_wchar *to, int len);
514 extern int pg_wchar2mb(const pg_wchar *from, char *to);
515 extern int pg_wchar2mb_with_len(const pg_wchar *from, char *to, int len);
517  const pg_wchar *from, char *to, int len);
518 extern int pg_char_and_wchar_strcmp(const char *s1, const pg_wchar *s2);
519 extern int pg_wchar_strncmp(const pg_wchar *s1, const pg_wchar *s2, size_t n);
520 extern int pg_char_and_wchar_strncmp(const char *s1, const pg_wchar *s2, size_t n);
521 extern size_t pg_wchar_strlen(const pg_wchar *wstr);
522 extern int pg_mblen(const char *mbstr);
523 extern int pg_dsplen(const char *mbstr);
524 extern int pg_encoding_mblen(int encoding, const char *mbstr);
525 extern int pg_encoding_dsplen(int encoding, const char *mbstr);
526 extern int pg_encoding_verifymb(int encoding, const char *mbstr, int len);
527 extern int pg_mule_mblen(const unsigned char *mbstr);
528 extern int pg_mic_mblen(const unsigned char *mbstr);
529 extern int pg_mbstrlen(const char *mbstr);
530 extern int pg_mbstrlen_with_len(const char *mbstr, int len);
531 extern int pg_mbcliplen(const char *mbstr, int len, int limit);
532 extern int pg_encoding_mbcliplen(int encoding, const char *mbstr,
533  int len, int limit);
534 extern int pg_mbcharcliplen(const char *mbstr, int len, int imit);
535 extern int pg_encoding_max_length(int encoding);
536 extern int pg_database_encoding_max_length(void);
538 
539 extern int PrepareClientEncoding(int encoding);
540 extern int SetClientEncoding(int encoding);
541 extern void InitializeClientEncoding(void);
542 extern int pg_get_client_encoding(void);
543 extern const char *pg_get_client_encoding_name(void);
544 
545 extern void SetDatabaseEncoding(int encoding);
546 extern int GetDatabaseEncoding(void);
547 extern const char *GetDatabaseEncodingName(void);
548 extern void SetMessageEncoding(int encoding);
549 extern int GetMessageEncoding(void);
550 
551 #ifdef ENABLE_NLS
552 extern int pg_bind_textdomain_codeset(const char *domainname);
553 #endif
554 
555 extern int pg_valid_client_encoding(const char *name);
556 extern int pg_valid_server_encoding(const char *name);
557 
558 extern unsigned char *unicode_to_utf8(pg_wchar c, unsigned char *utf8string);
559 extern pg_wchar utf8_to_unicode(const unsigned char *c);
560 extern int pg_utf_mblen(const unsigned char *);
561 extern unsigned char *pg_do_encoding_conversion(unsigned char *src, int len,
562  int src_encoding,
563  int dest_encoding);
564 
565 extern char *pg_client_to_server(const char *s, int len);
566 extern char *pg_server_to_client(const char *s, int len);
567 extern char *pg_any_to_server(const char *s, int len, int encoding);
568 extern char *pg_server_to_any(const char *s, int len, int encoding);
569 
570 extern unsigned short BIG5toCNS(unsigned short big5, unsigned char *lc);
571 extern unsigned short CNStoBIG5(unsigned short cns, unsigned char lc);
572 
573 extern void UtfToLocal(const unsigned char *utf, int len,
574  unsigned char *iso,
575  const pg_mb_radix_tree *map,
576  const pg_utf_to_local_combined *cmap, int cmapsize,
577  utf_local_conversion_func conv_func,
578  int encoding);
579 extern void LocalToUtf(const unsigned char *iso, int len,
580  unsigned char *utf,
581  const pg_mb_radix_tree *map,
582  const pg_local_to_utf_combined *cmap, int cmapsize,
583  utf_local_conversion_func conv_func,
584  int encoding);
585 
586 extern bool pg_verifymbstr(const char *mbstr, int len, bool noError);
587 extern bool pg_verify_mbstr(int encoding, const char *mbstr, int len,
588  bool noError);
589 extern int pg_verify_mbstr_len(int encoding, const char *mbstr, int len,
590  bool noError);
591 
592 extern void check_encoding_conversion_args(int src_encoding,
593  int dest_encoding,
594  int len,
595  int expected_src_encoding,
596  int expected_dest_encoding);
597 
598 extern void report_invalid_encoding(int encoding, const char *mbstr, int len) pg_attribute_noreturn();
599 extern void report_untranslatable_char(int src_encoding, int dest_encoding,
600  const char *mbstr, int len) pg_attribute_noreturn();
601 
602 extern void local2local(const unsigned char *l, unsigned char *p, int len,
603  int src_encoding, int dest_encoding, const unsigned char *tab);
604 extern void pg_ascii2mic(const unsigned char *l, unsigned char *p, int len);
605 extern void pg_mic2ascii(const unsigned char *mic, unsigned char *p, int len);
606 extern void latin2mic(const unsigned char *l, unsigned char *p, int len,
607  int lc, int encoding);
608 extern void mic2latin(const unsigned char *mic, unsigned char *p, int len,
609  int lc, int encoding);
610 extern void latin2mic_with_table(const unsigned char *l, unsigned char *p,
611  int len, int lc, int encoding,
612  const unsigned char *tab);
613 extern void mic2latin_with_table(const unsigned char *mic, unsigned char *p,
614  int len, int lc, int encoding,
615  const unsigned char *tab);
616 
617 extern bool pg_utf8_islegal(const unsigned char *source, int length);
618 
619 #ifdef WIN32
620 extern WCHAR *pgwin32_message_to_UTF16(const char *str, int len, int *utf16len);
621 #endif
622 
623 #endif /* PG_WCHAR_H */
int pg_mb2wchar(const char *from, pg_wchar *to)
Definition: mbutils.c:727
char * pg_any_to_server(const char *s, int len, int encoding)
Definition: mbutils.c:572
int length(const List *list)
Definition: list.c:1271
uint8 b2_2_upper
Definition: pg_wchar.h:433
uint8 b4_2_lower
Definition: pg_wchar.h:448
void latin2mic_with_table(const unsigned char *l, unsigned char *p, int len, int lc, int encoding, const unsigned char *tab)
Definition: conv.c:193
int pg_char_to_encoding(const char *name)
Definition: encnames.c:551
const pg_enc2gettext pg_enc2gettext_tbl[]
Definition: encnames.c:359
const pg_wchar_tbl pg_wchar_table[]
Definition: wchar.c:1729
void latin2mic(const unsigned char *l, unsigned char *p, int len, int lc, int encoding)
Definition: conv.c:71
const uint16 * chars16
Definition: pg_wchar.h:420
void InitializeClientEncoding(void)
Definition: mbutils.c:293
bool pg_utf8_islegal(const unsigned char *source, int length)
Definition: wchar.c:1458
const char * GetDatabaseEncodingName(void)
Definition: mbutils.c:1021
int pg_encoding_mb2wchar_with_len(int encoding, const char *from, pg_wchar *to, int len)
Definition: mbutils.c:741
uint8 b2_1_lower
Definition: pg_wchar.h:430
uint8 b4_4_upper
Definition: pg_wchar.h:453
unsigned short CNStoBIG5(unsigned short cns, unsigned char lc)
Definition: big5.c:345
void LocalToUtf(const unsigned char *iso, int len, unsigned char *utf, const pg_mb_radix_tree *map, const pg_local_to_utf_combined *cmap, int cmapsize, utf_local_conversion_func conv_func, int encoding)
Definition: conv.c:665
unsigned char uint8
Definition: c.h:266
int pg_mb2wchar_with_len(const char *from, pg_wchar *to, int len)
Definition: mbutils.c:734
int pg_mic_mblen(const unsigned char *mbstr)
Definition: wchar.c:1776
size_t pg_wchar_strlen(const pg_wchar *wstr)
Definition: wstrncmp.c:70
const char * name
Definition: pg_wchar.h:330
int pg_valid_server_encoding_id(int encoding)
Definition: encnames.c:515
const char * pg_encoding_to_char(int encoding)
Definition: encnames.c:607
int pg_mbcliplen(const char *mbstr, int len, int limit)
Definition: mbutils.c:831
int pg_encoding_mblen(int encoding, const char *mbstr)
Definition: wchar.c:1785
void pg_mic2ascii(const unsigned char *mic, unsigned char *p, int len)
Definition: conv.c:163
int pg_database_encoding_max_length(void)
Definition: wchar.c:1833
int pg_encoding_dsplen(int encoding, const char *mbstr)
Definition: wchar.c:1796
const uint32 * chars32
Definition: pg_wchar.h:421
uint8 b4_3_upper
Definition: pg_wchar.h:451
int pg_valid_server_encoding(const char *name)
Definition: encnames.c:501
int maxmblen
Definition: pg_wchar.h:369
uint8 b4_4_lower
Definition: pg_wchar.h:452
char bool
Definition: c.h:202
char * pg_server_to_client(const char *s, int len)
Definition: mbutils.c:634
uint8 b3_1_lower
Definition: pg_wchar.h:437
int pg_char_and_wchar_strncmp(const char *s1, const pg_wchar *s2, size_t n)
Definition: wstrncmp.c:55
mbdisplaylen_converter dsplen
Definition: pg_wchar.h:367
void report_untranslatable_char(int src_encoding, int dest_encoding, const char *mbstr, int len) pg_attribute_noreturn()
Definition: wchar.c:2027
mbcharacter_incrementer pg_database_encoding_character_incrementer(void)
Definition: wchar.c:1842
uint32(* utf_local_conversion_func)(uint32 code)
Definition: pg_wchar.h:482
void report_invalid_encoding(int encoding, const char *mbstr, int len) pg_attribute_noreturn()
Definition: wchar.c:1995
unsigned short uint16
Definition: c.h:267
uint8 b3_1_upper
Definition: pg_wchar.h:438
char * s1
unsigned short BIG5toCNS(unsigned short big5, unsigned char *lc)
Definition: big5.c:292
struct pg_enc2name pg_enc2name
unsigned int pg_wchar
Definition: pg_wchar.h:25
char * c
void local2local(const unsigned char *l, unsigned char *p, int len, int src_encoding, int dest_encoding, const unsigned char *tab)
Definition: conv.c:30
struct pg_enc2gettext pg_enc2gettext
void SetDatabaseEncoding(int encoding)
Definition: mbutils.c:909
uint8 b4_2_upper
Definition: pg_wchar.h:449
int pg_encoding_mbcliplen(int encoding, const char *mbstr, int len, int limit)
Definition: mbutils.c:841
int(* mb2wchar_with_len_converter)(const unsigned char *from, pg_wchar *to, int len)
Definition: pg_wchar.h:344
unsigned int uint32
Definition: c.h:268
const char * name
Definition: pg_wchar.h:315
uint8 b3_2_upper
Definition: pg_wchar.h:440
int SetClientEncoding(int encoding)
Definition: mbutils.c:212
bool pg_verifymbstr(const char *mbstr, int len, bool noError)
Definition: wchar.c:1866
unsigned int pg_wchar
Definition: mbprint.c:31
void mic2latin_with_table(const unsigned char *mic, unsigned char *p, int len, int lc, int encoding, const unsigned char *tab)
Definition: conv.c:241
int pg_utf_mblen(const unsigned char *)
Definition: wchar.c:541
void UtfToLocal(const unsigned char *utf, int len, unsigned char *iso, const pg_mb_radix_tree *map, const pg_utf_to_local_combined *cmap, int cmapsize, utf_local_conversion_func conv_func, int encoding)
Definition: conv.c:474
int pg_mbstrlen_with_len(const char *mbstr, int len)
Definition: mbutils.c:805
int pg_char_and_wchar_strcmp(const char *s1, const pg_wchar *s2)
Definition: wstrcmp.c:41
int pg_verify_mbstr_len(int encoding, const char *mbstr, int len, bool noError)
Definition: wchar.c:1894
int PrepareClientEncoding(int encoding)
Definition: mbutils.c:114
const char * pg_get_client_encoding_name(void)
Definition: mbutils.c:326
char * s2
int GetMessageEncoding(void)
Definition: mbutils.c:1045
int pg_wchar_strncmp(const pg_wchar *s1, const pg_wchar *s2, size_t n)
Definition: wstrncmp.c:40
void SetMessageEncoding(int encoding)
Definition: mbutils.c:919
uint8 b3_2_lower
Definition: pg_wchar.h:439
char * pg_client_to_server(const char *s, int len)
Definition: mbutils.c:556
wchar2mb_with_len_converter wchar2mb_with_len
Definition: pg_wchar.h:364
pg_enc encoding
Definition: pg_wchar.h:316
uint8 b3_3_upper
Definition: pg_wchar.h:442
static char * encoding
Definition: initdb.c:122
bool(* mbcharacter_incrementer)(unsigned char *mbstr, int len)
Definition: pg_wchar.h:356
int pg_mule_mblen(const unsigned char *mbstr)
Definition: wchar.c:866
const pg_enc2name pg_enc2name_tbl[]
Definition: encnames.c:307
uint8 b2_2_lower
Definition: pg_wchar.h:432
uint8 b4_1_upper
Definition: pg_wchar.h:447
int pg_wchar2mb(const pg_wchar *from, char *to)
Definition: mbutils.c:749
pg_enc
Definition: pg_wchar.h:236
int GetDatabaseEncoding(void)
Definition: mbutils.c:1015
int(* mblen_converter)(const unsigned char *mbstr)
Definition: pg_wchar.h:352
int pg_wchar2mb_with_len(const pg_wchar *from, char *to, int len)
Definition: mbutils.c:756
int pg_mblen(const char *mbstr)
Definition: mbutils.c:771
const char * name
Definition: encode.c:521
void mic2latin(const unsigned char *mic, unsigned char *p, int len, int lc, int encoding)
Definition: conv.c:99
int pg_dsplen(const char *mbstr)
Definition: mbutils.c:778
pg_wchar utf8_to_unicode(const unsigned char *c)
Definition: wchar.c:714
uint8 b4_3_lower
Definition: pg_wchar.h:450
int(* wchar2mb_with_len_converter)(const pg_wchar *from, unsigned char *to, int len)
Definition: pg_wchar.h:348
#define pg_attribute_noreturn()
Definition: c.h:653
uint8 b2_1_upper
Definition: pg_wchar.h:431
int pg_encoding_verifymb(int encoding, const char *mbstr, int len)
Definition: wchar.c:1809
int pg_get_client_encoding(void)
Definition: mbutils.c:317
bool pg_verify_mbstr(int encoding, const char *mbstr, int len, bool noError)
Definition: wchar.c:1877
unsigned char * unicode_to_utf8(pg_wchar c, unsigned char *utf8string)
Definition: wchar.c:475
char * pg_server_to_any(const char *s, int len, int encoding)
Definition: mbutils.c:645
int pg_mbcharcliplen(const char *mbstr, int len, int imit)
Definition: mbutils.c:873
uint8 b3_3_lower
Definition: pg_wchar.h:441
uint8 b4_1_lower
Definition: pg_wchar.h:446
int pg_valid_client_encoding(const char *name)
Definition: encnames.c:487
int(* mbverifier)(const unsigned char *mbstr, int len)
Definition: pg_wchar.h:358
pg_enc encoding
Definition: pg_wchar.h:329
const char * get_encoding_name_for_icu(int encoding)
Definition: encnames.c:461
void check_encoding_conversion_args(int src_encoding, int dest_encoding, int len, int expected_src_encoding, int expected_dest_encoding)
Definition: wchar.c:1966
int pg_encoding_max_length(int encoding)
Definition: wchar.c:1820
mb2wchar_with_len_converter mb2wchar_with_len
Definition: pg_wchar.h:362
mbverifier mbverify
Definition: pg_wchar.h:368
mblen_converter mblen
Definition: pg_wchar.h:366
bool is_encoding_supported_by_icu(int encoding)
Definition: encnames.c:455
int pg_mbstrlen(const char *mbstr)
Definition: mbutils.c:785
int(* mbdisplaylen_converter)(const unsigned char *mbstr)
Definition: pg_wchar.h:354
void pg_ascii2mic(const unsigned char *l, unsigned char *p, int len)
Definition: conv.c:143
unsigned char * pg_do_encoding_conversion(unsigned char *src, int len, int src_encoding, int dest_encoding)
Definition: mbutils.c:337
int pg_encoding_wchar2mb_with_len(int encoding, const pg_wchar *from, char *to, int len)
Definition: mbutils.c:763