Header And Logo
|
00001 /*------------------------------------------------------------------------- 00002 * 00003 * pg_wchar.h 00004 * multibyte-character support 00005 * 00006 * Portions Copyright (c) 1996-2013, PostgreSQL Global Development Group 00007 * Portions Copyright (c) 1994, Regents of the University of California 00008 * 00009 * src/include/mb/pg_wchar.h 00010 * 00011 * NOTES 00012 * This is used both by the backend and by libpq, but should not be 00013 * included by libpq client programs. In particular, a libpq client 00014 * should not assume that the encoding IDs used by the version of libpq 00015 * it's linked to match up with the IDs declared here. 00016 * 00017 *------------------------------------------------------------------------- 00018 */ 00019 #ifndef PG_WCHAR_H 00020 #define PG_WCHAR_H 00021 00022 /* 00023 * The pg_wchar type 00024 */ 00025 typedef unsigned int pg_wchar; 00026 00027 /* 00028 * Maximum byte length of multibyte characters in any backend encoding 00029 */ 00030 #define MAX_MULTIBYTE_CHAR_LEN 4 00031 00032 /* 00033 * various definitions for EUC 00034 */ 00035 #define SS2 0x8e /* single shift 2 (JIS0201) */ 00036 #define SS3 0x8f /* single shift 3 (JIS0212) */ 00037 00038 /* 00039 * SJIS validation macros 00040 */ 00041 #define ISSJISHEAD(c) (((c) >= 0x81 && (c) <= 0x9f) || ((c) >= 0xe0 && (c) <= 0xfc)) 00042 #define ISSJISTAIL(c) (((c) >= 0x40 && (c) <= 0x7e) || ((c) >= 0x80 && (c) <= 0xfc)) 00043 00044 /*---------------------------------------------------- 00045 * MULE Internal Encoding (MIC) 00046 * 00047 * This encoding follows the design used within XEmacs; it is meant to 00048 * subsume many externally-defined character sets. Each character includes 00049 * identification of the character set it belongs to, so the encoding is 00050 * general but somewhat bulky. 00051 * 00052 * Currently PostgreSQL supports 5 types of MULE character sets: 00053 * 00054 * 1) 1-byte ASCII characters. Each byte is below 0x80. 00055 * 00056 * 2) "Official" single byte charsets such as ISO-8859-1 (Latin1). 00057 * Each MULE character consists of 2 bytes: LC1 + C1, where LC1 is 00058 * an identifier for the charset (in the range 0x81 to 0x8d) and C1 00059 * is the character code (in the range 0xa0 to 0xff). 00060 * 00061 * 3) "Private" single byte charsets such as SISHENG. Each MULE 00062 * character consists of 3 bytes: LCPRV1 + LC12 + C1, where LCPRV1 00063 * is a private-charset flag, LC12 is an identifier for the charset, 00064 * and C1 is the character code (in the range 0xa0 to 0xff). 00065 * LCPRV1 is either 0x9a (if LC12 is in the range 0xa0 to 0xdf) 00066 * or 0x9b (if LC12 is in the range 0xe0 to 0xef). 00067 * 00068 * 4) "Official" multibyte charsets such as JIS X0208. Each MULE 00069 * character consists of 3 bytes: LC2 + C1 + C2, where LC2 is 00070 * an identifier for the charset (in the range 0x90 to 0x99) and C1 00071 * and C2 form the character code (each in the range 0xa0 to 0xff). 00072 * 00073 * 5) "Private" multibyte charsets such as CNS 11643-1992 Plane 3. 00074 * Each MULE character consists of 4 bytes: LCPRV2 + LC22 + C1 + C2, 00075 * where LCPRV2 is a private-charset flag, LC22 is an identifier for 00076 * the charset, and C1 and C2 form the character code (each in the range 00077 * 0xa0 to 0xff). LCPRV2 is either 0x9c (if LC22 is in the range 0xf0 00078 * to 0xf4) or 0x9d (if LC22 is in the range 0xf5 to 0xfe). 00079 * 00080 * "Official" encodings are those that have been assigned code numbers by 00081 * the XEmacs project; "private" encodings have Postgres-specific charset 00082 * identifiers. 00083 * 00084 * See the "XEmacs Internals Manual", available at http://www.xemacs.org, 00085 * for more details. Note that for historical reasons, Postgres' 00086 * private-charset flag values do not match what XEmacs says they should be, 00087 * so this isn't really exactly MULE (not that private charsets would be 00088 * interoperable anyway). 00089 * 00090 * Note that XEmacs's implementation is different from what emacs does. 00091 * We follow emacs's implementaion, rathter than XEmacs's. 00092 *---------------------------------------------------- 00093 */ 00094 00095 /* 00096 * Charset identifiers (also called "leading bytes" in the MULE documentation) 00097 */ 00098 00099 /* 00100 * Charset IDs for official single byte encodings (0x81-0x8e) 00101 */ 00102 #define LC_ISO8859_1 0x81 /* ISO8859 Latin 1 */ 00103 #define LC_ISO8859_2 0x82 /* ISO8859 Latin 2 */ 00104 #define LC_ISO8859_3 0x83 /* ISO8859 Latin 3 */ 00105 #define LC_ISO8859_4 0x84 /* ISO8859 Latin 4 */ 00106 #define LC_TIS620 0x85 /* Thai (not supported yet) */ 00107 #define LC_ISO8859_7 0x86 /* Greek (not supported yet) */ 00108 #define LC_ISO8859_6 0x87 /* Arabic (not supported yet) */ 00109 #define LC_ISO8859_8 0x88 /* Hebrew (not supported yet) */ 00110 #define LC_JISX0201K 0x89 /* Japanese 1 byte kana */ 00111 #define LC_JISX0201R 0x8a /* Japanese 1 byte Roman */ 00112 /* Note that 0x8b seems to be unused as of Emacs 20.7. 00113 * However, there might be a chance that 0x8b could be used 00114 * in later versions of Emacs. 00115 */ 00116 #define LC_KOI8_R 0x8b /* Cyrillic KOI8-R */ 00117 #define LC_ISO8859_5 0x8c /* ISO8859 Cyrillic */ 00118 #define LC_ISO8859_9 0x8d /* ISO8859 Latin 5 (not supported yet) */ 00119 #define LC_ISO8859_15 0x8e /* ISO8859 Latin 15 (not supported yet) */ 00120 /* #define CONTROL_1 0x8f control characters (unused) */ 00121 00122 /* Is a leading byte for "official" single byte encodings? */ 00123 #define IS_LC1(c) ((unsigned char)(c) >= 0x81 && (unsigned char)(c) <= 0x8d) 00124 00125 /* 00126 * Charset IDs for official multibyte encodings (0x90-0x99) 00127 * 0x9a-0x9d are free. 0x9e and 0x9f are reserved. 00128 */ 00129 #define LC_JISX0208_1978 0x90 /* Japanese Kanji, old JIS (not supported) */ 00130 #define LC_GB2312_80 0x91 /* Chinese */ 00131 #define LC_JISX0208 0x92 /* Japanese Kanji (JIS X 0208) */ 00132 #define LC_KS5601 0x93 /* Korean */ 00133 #define LC_JISX0212 0x94 /* Japanese Kanji (JIS X 0212) */ 00134 #define LC_CNS11643_1 0x95 /* CNS 11643-1992 Plane 1 */ 00135 #define LC_CNS11643_2 0x96 /* CNS 11643-1992 Plane 2 */ 00136 #define LC_JISX0213_1 0x97 /* Japanese Kanji (JIS X 0213 Plane 1) (not supported) */ 00137 #define LC_BIG5_1 0x98 /* Plane 1 Chinese traditional (not supported) */ 00138 #define LC_BIG5_2 0x99 /* Plane 1 Chinese traditional (not supported) */ 00139 00140 /* Is a leading byte for "official" multibyte encodings? */ 00141 #define IS_LC2(c) ((unsigned char)(c) >= 0x90 && (unsigned char)(c) <= 0x99) 00142 00143 /* 00144 * Postgres-specific prefix bytes for "private" single byte encodings 00145 * (According to the MULE docs, we should be using 0x9e for this) 00146 */ 00147 #define LCPRV1_A 0x9a 00148 #define LCPRV1_B 0x9b 00149 #define IS_LCPRV1(c) ((unsigned char)(c) == LCPRV1_A || (unsigned char)(c) == LCPRV1_B) 00150 #define IS_LCPRV1_A_RANGE(c) \ 00151 ((unsigned char)(c) >= 0xa0 && (unsigned char)(c) <= 0xdf) 00152 #define IS_LCPRV1_B_RANGE(c) \ 00153 ((unsigned char)(c) >= 0xe0 && (unsigned char)(c) <= 0xef) 00154 00155 /* 00156 * Postgres-specific prefix bytes for "private" multibyte encodings 00157 * (According to the MULE docs, we should be using 0x9f for this) 00158 */ 00159 #define LCPRV2_A 0x9c 00160 #define LCPRV2_B 0x9d 00161 #define IS_LCPRV2(c) ((unsigned char)(c) == LCPRV2_A || (unsigned char)(c) == LCPRV2_B) 00162 #define IS_LCPRV2_A_RANGE(c) \ 00163 ((unsigned char)(c) >= 0xf0 && (unsigned char)(c) <= 0xf4) 00164 #define IS_LCPRV2_B_RANGE(c) \ 00165 ((unsigned char)(c) >= 0xf5 && (unsigned char)(c) <= 0xfe) 00166 00167 /* 00168 * Charset IDs for private single byte encodings (0xa0-0xef) 00169 */ 00170 #define LC_SISHENG 0xa0 /* Chinese SiSheng characters for 00171 * PinYin/ZhuYin (not supported) */ 00172 #define LC_IPA 0xa1 /* IPA (International Phonetic Association) 00173 * (not supported) */ 00174 #define LC_VISCII_LOWER 0xa2 /* Vietnamese VISCII1.1 lower-case (not 00175 * supported) */ 00176 #define LC_VISCII_UPPER 0xa3 /* Vietnamese VISCII1.1 upper-case (not 00177 * supported) */ 00178 #define LC_ARABIC_DIGIT 0xa4 /* Arabic digit (not supported) */ 00179 #define LC_ARABIC_1_COLUMN 0xa5 /* Arabic 1-column (not supported) */ 00180 #define LC_ASCII_RIGHT_TO_LEFT 0xa6 /* ASCII (left half of ISO8859-1) with 00181 * right-to-left direction (not 00182 * supported) */ 00183 #define LC_LAO 0xa7 /* Lao characters (ISO10646 0E80..0EDF) 00184 * (not supported) */ 00185 #define LC_ARABIC_2_COLUMN 0xa8 /* Arabic 1-column (not supported) */ 00186 00187 /* 00188 * Charset IDs for private multibyte encodings (0xf0-0xff) 00189 */ 00190 #define LC_INDIAN_1_COLUMN 0xf0 /* Indian charset for 1-column width glyphs 00191 * (not supported) */ 00192 #define LC_TIBETAN_1_COLUMN 0xf1 /* Tibetan 1-column width glyphs 00193 * (not supported) */ 00194 #define LC_UNICODE_SUBSET_2 0xf2 /* Unicode characters of the range U+2500..U+33FF. 00195 * (not supported) */ 00196 #define LC_UNICODE_SUBSET_3 0xf3 /* Unicode characters of the range U+E000..U+FFFF. 00197 * (not supported) */ 00198 #define LC_UNICODE_SUBSET 0xf4 /* Unicode characters of the range U+0100..U+24FF. 00199 * (not supported) */ 00200 #define LC_ETHIOPIC 0xf5 /* Ethiopic characters (not supported) */ 00201 #define LC_CNS11643_3 0xf6 /* CNS 11643-1992 Plane 3 */ 00202 #define LC_CNS11643_4 0xf7 /* CNS 11643-1992 Plane 4 */ 00203 #define LC_CNS11643_5 0xf8 /* CNS 11643-1992 Plane 5 */ 00204 #define LC_CNS11643_6 0xf9 /* CNS 11643-1992 Plane 6 */ 00205 #define LC_CNS11643_7 0xfa /* CNS 11643-1992 Plane 7 */ 00206 #define LC_INDIAN_2_COLUMN 0xfb /* Indian charset for 2-column width glyphs 00207 * (not supported) */ 00208 #define LC_TIBETAN 0xfc /* Tibetan (not supported) */ 00209 /* #define FREE 0xfd free (unused) */ 00210 /* #define FREE 0xfe free (unused) */ 00211 /* #define FREE 0xff free (unused) */ 00212 00213 /*---------------------------------------------------- 00214 * end of MULE stuff 00215 *---------------------------------------------------- 00216 */ 00217 00218 /* 00219 * PostgreSQL encoding identifiers 00220 * 00221 * WARNING: the order of this enum must be same as order of entries 00222 * in the pg_enc2name_tbl[] array (in mb/encnames.c), and 00223 * in the pg_wchar_table[] array (in mb/wchar.c)! 00224 * 00225 * If you add some encoding don't forget to check 00226 * PG_ENCODING_BE_LAST macro. 00227 * 00228 * PG_SQL_ASCII is default encoding and must be = 0. 00229 * 00230 * XXX We must avoid renumbering any backend encoding until libpq's major 00231 * version number is increased beyond 5; it turns out that the backend 00232 * encoding IDs are effectively part of libpq's ABI as far as 8.2 initdb and 00233 * psql are concerned. 00234 */ 00235 typedef enum pg_enc 00236 { 00237 PG_SQL_ASCII = 0, /* SQL/ASCII */ 00238 PG_EUC_JP, /* EUC for Japanese */ 00239 PG_EUC_CN, /* EUC for Chinese */ 00240 PG_EUC_KR, /* EUC for Korean */ 00241 PG_EUC_TW, /* EUC for Taiwan */ 00242 PG_EUC_JIS_2004, /* EUC-JIS-2004 */ 00243 PG_UTF8, /* Unicode UTF8 */ 00244 PG_MULE_INTERNAL, /* Mule internal code */ 00245 PG_LATIN1, /* ISO-8859-1 Latin 1 */ 00246 PG_LATIN2, /* ISO-8859-2 Latin 2 */ 00247 PG_LATIN3, /* ISO-8859-3 Latin 3 */ 00248 PG_LATIN4, /* ISO-8859-4 Latin 4 */ 00249 PG_LATIN5, /* ISO-8859-9 Latin 5 */ 00250 PG_LATIN6, /* ISO-8859-10 Latin6 */ 00251 PG_LATIN7, /* ISO-8859-13 Latin7 */ 00252 PG_LATIN8, /* ISO-8859-14 Latin8 */ 00253 PG_LATIN9, /* ISO-8859-15 Latin9 */ 00254 PG_LATIN10, /* ISO-8859-16 Latin10 */ 00255 PG_WIN1256, /* windows-1256 */ 00256 PG_WIN1258, /* Windows-1258 */ 00257 PG_WIN866, /* (MS-DOS CP866) */ 00258 PG_WIN874, /* windows-874 */ 00259 PG_KOI8R, /* KOI8-R */ 00260 PG_WIN1251, /* windows-1251 */ 00261 PG_WIN1252, /* windows-1252 */ 00262 PG_ISO_8859_5, /* ISO-8859-5 */ 00263 PG_ISO_8859_6, /* ISO-8859-6 */ 00264 PG_ISO_8859_7, /* ISO-8859-7 */ 00265 PG_ISO_8859_8, /* ISO-8859-8 */ 00266 PG_WIN1250, /* windows-1250 */ 00267 PG_WIN1253, /* windows-1253 */ 00268 PG_WIN1254, /* windows-1254 */ 00269 PG_WIN1255, /* windows-1255 */ 00270 PG_WIN1257, /* windows-1257 */ 00271 PG_KOI8U, /* KOI8-U */ 00272 /* PG_ENCODING_BE_LAST points to the above entry */ 00273 00274 /* followings are for client encoding only */ 00275 PG_SJIS, /* Shift JIS (Winindows-932) */ 00276 PG_BIG5, /* Big5 (Windows-950) */ 00277 PG_GBK, /* GBK (Windows-936) */ 00278 PG_UHC, /* UHC (Windows-949) */ 00279 PG_GB18030, /* GB18030 */ 00280 PG_JOHAB, /* EUC for Korean JOHAB */ 00281 PG_SHIFT_JIS_2004, /* Shift-JIS-2004 */ 00282 _PG_LAST_ENCODING_ /* mark only */ 00283 00284 } pg_enc; 00285 00286 #define PG_ENCODING_BE_LAST PG_KOI8U 00287 00288 /* 00289 * Please use these tests before access to pg_encconv_tbl[] 00290 * or to other places... 00291 */ 00292 #define PG_VALID_BE_ENCODING(_enc) \ 00293 ((_enc) >= 0 && (_enc) <= PG_ENCODING_BE_LAST) 00294 00295 #define PG_ENCODING_IS_CLIENT_ONLY(_enc) \ 00296 ((_enc) > PG_ENCODING_BE_LAST && (_enc) < _PG_LAST_ENCODING_) 00297 00298 #define PG_VALID_ENCODING(_enc) \ 00299 ((_enc) >= 0 && (_enc) < _PG_LAST_ENCODING_) 00300 00301 /* On FE are possible all encodings */ 00302 #define PG_VALID_FE_ENCODING(_enc) PG_VALID_ENCODING(_enc) 00303 00304 /* 00305 * Encoding names with all aliases 00306 */ 00307 typedef struct pg_encname 00308 { 00309 char *name; 00310 pg_enc encoding; 00311 } pg_encname; 00312 00313 extern pg_encname pg_encname_tbl[]; 00314 extern unsigned int pg_encname_tbl_sz; 00315 00316 /* 00317 * Careful: 00318 * 00319 * if (PG_VALID_ENCODING(encoding)) 00320 * pg_enc2name_tbl[ encoding ]; 00321 */ 00322 typedef struct pg_enc2name 00323 { 00324 char *name; 00325 pg_enc encoding; 00326 #ifdef WIN32 00327 unsigned codepage; /* codepage for WIN32 */ 00328 #endif 00329 } pg_enc2name; 00330 00331 extern pg_enc2name pg_enc2name_tbl[]; 00332 00333 /* 00334 * Encoding names for gettext 00335 */ 00336 typedef struct pg_enc2gettext 00337 { 00338 pg_enc encoding; 00339 const char *name; 00340 } pg_enc2gettext; 00341 00342 extern pg_enc2gettext pg_enc2gettext_tbl[]; 00343 00344 /* 00345 * pg_wchar stuff 00346 */ 00347 typedef int (*mb2wchar_with_len_converter) (const unsigned char *from, 00348 pg_wchar *to, 00349 int len); 00350 00351 typedef int (*wchar2mb_with_len_converter) (const pg_wchar *from, 00352 unsigned char *to, 00353 int len); 00354 00355 typedef int (*mblen_converter) (const unsigned char *mbstr); 00356 00357 typedef int (*mbdisplaylen_converter) (const unsigned char *mbstr); 00358 00359 typedef bool (*mbcharacter_incrementer) (unsigned char *mbstr, int len); 00360 00361 typedef int (*mbverifier) (const unsigned char *mbstr, int len); 00362 00363 typedef struct 00364 { 00365 mb2wchar_with_len_converter mb2wchar_with_len; /* convert a multibyte 00366 * string to a wchar */ 00367 wchar2mb_with_len_converter wchar2mb_with_len; /* convert a wchar 00368 * string to a multibyte */ 00369 mblen_converter mblen; /* get byte length of a char */ 00370 mbdisplaylen_converter dsplen; /* get display width of a char */ 00371 mbverifier mbverify; /* verify multibyte sequence */ 00372 int maxmblen; /* max bytes for a char in this encoding */ 00373 } pg_wchar_tbl; 00374 00375 extern pg_wchar_tbl pg_wchar_table[]; 00376 00377 /* 00378 * UTF-8 to local code conversion map 00379 * Note that we limit the max length of UTF-8 to 4 bytes, 00380 * which is UCS-4 00010000-001FFFFF range. 00381 */ 00382 typedef struct 00383 { 00384 uint32 utf; /* UTF-8 */ 00385 uint32 code; /* local code */ 00386 } pg_utf_to_local; 00387 00388 /* 00389 * local code to UTF-8 conversion map 00390 */ 00391 typedef struct 00392 { 00393 uint32 code; /* local code */ 00394 uint32 utf; /* UTF-8 */ 00395 } pg_local_to_utf; 00396 00397 /* 00398 * UTF-8 to local code conversion map(combined characters) 00399 */ 00400 typedef struct 00401 { 00402 uint32 utf1; /* UTF-8 code 1 */ 00403 uint32 utf2; /* UTF-8 code 2 */ 00404 uint32 code; /* local code */ 00405 } pg_utf_to_local_combined; 00406 00407 /* 00408 * local code to UTF-8 conversion map(combined characters) 00409 */ 00410 typedef struct 00411 { 00412 uint32 code; /* local code */ 00413 uint32 utf1; /* UTF-8 code 1 */ 00414 uint32 utf2; /* UTF-8 code 2 */ 00415 } pg_local_to_utf_combined; 00416 00417 /* 00418 * Support macro for encoding conversion functions to validate their 00419 * arguments. (This could be made more compact if we included fmgr.h 00420 * here, but we don't want to do that because this header file is also 00421 * used by frontends.) 00422 */ 00423 #define CHECK_ENCODING_CONVERSION_ARGS(srcencoding,destencoding) \ 00424 check_encoding_conversion_args(PG_GETARG_INT32(0), \ 00425 PG_GETARG_INT32(1), \ 00426 PG_GETARG_INT32(4), \ 00427 (srcencoding), \ 00428 (destencoding)) 00429 00430 00431 /* 00432 * These functions are considered part of libpq's exported API and 00433 * are also declared in libpq-fe.h. 00434 */ 00435 extern int pg_char_to_encoding(const char *name); 00436 extern const char *pg_encoding_to_char(int encoding); 00437 extern int pg_valid_server_encoding_id(int encoding); 00438 00439 /* 00440 * Remaining functions are not considered part of libpq's API, though many 00441 * of them do exist inside libpq. 00442 */ 00443 extern pg_encname *pg_char_to_encname_struct(const char *name); 00444 00445 extern int pg_mb2wchar(const char *from, pg_wchar *to); 00446 extern int pg_mb2wchar_with_len(const char *from, pg_wchar *to, int len); 00447 extern int pg_encoding_mb2wchar_with_len(int encoding, 00448 const char *from, pg_wchar *to, int len); 00449 extern int pg_wchar2mb(const pg_wchar *from, char *to); 00450 extern int pg_wchar2mb_with_len(const pg_wchar *from, char *to, int len); 00451 extern int pg_encoding_wchar2mb_with_len(int encoding, 00452 const pg_wchar *from, char *to, int len); 00453 extern int pg_char_and_wchar_strcmp(const char *s1, const pg_wchar *s2); 00454 extern int pg_wchar_strncmp(const pg_wchar *s1, const pg_wchar *s2, size_t n); 00455 extern int pg_char_and_wchar_strncmp(const char *s1, const pg_wchar *s2, size_t n); 00456 extern size_t pg_wchar_strlen(const pg_wchar *wstr); 00457 extern int pg_mblen(const char *mbstr); 00458 extern int pg_dsplen(const char *mbstr); 00459 extern int pg_encoding_mblen(int encoding, const char *mbstr); 00460 extern int pg_encoding_dsplen(int encoding, const char *mbstr); 00461 extern int pg_encoding_verifymb(int encoding, const char *mbstr, int len); 00462 extern int pg_mule_mblen(const unsigned char *mbstr); 00463 extern int pg_mic_mblen(const unsigned char *mbstr); 00464 extern int pg_mbstrlen(const char *mbstr); 00465 extern int pg_mbstrlen_with_len(const char *mbstr, int len); 00466 extern int pg_mbcliplen(const char *mbstr, int len, int limit); 00467 extern int pg_encoding_mbcliplen(int encoding, const char *mbstr, 00468 int len, int limit); 00469 extern int pg_mbcharcliplen(const char *mbstr, int len, int imit); 00470 extern int pg_encoding_max_length(int encoding); 00471 extern int pg_database_encoding_max_length(void); 00472 extern mbcharacter_incrementer pg_database_encoding_character_incrementer(void); 00473 00474 extern int PrepareClientEncoding(int encoding); 00475 extern int SetClientEncoding(int encoding); 00476 extern void InitializeClientEncoding(void); 00477 extern int pg_get_client_encoding(void); 00478 extern const char *pg_get_client_encoding_name(void); 00479 00480 extern void SetDatabaseEncoding(int encoding); 00481 extern int GetDatabaseEncoding(void); 00482 extern const char *GetDatabaseEncodingName(void); 00483 extern int GetPlatformEncoding(void); 00484 extern void pg_bind_textdomain_codeset(const char *domainname); 00485 00486 extern int pg_valid_client_encoding(const char *name); 00487 extern int pg_valid_server_encoding(const char *name); 00488 00489 extern unsigned char *unicode_to_utf8(pg_wchar c, unsigned char *utf8string); 00490 extern pg_wchar utf8_to_unicode(const unsigned char *c); 00491 extern int pg_utf_mblen(const unsigned char *); 00492 extern unsigned char *pg_do_encoding_conversion(unsigned char *src, int len, 00493 int src_encoding, 00494 int dest_encoding); 00495 00496 extern char *pg_client_to_server(const char *s, int len); 00497 extern char *pg_server_to_client(const char *s, int len); 00498 extern char *pg_any_to_server(const char *s, int len, int encoding); 00499 extern char *pg_server_to_any(const char *s, int len, int encoding); 00500 00501 extern unsigned short BIG5toCNS(unsigned short big5, unsigned char *lc); 00502 extern unsigned short CNStoBIG5(unsigned short cns, unsigned char lc); 00503 00504 extern void LocalToUtf(const unsigned char *iso, unsigned char *utf, 00505 const pg_local_to_utf *map, const pg_local_to_utf_combined *cmap, 00506 int size1, int size2, int encoding, int len); 00507 00508 extern void UtfToLocal(const unsigned char *utf, unsigned char *iso, 00509 const pg_utf_to_local *map, const pg_utf_to_local_combined *cmap, 00510 int size1, int size2, int encoding, int len); 00511 00512 extern bool pg_verifymbstr(const char *mbstr, int len, bool noError); 00513 extern bool pg_verify_mbstr(int encoding, const char *mbstr, int len, 00514 bool noError); 00515 extern int pg_verify_mbstr_len(int encoding, const char *mbstr, int len, 00516 bool noError); 00517 00518 extern void check_encoding_conversion_args(int src_encoding, 00519 int dest_encoding, 00520 int len, 00521 int expected_src_encoding, 00522 int expected_dest_encoding); 00523 00524 extern void report_invalid_encoding(int encoding, const char *mbstr, int len) __attribute__((noreturn)); 00525 extern void report_untranslatable_char(int src_encoding, int dest_encoding, 00526 const char *mbstr, int len) __attribute__((noreturn)); 00527 00528 extern void pg_ascii2mic(const unsigned char *l, unsigned char *p, int len); 00529 extern void pg_mic2ascii(const unsigned char *mic, unsigned char *p, int len); 00530 extern void latin2mic(const unsigned char *l, unsigned char *p, int len, 00531 int lc, int encoding); 00532 extern void mic2latin(const unsigned char *mic, unsigned char *p, int len, 00533 int lc, int encoding); 00534 extern void latin2mic_with_table(const unsigned char *l, unsigned char *p, 00535 int len, int lc, int encoding, 00536 const unsigned char *tab); 00537 extern void mic2latin_with_table(const unsigned char *mic, unsigned char *p, 00538 int len, int lc, int encoding, 00539 const unsigned char *tab); 00540 00541 extern bool pg_utf8_islegal(const unsigned char *source, int length); 00542 00543 #ifdef WIN32 00544 extern WCHAR *pgwin32_toUTF16(const char *str, int len, int *utf16len); 00545 #endif 00546 00547 #endif /* PG_WCHAR_H */
1.7.1