/*-------------------------------------------------------------------------
*
* pg_wchar . h
* multibyte - character support
*
* Portions Copyright ( c ) 1996 - 2025 , PostgreSQL Global Development Group
* Portions Copyright ( c ) 1994 , Regents of the University of California
*
* src / include / mb / pg_wchar . h
*
* NOTES
* This is used both by the backend and by frontends , but should not be
* included by libpq client programs . In particular , a libpq client
* should not assume that the encoding IDs used by the version of libpq
* it ' s linked to match up with the IDs declared here .
* To help prevent mistakes , relevant functions that are exported by
* libpq have a physically different name when being referenced
* statically .
*
* - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
*/
#ifndef PG_WCHAR_H
#define PG_WCHAR_H
/*
* The pg_wchar type
*/
typedef unsigned int pg_wchar;
/*
* Maximum byte length of multibyte characters in any backend encoding
*/
#define MAX_MULTIBYTE_CHAR_LEN 4
/*
* various definitions for EUC
*/
#define SS2 0 x8e /* single shift 2 (JIS0201) */
#define SS3 0 x8f /* single shift 3 (JIS0212) */
/*
* SJIS validation macros
*/
#define ISSJISHEAD(c) (((c) >= 0 x81 && (c) <= 0 x9f) || ((c) >= 0 xe0 && (c) <= 0 xfc))
#define ISSJISTAIL(c) (((c) >= 0 x40 && (c) <= 0 x7e) || ((c) >= 0 x80 && (c) <= 0 xfc))
/*----------------------------------------------------
* MULE Internal Encoding ( MIC )
*
* This encoding follows the design used within XEmacs ; it is meant to
* subsume many externally - defined character sets . Each character includes
* identification of the character set it belongs to , so the encoding is
* general but somewhat bulky .
*
* Currently PostgreSQL supports 5 types of MULE character sets :
*
* 1 ) 1 - byte ASCII characters . Each byte is below 0 x80 .
*
* 2 ) " Official " single byte charsets such as ISO - 8859 - 1 ( Latin1 ) .
* Each MULE character consists of 2 bytes : LC1 + C1 , where LC1 is
* an identifier for the charset ( in the range 0 x81 to 0 x8d ) and C1
* is the character code ( in the range 0 xa0 to 0 xff ) .
*
* 3 ) " Private " single byte charsets such as SISHENG . Each MULE
* character consists of 3 bytes : LCPRV1 + LC12 + C1 , where LCPRV1
* is a private - charset flag , LC12 is an identifier for the charset ,
* and C1 is the character code ( in the range 0 xa0 to 0 xff ) .
* LCPRV1 is either 0 x9a ( if LC12 is in the range 0 xa0 to 0 xdf )
* or 0 x9b ( if LC12 is in the range 0 xe0 to 0 xef ) .
*
* 4 ) " Official " multibyte charsets such as JIS X0208 . Each MULE
* character consists of 3 bytes : LC2 + C1 + C2 , where LC2 is
* an identifier for the charset ( in the range 0 x90 to 0 x99 ) and C1
* and C2 form the character code ( each in the range 0 xa0 to 0 xff ) .
*
* 5 ) " Private " multibyte charsets such as CNS 11643 - 1992 Plane 3 .
* Each MULE character consists of 4 bytes : LCPRV2 + LC22 + C1 + C2 ,
* where LCPRV2 is a private - charset flag , LC22 is an identifier for
* the charset , and C1 and C2 form the character code ( each in the range
* 0 xa0 to 0 xff ) . LCPRV2 is either 0 x9c ( if LC22 is in the range 0 xf0
* to 0 xf4 ) or 0 x9d ( if LC22 is in the range 0 xf5 to 0 xfe ) .
*
* " Official " encodings are those that have been assigned code numbers by
* the XEmacs project ; " private " encodings have Postgres - specific charset
* identifiers .
*
* See the " XEmacs Internals Manual " , available at http : //www.xemacs.org,
* for more details . Note that for historical reasons , Postgres '
* private - charset flag values do not match what XEmacs says they should be ,
* so this isn ' t really exactly MULE ( not that private charsets would be
* interoperable anyway ) .
*
* Note that XEmacs ' s implementation is different from what emacs does .
* We follow emacs ' s implementation , rather than XEmacs ' s .
* - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
*/
/*
* Charset identifiers ( also called " leading bytes " in the MULE documentation )
*/
/*
* Charset IDs for official single byte encodings ( 0 x81 - 0 x8e )
*/
#define LC_ISO8859_1 0 x81 /* ISO8859 Latin 1 */
#define LC_ISO8859_2 0 x82 /* ISO8859 Latin 2 */
#define LC_ISO8859_3 0 x83 /* ISO8859 Latin 3 */
#define LC_ISO8859_4 0 x84 /* ISO8859 Latin 4 */
#define LC_TIS620 0 x85 /* Thai (not supported yet) */
#define LC_ISO8859_7 0 x86 /* Greek (not supported yet) */
#define LC_ISO8859_6 0 x87 /* Arabic (not supported yet) */
#define LC_ISO8859_8 0 x88 /* Hebrew (not supported yet) */
#define LC_JISX0201K 0 x89 /* Japanese 1 byte kana */
#define LC_JISX0201R 0 x8a /* Japanese 1 byte Roman */
/* Note that 0x8b seems to be unused as of Emacs 20.7.
* However , there might be a chance that 0 x8b could be used
* in later versions of Emacs .
*/
#define LC_KOI8_R 0 x8b /* Cyrillic KOI8-R */
#define LC_ISO8859_5 0 x8c /* ISO8859 Cyrillic */
#define LC_ISO8859_9 0 x8d /* ISO8859 Latin 5 (not supported yet) */
#define LC_ISO8859_15 0 x8e /* ISO8859 Latin 15 (not supported yet) */
/* #define CONTROL_1 0x8f control characters (unused) */
/* Is a leading byte for "official" single byte encodings? */
#define IS_LC1(c) ((unsigned char )(c) >= 0 x81 && (unsigned char )(c) <= 0 x8d)
/*
* Charset IDs for official multibyte encodings ( 0 x90 - 0 x99 )
* 0 x9a - 0 x9d are free . 0 x9e and 0 x9f are reserved .
*/
#define LC_JISX0208_1978 0 x90 /* Japanese Kanji, old JIS (not supported) */
#define LC_GB2312_80 0 x91 /* Chinese */
#define LC_JISX0208 0 x92 /* Japanese Kanji (JIS X 0208) */
#define LC_KS5601 0 x93 /* Korean */
#define LC_JISX0212 0 x94 /* Japanese Kanji (JIS X 0212) */
#define LC_CNS11643_1 0 x95 /* CNS 11643-1992 Plane 1 */
#define LC_CNS11643_2 0 x96 /* CNS 11643-1992 Plane 2 */
#define LC_JISX0213_1 0 x97 /* Japanese Kanji (JIS X 0213 Plane 1)
* (not supported) */
#define LC_BIG5_1 0 x98 /* Plane 1 Chinese traditional (not
* supported) */
#define LC_BIG5_2 0 x99 /* Plane 1 Chinese traditional (not
* supported) */
/* Is a leading byte for "official" multibyte encodings? */
#define IS_LC2(c) ((unsigned char )(c) >= 0 x90 && (unsigned char )(c) <= 0 x99)
/*
* Postgres - specific prefix bytes for " private " single byte encodings
* ( According to the MULE docs , we should be using 0 x9e for this )
*/
#define LCPRV1_A 0 x9a
#define LCPRV1_B 0 x9b
#define IS_LCPRV1(c) ((unsigned char )(c) == LCPRV1_A || (unsigned char )(c) == LCPRV1_B)
#define IS_LCPRV1_A_RANGE(c) \
((unsigned char )(c) >= 0 xa0 && (unsigned char )(c) <= 0 xdf)
#define IS_LCPRV1_B_RANGE(c) \
((unsigned char )(c) >= 0 xe0 && (unsigned char )(c) <= 0 xef)
/*
* Postgres - specific prefix bytes for " private " multibyte encodings
* ( According to the MULE docs , we should be using 0 x9f for this )
*/
#define LCPRV2_A 0 x9c
#define LCPRV2_B 0 x9d
#define IS_LCPRV2(c) ((unsigned char )(c) == LCPRV2_A || (unsigned char )(c) == LCPRV2_B)
#define IS_LCPRV2_A_RANGE(c) \
((unsigned char )(c) >= 0 xf0 && (unsigned char )(c) <= 0 xf4)
#define IS_LCPRV2_B_RANGE(c) \
((unsigned char )(c) >= 0 xf5 && (unsigned char )(c) <= 0 xfe)
/*
* Charset IDs for private single byte encodings ( 0 xa0 - 0 xef )
*/
#define LC_SISHENG 0 xa0 /* Chinese SiSheng characters for
* PinYin/ZhuYin (not supported) */
#define LC_IPA 0 xa1 /* IPA (International Phonetic
* Association) (not supported) */
#define LC_VISCII_LOWER 0 xa2 /* Vietnamese VISCII1.1 lower-case (not
* supported) */
#define LC_VISCII_UPPER 0 xa3 /* Vietnamese VISCII1.1 upper-case (not
* supported) */
#define LC_ARABIC_DIGIT 0 xa4 /* Arabic digit (not supported) */
#define LC_ARABIC_1_COLUMN 0 xa5 /* Arabic 1-column (not supported) */
#define LC_ASCII_RIGHT_TO_LEFT 0 xa6 /* ASCII (left half of ISO8859-1) with
* right - to - left direction ( not
* supported) */
#define LC_LAO 0 xa7 /* Lao characters (ISO10646 0E80..0EDF)
* (not supported) */
#define LC_ARABIC_2_COLUMN 0 xa8 /* Arabic 1-column (not supported) */
/*
* Charset IDs for private multibyte encodings ( 0 xf0 - 0 xff )
*/
#define LC_INDIAN_1_COLUMN 0 xf0 /* Indian charset for 1-column width
* glyphs (not supported) */
#define LC_TIBETAN_1_COLUMN 0 xf1 /* Tibetan 1-column width glyphs (not
* supported) */
#define LC_UNICODE_SUBSET_2 0 xf2 /* Unicode characters of the range
* U+2500..U+33FF. (not supported) */
#define LC_UNICODE_SUBSET_3 0 xf3 /* Unicode characters of the range
* U+E000..U+FFFF. (not supported) */
#define LC_UNICODE_SUBSET 0 xf4 /* Unicode characters of the range
* U+0100..U+24FF. (not supported) */
#define LC_ETHIOPIC 0 xf5 /* Ethiopic characters (not supported) */
#define LC_CNS11643_3 0 xf6 /* CNS 11643-1992 Plane 3 */
#define LC_CNS11643_4 0 xf7 /* CNS 11643-1992 Plane 4 */
#define LC_CNS11643_5 0 xf8 /* CNS 11643-1992 Plane 5 */
#define LC_CNS11643_6 0 xf9 /* CNS 11643-1992 Plane 6 */
#define LC_CNS11643_7 0 xfa /* CNS 11643-1992 Plane 7 */
#define LC_INDIAN_2_COLUMN 0 xfb /* Indian charset for 2-column width
* glyphs (not supported) */
#define LC_TIBETAN 0 xfc /* Tibetan (not supported) */
/* #define FREE 0xfd free (unused) */
/* #define FREE 0xfe free (unused) */
/* #define FREE 0xff free (unused) */
/*----------------------------------------------------
* end of MULE stuff
* - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
*/
/*
* PostgreSQL encoding identifiers
*
* WARNING : If you add some encoding don ' t forget to update
* the pg_enc2name_tbl [ ] array ( in src / common / encnames . c ) ,
* the pg_enc2gettext_tbl [ ] array ( in src / common / encnames . c ) and
* the pg_wchar_table [ ] array ( in src / common / wchar . c ) and to check
* PG_ENCODING_BE_LAST macro .
*
* PG_SQL_ASCII is default encoding and must be = 0 .
*
* XXX We must avoid renumbering any backend encoding until libpq ' s major
* version number is increased beyond 5 ; it turns out that the backend
* encoding IDs are effectively part of libpq ' s ABI as far as 8 . 2 initdb and
* psql are concerned .
*/
typedef enum pg_enc
{
PG_SQL_ASCII = 0 , /* SQL/ASCII */
PG_EUC_JP, /* EUC for Japanese */
PG_EUC_CN, /* EUC for Chinese */
PG_EUC_KR, /* EUC for Korean */
PG_EUC_TW, /* EUC for Taiwan */
PG_EUC_JIS_2004, /* EUC-JIS-2004 */
PG_UTF8, /* Unicode UTF8 */
PG_MULE_INTERNAL, /* Mule internal code */
PG_LATIN1, /* ISO-8859-1 Latin 1 */
PG_LATIN2, /* ISO-8859-2 Latin 2 */
PG_LATIN3, /* ISO-8859-3 Latin 3 */
PG_LATIN4, /* ISO-8859-4 Latin 4 */
PG_LATIN5, /* ISO-8859-9 Latin 5 */
PG_LATIN6, /* ISO-8859-10 Latin6 */
PG_LATIN7, /* ISO-8859-13 Latin7 */
PG_LATIN8, /* ISO-8859-14 Latin8 */
PG_LATIN9, /* ISO-8859-15 Latin9 */
PG_LATIN10, /* ISO-8859-16 Latin10 */
PG_WIN1256, /* windows-1256 */
PG_WIN1258, /* Windows-1258 */
PG_WIN866, /* (MS-DOS CP866) */
PG_WIN874, /* windows-874 */
PG_KOI8R, /* KOI8-R */
PG_WIN1251, /* windows-1251 */
PG_WIN1252, /* windows-1252 */
PG_ISO_8859_5, /* ISO-8859-5 */
PG_ISO_8859_6, /* ISO-8859-6 */
PG_ISO_8859_7, /* ISO-8859-7 */
PG_ISO_8859_8, /* ISO-8859-8 */
PG_WIN1250, /* windows-1250 */
PG_WIN1253, /* windows-1253 */
PG_WIN1254, /* windows-1254 */
PG_WIN1255, /* windows-1255 */
PG_WIN1257, /* windows-1257 */
PG_KOI8U, /* KOI8-U */
/* PG_ENCODING_BE_LAST points to the above entry */
/* followings are for client encoding only */
PG_SJIS, /* Shift JIS (Windows-932) */
PG_BIG5, /* Big5 (Windows-950) */
PG_GBK, /* GBK (Windows-936) */
PG_UHC, /* UHC (Windows-949) */
PG_GB18030, /* GB18030 */
PG_JOHAB, /* EUC for Korean JOHAB */
PG_SHIFT_JIS_2004, /* Shift-JIS-2004 */
_PG_LAST_ENCODING_ /* mark only */
} pg_enc;
#define PG_ENCODING_BE_LAST PG_KOI8U
/*
* Please use these tests before access to pg_enc2name_tbl [ ]
* or to other places . . .
*/
#define PG_VALID_BE_ENCODING(_enc) \
((_enc) >= 0 && (_enc) <= PG_ENCODING_BE_LAST)
#define PG_ENCODING_IS_CLIENT_ONLY(_enc) \
((_enc) > PG_ENCODING_BE_LAST && (_enc) < _PG_LAST_ENCODING_)
#define PG_VALID_ENCODING(_enc) \
((_enc) >= 0 && (_enc) < _PG_LAST_ENCODING_)
/* On FE are possible all encodings */
#define PG_VALID_FE_ENCODING(_enc) PG_VALID_ENCODING(_enc)
/*
* When converting strings between different encodings , we assume that space
* for converted result is 4 - to - 1 growth in the worst case . The rate for
* currently supported encoding pairs are within 3 ( SJIS JIS X0201 half width
* kana - > UTF8 is the worst case ) . So " 4 " should be enough for the moment .
*
* Note that this is not the same as the maximum character width in any
* particular encoding .
*/
#define MAX_CONVERSION_GROWTH 4
/*
* Maximum byte length of a string that ' s required in any encoding to convert
* at least one character to any other encoding . In other words , if you feed
* MAX_CONVERSION_INPUT_LENGTH bytes to any encoding conversion function , it
* is guaranteed to be able to convert something without needing more input
* ( assuming the input is valid ) .
*
* Currently , the maximum case is the conversion UTF8 - > SJIS JIS X0201 half
* width kana , where a pair of UTF - 8 characters is converted into a single
* SHIFT_JIS_2004 character ( the reverse of the worst case for
* MAX_CONVERSION_GROWTH ) . It needs 6 bytes of input . In theory , a
* user - defined conversion function might have more complicated cases , although
* for the reverse mapping you would probably also need to bump up
* MAX_CONVERSION_GROWTH . But there is no need to be stingy here , so make it
* generous .
*/
#define MAX_CONVERSION_INPUT_LENGTH 16
/*
* Maximum byte length of the string equivalent to any one Unicode code point ,
* in any backend encoding . The current value assumes that a 4 - byte UTF - 8
* character might expand by MAX_CONVERSION_GROWTH , which is a huge
* overestimate . But in current usage we don ' t allocate large multiples of
* this , so there ' s little point in being stingy .
*/
#define MAX_UNICODE_EQUIVALENT_STRING 16
/*
* Table for mapping an encoding number to official encoding name and
* possibly other subsidiary data . Be careful to check encoding number
* before accessing a table entry !
*
* if ( PG_VALID_ENCODING ( encoding ) )
* pg_enc2name_tbl [ encoding ] ;
*/
typedef struct pg_enc2name
{
const char *name;
pg_enc encoding;
#ifdef WIN32
unsigned codepage; /* codepage for WIN32 */
#endif
} pg_enc2name;
extern PGDLLIMPORT const pg_enc2name pg_enc2name_tbl[];
/*
* Encoding names for gettext
*/
extern PGDLLIMPORT const char *pg_enc2gettext_tbl[];
/*
* pg_wchar stuff
*/
typedef int (*mb2wchar_with_len_converter) (const unsigned char *from,
pg_wchar *to,
int len);
typedef int (*wchar2mb_with_len_converter) (const pg_wchar *from,
unsigned char *to,
int len);
typedef int (*mblen_converter) (const unsigned char *mbstr);
typedef int (*mbdisplaylen_converter) (const unsigned char *mbstr);
typedef bool (*mbcharacter_incrementer) (unsigned char *mbstr, int len);
typedef int (*mbchar_verifier) (const unsigned char *mbstr, int len);
typedef int (*mbstr_verifier) (const unsigned char *mbstr, int len);
typedef struct
{
mb2wchar_with_len_converter mb2wchar_with_len; /* convert a multibyte
* string to a wchar */
wchar2mb_with_len_converter wchar2mb_with_len; /* convert a wchar string
* to a multibyte */
mblen_converter mblen; /* get byte length of a char */
mbdisplaylen_converter dsplen; /* get display width of a char */
mbchar_verifier mbverifychar; /* verify multibyte character */
mbstr_verifier mbverifystr; /* verify multibyte string */
int maxmblen; /* max bytes for a char in this encoding */
} pg_wchar_tbl;
extern PGDLLIMPORT const pg_wchar_tbl pg_wchar_table[];
/*
* Data structures for conversions between UTF - 8 and other encodings
* ( UtfToLocal ( ) and LocalToUtf ( ) ) . In these data structures , characters of
* either encoding are represented by uint32 words ; hence we can only support
* characters up to 4 bytes long . For example , the byte sequence 0 xC2 0 x89
* would be represented by 0 x0000C289 , and 0 xE8 0 xA2 0 xB4 by 0 x00E8A2B4 .
*
* There are three possible ways a character can be mapped :
*
* 1 . Using a radix tree , from source to destination code .
* 2 . Using a sorted array of source - > destination code pairs . This
* method is used for " combining " characters . There are so few of
* them that building a radix tree would be wasteful .
* 3 . Using a conversion function .
*/
/*
* Radix tree for character conversion .
*
* Logically , this is actually four different radix trees , for 1 - byte ,
* 2 - byte , 3 - byte and 4 - byte inputs . The 1 - byte tree is a simple lookup
* table from source to target code . The 2 - byte tree consists of two levels :
* one lookup table for the first byte , where the value in the lookup table
* points to a lookup table for the second byte . And so on .
*
* Physically , all the trees are stored in one big array , in ' chars16 ' or
* ' chars32 ' , depending on the maximum value that needs to be represented . For
* each level in each tree , we also store lower and upper bound of allowed
* values - values outside those bounds are considered invalid , and are left
* out of the tables .
*
* In the intermediate levels of the trees , the values stored are offsets
* into the chars [ 16 | 32 ] array .
*
* In the beginning of the chars [ 16 | 32 ] array , there is always a number of
* zeros , so that you safely follow an index from an intermediate table
* without explicitly checking for a zero . Following a zero any number of
* times will always bring you to the dummy , all - zeros table in the
* beginning . This helps to shave some cycles when looking up values .
*/
typedef struct
{
/*
* Array containing all the values . Only one of chars16 or chars32 is
* used , depending on how wide the values we need to represent are .
*/
const uint16 *chars16;
const uint32 *chars32;
/* Radix tree for 1-byte inputs */
uint32 b1root; /* offset of table in the chars[16|32] array */
uint8 b1_lower; /* min allowed value for a single byte input */
uint8 b1_upper; /* max allowed value for a single byte input */
/* Radix tree for 2-byte inputs */
uint32 b2root; /* offset of 1st byte's table */
uint8 b2_1_lower; /* min/max allowed value for 1st input byte */
uint8 b2_1_upper;
uint8 b2_2_lower; /* min/max allowed value for 2nd input byte */
uint8 b2_2_upper;
/* Radix tree for 3-byte inputs */
uint32 b3root; /* offset of 1st byte's table */
uint8 b3_1_lower; /* min/max allowed value for 1st input byte */
uint8 b3_1_upper;
uint8 b3_2_lower; /* min/max allowed value for 2nd input byte */
uint8 b3_2_upper;
uint8 b3_3_lower; /* min/max allowed value for 3rd input byte */
uint8 b3_3_upper;
/* Radix tree for 4-byte inputs */
uint32 b4root; /* offset of 1st byte's table */
uint8 b4_1_lower; /* min/max allowed value for 1st input byte */
uint8 b4_1_upper;
uint8 b4_2_lower; /* min/max allowed value for 2nd input byte */
uint8 b4_2_upper;
uint8 b4_3_lower; /* min/max allowed value for 3rd input byte */
uint8 b4_3_upper;
uint8 b4_4_lower; /* min/max allowed value for 4th input byte */
uint8 b4_4_upper;
} pg_mb_radix_tree;
/*
* UTF - 8 to local code conversion map ( for combined characters )
*/
typedef struct
{
uint32 utf1; /* UTF-8 code 1 */
uint32 utf2; /* UTF-8 code 2 */
uint32 code; /* local code */
} pg_utf_to_local_combined;
/*
* local code to UTF - 8 conversion map ( for combined characters )
*/
typedef struct
{
uint32 code; /* local code */
uint32 utf1; /* UTF-8 code 1 */
uint32 utf2; /* UTF-8 code 2 */
} pg_local_to_utf_combined;
/*
* callback function for algorithmic encoding conversions ( in either direction )
*
* if function returns zero , it does not know how to convert the code
*/
typedef uint32 (*utf_local_conversion_func) (uint32 code);
/*
* Support macro for encoding conversion functions to validate their
* arguments . ( This could be made more compact if we included fmgr . h
* here , but we don ' t want to do that because this header file is also
* used by frontends . )
*/
#define CHECK_ENCODING_CONVERSION_ARGS(srcencoding,destencoding) \
check_encoding_conversion_args(PG_GETARG_INT32(0 ), \
PG_GETARG_INT32(1 ), \
PG_GETARG_INT32(4 ), \
(srcencoding), \
(destencoding))
/*
* Some handy functions for Unicode - specific tests .
*/
static inline bool
is_valid_unicode_codepoint(pg_wchar c)
{
return (c > 0 && c <= 0 x10FFFF);
}
static inline bool
is_utf16_surrogate_first(pg_wchar c)
{
return (c >= 0 xD800 && c <= 0 xDBFF);
}
static inline bool
is_utf16_surrogate_second(pg_wchar c)
{
return (c >= 0 xDC00 && c <= 0 xDFFF);
}
static inline pg_wchar
surrogate_pair_to_codepoint(pg_wchar first, pg_wchar second)
{
return ((first & 0 x3FF) << 10 ) + 0 x10000 + (second & 0 x3FF);
}
/*
* Convert a UTF - 8 character to a Unicode code point .
* This is a one - character version of pg_utf2wchar_with_len .
*
* No error checks here , c must point to a long - enough string .
*/
static inline pg_wchar
utf8_to_unicode(const unsigned char *c)
{
if ((*c & 0 x80) == 0 )
return (pg_wchar) c[0 ];
else if ((*c & 0 xe0) == 0 xc0)
return (pg_wchar) (((c[0 ] & 0 x1f) << 6 ) |
(c[1 ] & 0 x3f));
else if ((*c & 0 xf0) == 0 xe0)
return (pg_wchar) (((c[0 ] & 0 x0f) << 12 ) |
((c[1 ] & 0 x3f) << 6 ) |
(c[2 ] & 0 x3f));
else if ((*c & 0 xf8) == 0 xf0)
return (pg_wchar) (((c[0 ] & 0 x07) << 18 ) |
((c[1 ] & 0 x3f) << 12 ) |
((c[2 ] & 0 x3f) << 6 ) |
(c[3 ] & 0 x3f));
else
/* that is an invalid code on purpose */
return 0 xffffffff;
}
/*
* Map a Unicode code point to UTF - 8 . utf8string must have at least
* unicode_utf8len ( c ) bytes available .
*/
static inline unsigned char *
unicode_to_utf8(pg_wchar c, unsigned char *utf8string)
{
if (c <= 0 x7F)
{
utf8string[0 ] = c;
}
else if (c <= 0 x7FF)
{
utf8string[0 ] = 0 xC0 | ((c >> 6 ) & 0 x1F);
utf8string[1 ] = 0 x80 | (c & 0 x3F);
}
else if (c <= 0 xFFFF)
{
utf8string[0 ] = 0 xE0 | ((c >> 12 ) & 0 x0F);
utf8string[1 ] = 0 x80 | ((c >> 6 ) & 0 x3F);
utf8string[2 ] = 0 x80 | (c & 0 x3F);
}
else
{
utf8string[0 ] = 0 xF0 | ((c >> 18 ) & 0 x07);
utf8string[1 ] = 0 x80 | ((c >> 12 ) & 0 x3F);
utf8string[2 ] = 0 x80 | ((c >> 6 ) & 0 x3F);
utf8string[3 ] = 0 x80 | (c & 0 x3F);
}
return utf8string;
}
/*
* Number of bytes needed to represent the given char in UTF8 .
*/
static inline int
unicode_utf8len(pg_wchar c)
{
if (c <= 0 x7F)
return 1 ;
else if (c <= 0 x7FF)
return 2 ;
else if (c <= 0 xFFFF)
return 3 ;
else
return 4 ;
}
/*
* The functions in this list are exported by libpq , and we need to be sure
* that we know which calls are satisfied by libpq and which are satisfied
* by static linkage to libpgcommon . ( This is because we might be using a
* libpq . so that ' s of a different major version and has encoding IDs that
* differ from the current version ' s . ) The nominal function names are what
* are actually used in and exported by libpq , while the names exported by
* libpgcommon . a and libpgcommon_srv . a end in " _ private " .
*/
#if defined (USE_PRIVATE_ENCODING_FUNCS) || !defined (FRONTEND)
#define pg_char_to_encoding pg_char_to_encoding_private
#define pg_encoding_to_char pg_encoding_to_char_private
#define pg_valid_server_encoding pg_valid_server_encoding_private
#define pg_valid_server_encoding_id pg_valid_server_encoding_id_private
#define pg_utf_mblen pg_utf_mblen_private
#endif
/*
* These functions are considered part of libpq ' s exported API and
* are also declared in libpq - fe . h .
*/
extern int pg_char_to_encoding(const char *name);
extern const char *pg_encoding_to_char(int encoding);
extern int pg_valid_server_encoding_id(int encoding);
/*
* These functions are available to frontend code that links with libpgcommon
* ( in addition to the ones just above ) . The constant tables declared
* earlier in this file are also available from libpgcommon .
*/
extern void pg_encoding_set_invalid(int encoding, char *dst);
extern int pg_encoding_mblen(int encoding, const char *mbstr);
extern int pg_encoding_mblen_or_incomplete(int encoding, const char *mbstr,
size_t remaining);
extern int pg_encoding_mblen_bounded(int encoding, const char *mbstr);
extern int pg_encoding_dsplen(int encoding, const char *mbstr);
extern int pg_encoding_verifymbchar(int encoding, const char *mbstr, int len);
extern int pg_encoding_verifymbstr(int encoding, const char *mbstr, int len);
extern int pg_encoding_max_length(int encoding);
extern int pg_valid_client_encoding(const char *name);
extern int pg_valid_server_encoding(const char *name);
extern bool is_encoding_supported_by_icu(int encoding);
extern const char *get_encoding_name_for_icu(int encoding);
extern unsigned char *unicode_to_utf8(pg_wchar c, unsigned char *utf8string);
extern pg_wchar utf8_to_unicode(const unsigned char *c);
extern bool pg_utf8_islegal(const unsigned char *source, int length);
extern int pg_utf_mblen(const unsigned char *s);
extern int pg_mule_mblen(const unsigned char *s);
/*
* The remaining functions are backend - only .
*/
extern int pg_mb2wchar(const char *from, pg_wchar *to);
extern int pg_mb2wchar_with_len(const char *from, pg_wchar *to, int len);
extern int pg_encoding_mb2wchar_with_len(int encoding,
const char *from, pg_wchar *to, int len);
extern int pg_wchar2mb(const pg_wchar *from, char *to);
extern int pg_wchar2mb_with_len(const pg_wchar *from, char *to, int len);
extern int pg_encoding_wchar2mb_with_len(int encoding,
const pg_wchar *from, char *to, int len);
extern int pg_char_and_wchar_strcmp(const char *s1, const pg_wchar *s2);
extern int pg_wchar_strncmp(const pg_wchar *s1, const pg_wchar *s2, size_t n);
extern int pg_char_and_wchar_strncmp(const char *s1, const pg_wchar *s2, size_t n);
extern size_t pg_wchar_strlen(const pg_wchar *str);
extern int pg_mblen_cstr(const char *mbstr);
extern int pg_mblen_range(const char *mbstr, const char *end);
extern int pg_mblen_with_len(const char *mbstr, int limit);
extern int pg_mblen_unbounded(const char *mbstr);
/* deprecated */
extern int pg_mblen(const char *mbstr);
extern int pg_dsplen(const char *mbstr);
extern int pg_mbstrlen(const char *mbstr);
extern int pg_mbstrlen_with_len(const char *mbstr, int limit);
extern int pg_mbcliplen(const char *mbstr, int len, int limit);
extern int pg_encoding_mbcliplen(int encoding, const char *mbstr,
int len, int limit);
extern int pg_mbcharcliplen(const char *mbstr, int len, int limit);
extern int pg_database_encoding_max_length(void );
extern mbcharacter_incrementer pg_database_encoding_character_incrementer(void );
extern int PrepareClientEncoding(int encoding);
extern int SetClientEncoding(int encoding);
extern void InitializeClientEncoding(void );
extern int pg_get_client_encoding(void );
extern const char *pg_get_client_encoding_name(void );
extern void SetDatabaseEncoding(int encoding);
extern int GetDatabaseEncoding(void );
extern const char *GetDatabaseEncodingName(void );
extern void SetMessageEncoding(int encoding);
extern int GetMessageEncoding(void );
#ifdef ENABLE_NLS
extern int pg_bind_textdomain_codeset(const char *domainname);
#endif
extern unsigned char *pg_do_encoding_conversion(unsigned char *src, int len,
int src_encoding,
int dest_encoding);
extern int pg_do_encoding_conversion_buf(Oid proc,
int src_encoding,
int dest_encoding,
unsigned char *src, int srclen,
unsigned char *dest, int destlen,
bool noError);
extern char *pg_client_to_server(const char *s, int len);
extern char *pg_server_to_client(const char *s, int len);
extern char *pg_any_to_server(const char *s, int len, int encoding);
extern char *pg_server_to_any(const char *s, int len, int encoding);
extern void pg_unicode_to_server(pg_wchar c, unsigned char *s);
extern bool pg_unicode_to_server_noerror(pg_wchar c, unsigned char *s);
extern unsigned short BIG5toCNS(unsigned short big5, unsigned char *lc);
extern unsigned short CNStoBIG5(unsigned short cns, unsigned char lc);
extern int 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, bool noError);
extern int 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, bool noError);
extern bool pg_verifymbstr(const char *mbstr, int len, bool noError);
extern bool pg_verify_mbstr(int encoding, const char *mbstr, int len,
bool noError);
extern int pg_verify_mbstr_len(int encoding, const char *mbstr, int len,
bool noError);
extern void check_encoding_conversion_args(int src_encoding,
int dest_encoding,
int len,
int expected_src_encoding,
int expected_dest_encoding);
pg_noreturn extern void report_invalid_encoding(int encoding, const char *mbstr, int len);
pg_noreturn extern void report_untranslatable_char(int src_encoding, int dest_encoding,
const char *mbstr, int len);
extern int local2local(const unsigned char *l, unsigned char *p, int len,
int src_encoding, int dest_encoding,
const unsigned char *tab, bool noError);
extern int latin2mic(const unsigned char *l, unsigned char *p, int len,
int lc, int encoding, bool noError);
extern int mic2latin(const unsigned char *mic, unsigned char *p, int len,
int lc, int encoding, bool noError);
extern int latin2mic_with_table(const unsigned char *l, unsigned char *p,
int len, int lc, int encoding,
const unsigned char *tab, bool noError);
extern int mic2latin_with_table(const unsigned char *mic, unsigned char *p,
int len, int lc, int encoding,
const unsigned char *tab, bool noError);
#ifdef WIN32
extern WCHAR *pgwin32_message_to_UTF16(const char *str, int len, int *utf16len);
#endif
#endif /* PG_WCHAR_H */
Messung V0.5 in Prozent C=80 H=100 G=90
¤ Dauer der Verarbeitung: 0.23 Sekunden
(vorverarbeitet am 2026-08-05)
¤
*© Formatika GbR, Deutschland