diff options
Diffstat (limited to 'lib/mbchar.h')
| -rw-r--r-- | lib/mbchar.h | 165 |
1 files changed, 93 insertions, 72 deletions
diff --git a/lib/mbchar.h b/lib/mbchar.h index 5ca8df61..c06ef11b 100644 --- a/lib/mbchar.h +++ b/lib/mbchar.h @@ -1,5 +1,5 @@ /* Multibyte character data type. - Copyright (C) 2001, 2005-2007, 2009-2022 Free Software Foundation, Inc. + Copyright (C) 2001, 2005-2007, 2009-2024 Free Software Foundation, Inc. This file is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as @@ -17,10 +17,10 @@ /* Written by Bruno Haible <bruno@clisp.org>. */ /* A multibyte character is a short subsequence of a char* string, - representing a single wide character. + representing a single 32-bit wide character. - We use multibyte characters instead of wide characters because of - the following goals: + We use multibyte characters instead of 32-bit wide characters because + of the following goals: 1) correct multibyte handling, i.e. operate according to the LC_CTYPE locale, 2) ease of maintenance, i.e. the maintainer needs not know all details @@ -28,8 +28,7 @@ 3) don't fail grossly if the input is not in the encoding set by the locale, because often different encodings are in use in the same countries (ISO-8859-1/UTF-8, EUC-JP/Shift_JIS, ...), - 4) fast in the case of ASCII characters, - 5) portability, i.e. don't make unportable assumptions about wchar_t. + 4) fast in the case of ASCII characters. Multibyte characters are only accessed through the mb* macros. @@ -108,6 +107,7 @@ mb_setascii (&mbc, sc) assigns the standard ASCII character sc to mbc. + (Only available if the 'mbfile' module is in use.) mb_copy (&destmbc, &srcmbc) copies srcmbc to destmbc. @@ -144,27 +144,32 @@ #ifndef _MBCHAR_H #define _MBCHAR_H 1 -#include <string.h> -#include <wchar.h> -#include <wctype.h> - -#ifndef _GL_INLINE_HEADER_BEGIN +/* This file uses _GL_INLINE_HEADER_BEGIN, _GL_INLINE. */ +#if !_GL_CONFIG_H_INCLUDED #error "Please include config.h first." #endif + +#include <string.h> +#include <uchar.h> + _GL_INLINE_HEADER_BEGIN #ifndef MBCHAR_INLINE # define MBCHAR_INLINE _GL_INLINE #endif -#define MBCHAR_BUF_SIZE 24 +/* The longest multibyte characters, nowadays, are 4 bytes long. + Regardless of the values of MB_CUR_MAX and MB_LEN_MAX. */ +#define MBCHAR_BUF_SIZE 4 struct mbchar { const char *ptr; /* pointer to current character */ size_t bytes; /* number of bytes of current character, > 0 */ - bool wc_valid; /* true if wc is a valid wide character */ - wchar_t wc; /* if wc_valid: the current character */ + bool wc_valid; /* true if wc is a valid 32-bit wide character */ + char32_t wc; /* if wc_valid: the current character */ +#if defined GNULIB_MBFILE char buf[MBCHAR_BUF_SIZE]; /* room for the bytes, used for file input only */ +#endif }; /* EOF (not a real character) is represented with bytes = 0 and @@ -182,7 +187,7 @@ typedef struct mbchar mbchar_t; #define mb_cmp(mbc1, mbc2) \ ((mbc1).wc_valid \ ? ((mbc2).wc_valid \ - ? (int) (mbc1).wc - (int) (mbc2).wc \ + ? _GL_CMP ((mbc1).wc, (mbc2).wc) \ : -1) \ : ((mbc2).wc_valid \ ? 1 \ @@ -194,7 +199,7 @@ typedef struct mbchar mbchar_t; #define mb_casecmp(mbc1, mbc2) \ ((mbc1).wc_valid \ ? ((mbc2).wc_valid \ - ? (int) towlower ((mbc1).wc) - (int) towlower ((mbc2).wc) \ + ? _GL_CMP (c32tolower ((mbc1).wc), c32tolower ((mbc2).wc)) \ : -1) \ : ((mbc2).wc_valid \ ? 1 \ @@ -210,25 +215,25 @@ typedef struct mbchar mbchar_t; && memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) == 0) #define mb_caseequal(mbc1, mbc2) \ ((mbc1).wc_valid && (mbc2).wc_valid \ - ? towlower ((mbc1).wc) == towlower ((mbc2).wc) \ + ? c32tolower ((mbc1).wc) == c32tolower ((mbc2).wc) \ : (mbc1).bytes == (mbc2).bytes \ && memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) == 0) /* <ctype.h>, <wctype.h> classification. */ #define mb_isascii(mbc) \ ((mbc).wc_valid && (mbc).wc >= 0 && (mbc).wc <= 127) -#define mb_isalnum(mbc) ((mbc).wc_valid && iswalnum ((mbc).wc)) -#define mb_isalpha(mbc) ((mbc).wc_valid && iswalpha ((mbc).wc)) -#define mb_isblank(mbc) ((mbc).wc_valid && iswblank ((mbc).wc)) -#define mb_iscntrl(mbc) ((mbc).wc_valid && iswcntrl ((mbc).wc)) -#define mb_isdigit(mbc) ((mbc).wc_valid && iswdigit ((mbc).wc)) -#define mb_isgraph(mbc) ((mbc).wc_valid && iswgraph ((mbc).wc)) -#define mb_islower(mbc) ((mbc).wc_valid && iswlower ((mbc).wc)) -#define mb_isprint(mbc) ((mbc).wc_valid && iswprint ((mbc).wc)) -#define mb_ispunct(mbc) ((mbc).wc_valid && iswpunct ((mbc).wc)) -#define mb_isspace(mbc) ((mbc).wc_valid && iswspace ((mbc).wc)) -#define mb_isupper(mbc) ((mbc).wc_valid && iswupper ((mbc).wc)) -#define mb_isxdigit(mbc) ((mbc).wc_valid && iswxdigit ((mbc).wc)) +#define mb_isalnum(mbc) ((mbc).wc_valid && c32isalnum ((mbc).wc)) +#define mb_isalpha(mbc) ((mbc).wc_valid && c32isalpha ((mbc).wc)) +#define mb_isblank(mbc) ((mbc).wc_valid && c32isblank ((mbc).wc)) +#define mb_iscntrl(mbc) ((mbc).wc_valid && c32iscntrl ((mbc).wc)) +#define mb_isdigit(mbc) ((mbc).wc_valid && c32isdigit ((mbc).wc)) +#define mb_isgraph(mbc) ((mbc).wc_valid && c32isgraph ((mbc).wc)) +#define mb_islower(mbc) ((mbc).wc_valid && c32islower ((mbc).wc)) +#define mb_isprint(mbc) ((mbc).wc_valid && c32isprint ((mbc).wc)) +#define mb_ispunct(mbc) ((mbc).wc_valid && c32ispunct ((mbc).wc)) +#define mb_isspace(mbc) ((mbc).wc_valid && c32isspace ((mbc).wc)) +#define mb_isupper(mbc) ((mbc).wc_valid && c32isupper ((mbc).wc)) +#define mb_isxdigit(mbc) ((mbc).wc_valid && c32isxdigit ((mbc).wc)) /* Extra <wchar.h> function. */ @@ -236,12 +241,12 @@ typedef struct mbchar mbchar_t; #define MB_UNPRINTABLE_WIDTH 1 MBCHAR_INLINE int -mb_width_aux (wint_t wc) +mb_width_aux (char32_t wc) { - int w = wcwidth (wc); + int w = c32width (wc); /* For unprintable characters, arbitrarily return 0 for control characters and MB_UNPRINTABLE_WIDTH otherwise. */ - return (w >= 0 ? w : iswcntrl (wc) ? 0 : MB_UNPRINTABLE_WIDTH); + return (w >= 0 ? w : c32iscntrl (wc) ? 0 : MB_UNPRINTABLE_WIDTH); } #define mb_width(mbc) \ @@ -250,21 +255,25 @@ mb_width_aux (wint_t wc) /* Output. */ #define mb_putc(mbc, stream) fwrite ((mbc).ptr, 1, (mbc).bytes, (stream)) +#if defined GNULIB_MBFILE /* Assignment. */ -#define mb_setascii(mbc, sc) \ - ((mbc)->ptr = (mbc)->buf, (mbc)->bytes = 1, (mbc)->wc_valid = 1, \ - (mbc)->wc = (mbc)->buf[0] = (sc)) +# define mb_setascii(mbc, sc) \ + ((mbc)->ptr = (mbc)->buf, (mbc)->bytes = 1, (mbc)->wc_valid = 1, \ + (mbc)->wc = (mbc)->buf[0] = (sc)) +#endif /* Copying a character. */ MBCHAR_INLINE void mb_copy (mbchar_t *new_mbc, const mbchar_t *old_mbc) { +#if defined GNULIB_MBFILE if (old_mbc->ptr == &old_mbc->buf[0]) { memcpy (&new_mbc->buf[0], &old_mbc->buf[0], old_mbc->bytes); new_mbc->ptr = &new_mbc->buf[0]; } else +#endif new_mbc->ptr = old_mbc->ptr; new_mbc->bytes = old_mbc->bytes; if ((new_mbc->wc_valid = old_mbc->wc_valid)) @@ -272,44 +281,54 @@ mb_copy (mbchar_t *new_mbc, const mbchar_t *old_mbc) } -/* is_basic(c) tests whether the single-byte character c is in the - ISO C "basic character set". +/* is_basic(c) tests whether the single-byte character c is + - in the ISO C "basic character set" or is one of '@', '$', and '`' + which ISO C 23 ยง 5.2.1.1.(1) guarantees to be single-byte and in + practice are safe to treat as basic in the execution character set, + or + - in the POSIX "portable character set", which + <https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap06.html> + equally guarantees to be single-byte. This is a convenience function, and is in this file only to share code - between mbiter_multi.h and mbfile_multi.h. */ + between mbiter.h, mbuiter.h, and mbfile.h. */ #if (' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \ - && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \ - && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \ - && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \ - && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \ - && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \ - && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \ - && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \ - && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \ - && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \ - && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \ - && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \ - && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \ - && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \ - && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \ - && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \ - && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \ - && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \ - && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \ - && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \ - && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \ - && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \ - && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126) + && ('$' == 36) && ('%' == 37) && ('&' == 38) && ('\'' == 39) \ + && ('(' == 40) && (')' == 41) && ('*' == 42) && ('+' == 43) \ + && (',' == 44) && ('-' == 45) && ('.' == 46) && ('/' == 47) \ + && ('0' == 48) && ('1' == 49) && ('2' == 50) && ('3' == 51) \ + && ('4' == 52) && ('5' == 53) && ('6' == 54) && ('7' == 55) \ + && ('8' == 56) && ('9' == 57) && (':' == 58) && (';' == 59) \ + && ('<' == 60) && ('=' == 61) && ('>' == 62) && ('?' == 63) \ + && ('@' == 64) && ('A' == 65) && ('B' == 66) && ('C' == 67) \ + && ('D' == 68) && ('E' == 69) && ('F' == 70) && ('G' == 71) \ + && ('H' == 72) && ('I' == 73) && ('J' == 74) && ('K' == 75) \ + && ('L' == 76) && ('M' == 77) && ('N' == 78) && ('O' == 79) \ + && ('P' == 80) && ('Q' == 81) && ('R' == 82) && ('S' == 83) \ + && ('T' == 84) && ('U' == 85) && ('V' == 86) && ('W' == 87) \ + && ('X' == 88) && ('Y' == 89) && ('Z' == 90) && ('[' == 91) \ + && ('\\' == 92) && (']' == 93) && ('^' == 94) && ('_' == 95) \ + && ('`' == 96) && ('a' == 97) && ('b' == 98) && ('c' == 99) \ + && ('d' == 100) && ('e' == 101) && ('f' == 102) && ('g' == 103) \ + && ('h' == 104) && ('i' == 105) && ('j' == 106) && ('k' == 107) \ + && ('l' == 108) && ('m' == 109) && ('n' == 110) && ('o' == 111) \ + && ('p' == 112) && ('q' == 113) && ('r' == 114) && ('s' == 115) \ + && ('t' == 116) && ('u' == 117) && ('v' == 118) && ('w' == 119) \ + && ('x' == 120) && ('y' == 121) && ('z' == 122) && ('{' == 123) \ + && ('|' == 124) && ('}' == 125) && ('~' == 126) /* The character set is ISO-646, not EBCDIC. */ # define IS_BASIC_ASCII 1 -extern const unsigned int is_basic_table[]; - -MBCHAR_INLINE bool -is_basic (char c) -{ - return (is_basic_table [(unsigned char) c >> 5] >> ((unsigned char) c & 31)) - & 1; -} +/* All locale encodings (see localcharset.h) map the characters 0x00..0x7F + to U+0000..U+007F, like ASCII, except for + CP864 different mapping of '%' + SHIFT_JIS different mappings of 0x5C, 0x7E + JOHAB different mapping of 0x5C + However, these characters in the range 0x20..0x7E are in the ISO C + "basic character set" and in the POSIX "portable character set", which + ISO C and POSIX guarantee to be single-byte. Thus, locales with these + encodings are not POSIX compliant. And they are most likely not in use + any more (as of 2023). */ +# define is_basic(c) ((unsigned char) (c) < 0x80) #else @@ -318,21 +337,23 @@ is_basic (char c) { switch (c) { - case '\t': case '\v': case '\f': - case ' ': case '!': case '"': case '#': case '%': + case '\0': + case '\007': case '\010': + case '\t': case '\n': case '\v': case '\f': case '\r': + case ' ': case '!': case '"': case '#': case '$': case '%': case '&': case '\'': case '(': case ')': case '*': case '+': case ',': case '-': case '.': case '/': case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': case '8': case '9': case ':': case ';': case '<': case '=': case '>': - case '?': + case '?': case '@': case 'A': case 'B': case 'C': case 'D': case 'E': case 'F': case 'G': case 'H': case 'I': case 'J': case 'K': case 'L': case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R': case 'S': case 'T': case 'U': case 'V': case 'W': case 'X': case 'Y': case 'Z': - case '[': case '\\': case ']': case '^': case '_': + case '[': case '\\': case ']': case '^': case '_': case '`': case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': case 'g': case 'h': case 'i': case 'j': case 'k': case 'l': case 'm': case 'n': case 'o': |