include/charset.h

   1 /* SPDX-License-Identifier: GPL-2.0+ */
   2 /*
   3  *  charset conversion utils
   4  *
   5  *  Copyright (c) 2017 Rob Clark
   6  */
   7
   8 #ifndef __CHARSET_H_
   9 #define __CHARSET_H_
  10
  11 #include <linux/kernel.h>
  12 #include <linux/types.h>
  13
  14 #define MAX_UTF8_PER_UTF16 3
  15
  16 /**
  17  * console_read_unicode() - read Unicode code point from console
  18  *
  19  * @code:       pointer to store Unicode code point
  20  * Return:      0 = success
  21  */
  22 int console_read_unicode(s32 *code);
  23
  24 /**
  25  * utf8_get() - get next UTF-8 code point from buffer
  26  *
  27  * @src:                pointer to current byte, updated to point to next byte
  28  * Return:              code point, or 0 for end of string, or -1 if no legal
  29  *                      code point is found. In case of an error src points to
  30  *                      the incorrect byte.
  31  */
  32 s32 utf8_get(const char **src);
  33
  34 /**
  35  * utf8_put() - write UTF-8 code point to buffer
  36  *
  37  * @code:               code point
  38  * @dst:                pointer to destination buffer, updated to next position
  39  * Return:              -1 if the input parameters are invalid
  40  */
  41 int utf8_put(s32 code, char **dst);
  42
  43 /**
  44  * utf8_utf16_strnlen() - length of a truncated utf-8 string after conversion
  45  *                        to utf-16
  46  *
  47  * @src:                utf-8 string
  48  * @count:              maximum number of code points to convert
  49  * Return:              length in bytes after conversion to utf-16 without the
  50  *                      trailing \0. If an invalid UTF-8 sequence is hit one
  51  *                      word will be reserved for a replacement character.
  52  */
  53 size_t utf8_utf16_strnlen(const char *src, size_t count);
  54
  55 /**
  56  * utf8_utf16_strlen() - length of a utf-8 string after conversion to utf-16
  57  *
  58  * @src:                utf-8 string
  59  * Return:              length in bytes after conversion to utf-16 without the
  60  *                      trailing \0. -1 if the utf-8 string is not valid.
  61  */
  62 #define utf8_utf16_strlen(a) utf8_utf16_strnlen((a), SIZE_MAX)
  63
  64 /**
  65  * utf8_utf16_strncpy() - copy utf-8 string to utf-16 string
  66  *
  67  * @dst:                destination buffer
  68  * @src:                source buffer
  69  * @count:              maximum number of code points to copy
  70  * Return:              -1 if the input parameters are invalid
  71  */
  72 int utf8_utf16_strncpy(u16 **dst, const char *src, size_t count);
  73
  74 /**
  75  * utf8_utf16_strcpy() - copy utf-8 string to utf-16 string
  76  *
  77  * @dst:                destination buffer
  78  * @src:                source buffer
  79  * Return:              -1 if the input parameters are invalid
  80  */
  81 #define utf8_utf16_strcpy(d, s) utf8_utf16_strncpy((d), (s), SIZE_MAX)
  82
  83 /**
  84  * utf16_get() - get next UTF-16 code point from buffer
  85  *
  86  * @src:                pointer to current word, updated to point to next word
  87  * Return:              code point, or 0 for end of string, or -1 if no legal
  88  *                      code point is found. In case of an error src points to
  89  *                      the incorrect word.
  90  */
  91 s32 utf16_get(const u16 **src);
  92
  93 /**
  94  * utf16_put() - write UTF-16 code point to buffer
  95  *
  96  * @code:               code point
  97  * @dst:                pointer to destination buffer, updated to next position
  98  * Return:              -1 if the input parameters are invalid
  99  */
 100 int utf16_put(s32 code, u16 **dst);
 101
 102 /**
 103  * utf16_strnlen() - length of a truncated utf-16 string
 104  *
 105  * @src:                utf-16 string
 106  * @count:              maximum number of code points to convert
 107  * Return:              length in code points. If an invalid UTF-16 sequence is
 108  *                      hit one position will be reserved for a replacement
 109  *                      character.
 110  */
 111 size_t utf16_strnlen(const u16 *src, size_t count);
 112
 113 /**
 114  * utf16_utf8_strnlen() - length of a truncated utf-16 string after conversion
 115  *                        to utf-8
 116  *
 117  * @src:                utf-16 string
 118  * @count:              maximum number of code points to convert
 119  * Return:              length in bytes after conversion to utf-8 without the
 120  *                      trailing \0. If an invalid UTF-16 sequence is hit one
 121  *                      byte will be reserved for a replacement character.
 122  */
 123 size_t utf16_utf8_strnlen(const u16 *src, size_t count);
 124
 125 /**
 126  * utf16_utf8_strlen() - length of a utf-16 string after conversion to utf-8
 127  *
 128  * @src:                utf-16 string
 129  * Return:              length in bytes after conversion to utf-8 without the
 130  *                      trailing \0. -1 if the utf-16 string is not valid.
 131  */
 132 #define utf16_utf8_strlen(a) utf16_utf8_strnlen((a), SIZE_MAX)
 133
 134 /**
 135  * utf16_utf8_strncpy() - copy utf-16 string to utf-8 string
 136  *
 137  * @dst:                destination buffer
 138  * @src:                source buffer
 139  * @count:              maximum number of code points to copy
 140  * Return:              -1 if the input parameters are invalid
 141  */
 142 int utf16_utf8_strncpy(char **dst, const u16 *src, size_t count);
 143
 144 /**
 145  * utf16_utf8_strcpy() - copy utf-16 string to utf-8 string
 146  *
 147  * @dst:                destination buffer
 148  * @src:                source buffer
 149  * Return:              -1 if the input parameters are invalid
 150  */
 151 #define utf16_utf8_strcpy(d, s) utf16_utf8_strncpy((d), (s), SIZE_MAX)
 152
 153 /**
 154  * utf_to_lower() - convert a Unicode letter to lower case
 155  *
 156  * @code:               letter to convert
 157  * Return:              lower case letter or unchanged letter
 158  */
 159 s32 utf_to_lower(const s32 code);
 160
 161 /**
 162  * utf_to_upper() - convert a Unicode letter to upper case
 163  *
 164  * @code:               letter to convert
 165  * Return:              upper case letter or unchanged letter
 166  */
 167 s32 utf_to_upper(const s32 code);
 168
 169 /**
 170  * u16_strlen - count non-zero words
 171  *
 172  * This function matches wsclen() if the -fshort-wchar compiler flag is set.
 173  * In the EFI context we explicitly need a function handling u16 strings.
 174  *
 175  * @in:                 null terminated u16 string
 176  * ReturnValue:         number of non-zero words.
 177  *                      This is not the number of utf-16 letters!
 178  */
 179 size_t u16_strlen(const u16 *in);
 180
 181 /**
 182  * u16_strlen - count non-zero words
 183  *
 184  * This function matches wscnlen_s() if the -fshort-wchar compiler flag is set.
 185  * In the EFI context we explicitly need a function handling u16 strings.
 186  *
 187  * @in:                 null terminated u16 string
 188  * @count:              maximum number of words to count
 189  * ReturnValue:         number of non-zero words.
 190  *                      This is not the number of utf-16 letters!
 191  */
 192 size_t u16_strnlen(const u16 *in, size_t count);
 193
 194 /**
 195  * utf16_to_utf8() - Convert an utf16 string to utf8
 196  *
 197  * Converts 'size' characters of the utf16 string 'src' to utf8
 198  * written to the 'dest' buffer.
 199  *
 200  * NOTE that a single utf16 character can generate up to 3 utf8
 201  * characters.  See MAX_UTF8_PER_UTF16.
 202  *
 203  * @dest   the destination buffer to write the utf8 characters
 204  * @src    the source utf16 string
 205  * @size   the number of utf16 characters to convert
 206  * @return the pointer to the first unwritten byte in 'dest'
 207  */
 208 uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size);
 209
 210 #endif /* __CHARSET_H_ */