blob: 4d45e246e5158fdb6d7c8acf6133d42c4a8ce0e2 [file] [log] [blame]
Tom Rinif739fcd2018-05-07 17:02:21 -04001/* SPDX-License-Identifier: GPL-2.0+ */
Rob Clark78178bb2017-09-09 06:47:40 -04002/*
3 * charset conversion utils
4 *
5 * Copyright (c) 2017 Rob Clark
Rob Clark78178bb2017-09-09 06:47:40 -04006 */
7
8#ifndef __CHARSET_H_
9#define __CHARSET_H_
10
Heinrich Schuchardtd8c28232018-08-31 21:31:27 +020011#include <linux/kernel.h>
Heinrich Schuchardtf58c5ec2017-10-18 18:13:06 +020012#include <linux/types.h>
13
Heinrich Schuchardt984f2512017-10-09 21:09:05 +020014#define MAX_UTF8_PER_UTF16 3
Rob Clark78178bb2017-09-09 06:47:40 -040015
16/**
Heinrich Schuchardt35cbb792018-09-12 00:05:32 +020017 * console_read_unicode() - read Unicode code point from console
18 *
19 * @code: pointer to store Unicode code point
20 * Return: 0 = success
21 */
22int console_read_unicode(s32 *code);
23
24/**
Heinrich Schuchardtd8c28232018-08-31 21:31:27 +020025 * 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 */
32s32 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 */
41int 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 */
53size_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 */
72int 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 */
91s32 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 */
100int 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 */
111size_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 */
123size_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 */
142int 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/**
Heinrich Schuchardtb5130a82018-09-04 19:34:56 +0200154 * 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 */
159s32 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 */
167s32 utf_to_upper(const s32 code);
168
169/**
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200170 * u16_strlen - count non-zero words
Rob Clark78178bb2017-09-09 06:47:40 -0400171 *
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200172 * 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.
Rob Clark78178bb2017-09-09 06:47:40 -0400174 *
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200175 * @in: null terminated u16 string
176 * ReturnValue: number of non-zero words.
177 * This is not the number of utf-16 letters!
Rob Clark78178bb2017-09-09 06:47:40 -0400178 */
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200179size_t u16_strlen(const u16 *in);
Rob Clark78178bb2017-09-09 06:47:40 -0400180
181/**
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200182 * u16_strlen - count non-zero words
Rob Clark78178bb2017-09-09 06:47:40 -0400183 *
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200184 * 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.
Rob Clark78178bb2017-09-09 06:47:40 -0400186 *
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200187 * @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!
Rob Clark78178bb2017-09-09 06:47:40 -0400191 */
Heinrich Schuchardt1dde0d52018-08-31 21:31:26 +0200192size_t u16_strnlen(const u16 *in, size_t count);
Rob Clark78178bb2017-09-09 06:47:40 -0400193
194/**
Rob Clark78178bb2017-09-09 06:47:40 -0400195 * 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 *
Heinrich Schuchardt984f2512017-10-09 21:09:05 +0200200 * NOTE that a single utf16 character can generate up to 3 utf8
Rob Clark78178bb2017-09-09 06:47:40 -0400201 * 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 */
208uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size);
209
Rob Clark78178bb2017-09-09 06:47:40 -0400210#endif /* __CHARSET_H_ */