1 .\" Copyright (C) Markus Kuhn, 1996, 2001
3 .\" %%%LICENSE_START(GPLv2+_DOC_FULL)
4 .\" This is free documentation; you can redistribute it and/or
5 .\" modify it under the terms of the GNU General Public License as
6 .\" published by the Free Software Foundation; either version 2 of
7 .\" the License, or (at your option) any later version.
9 .\" The GNU General Public License's references to "object code"
10 .\" and "executables" are to be interpreted as the output of any
11 .\" document formatting or typesetting system, including
12 .\" intermediate and printed output.
14 .\" This manual is distributed in the hope that it will be useful,
15 .\" but WITHOUT ANY WARRANTY; without even the implied warranty of
16 .\" MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 .\" GNU General Public License for more details.
19 .\" You should have received a copy of the GNU General Public
20 .\" License along with this manual; if not, see
21 .\" <http://www.gnu.org/licenses/>.
24 .\" 1995-11-26 Markus Kuhn <mskuhn@cip.informatik.uni-erlangen.de>
25 .\" First version written
26 .\" 2001-05-11 Markus Kuhn <mgk25@cl.cam.ac.uk>
29 .TH UTF-8 7 2014-02-26 "GNU" "Linux Programmer's Manual"
31 UTF-8 \- an ASCII compatible multibyte Unicode encoding
35 character set occupies a 16-bit code space.
37 Unicode encoding (known as
39 consists of a sequence of 16-bit words.
40 Such strings can contain\(emas part of many 16-bit characters\(embytes
41 such as \(aq\\0\(aq or \(aq/\(aq, which have a
42 special meaning in filenames and other C library function arguments.
43 In addition, the majority of UNIX tools expect ASCII files and can't
44 read 16-bit words as characters without major modifications.
47 is not a suitable external encoding of
49 in filenames, text files, environment variables, and so on.
51 .BR "ISO 10646 Universal Character Set (UCS)" ,
52 a superset of Unicode, occupies an even larger code
53 space\(em31\ bits\(emand the obvious
55 encoding for it (a sequence of 32-bit words) has the same problems.
63 does not have these problems and is the common way in which
65 is used on UNIX-style operating systems.
69 encoding has the following nice properties:
73 characters 0x00000000 to 0x0000007f (the classic
75 characters) are encoded simply as bytes 0x00 to 0x7f (ASCII
77 This means that files and strings which contain only
78 7-bit ASCII characters have the same encoding under both
86 characters greater than 0x7f are encoded as a multibyte sequence
87 consisting only of bytes in the range 0x80 to 0xfd, so no ASCII
88 byte can appear as part of another character and there are no
89 problems with, for example, \(aq\\0\(aq or \(aq/\(aq.
92 The lexicographic sorting order of
97 All possible 2^31 UCS codes can be encoded using
101 The bytes 0xc0, 0xc1, 0xfe, and 0xff are never used in the
106 The first byte of a multibyte sequence which represents a single non-ASCII
108 character is always in the range 0xc2 to 0xfd and indicates how long
109 this multibyte sequence is.
110 All further bytes in a multibyte sequence
111 are in the range 0x80 to 0xbf.
112 This allows easy resynchronization and
113 makes the encoding stateless and robust against missing bytes.
119 characters may be up to six bytes long, however the
121 standard specifies no characters above 0x10ffff, so Unicode characters
122 can be only up to four bytes long in
125 The following byte sequences are used to represent a character.
126 The sequence to be used depends on the UCS code number of the character:
128 0x00000000 \- 0x0000007F:
131 0x00000080 \- 0x000007FF:
135 0x00000800 \- 0x0000FFFF:
140 0x00010000 \- 0x001FFFFF:
146 0x00200000 \- 0x03FFFFFF:
153 0x04000000 \- 0x7FFFFFFF:
163 bit positions are filled with the bits of the character code number in
164 binary representation.
165 Only the shortest possible multibyte sequence
166 which can represent the code number of the character can be used.
170 code values 0xd800\(en0xdfff (UTF-16 surrogates) as well as 0xfffe and
171 0xffff (UCS noncharacters) should not appear in conforming
177 character 0xa9 = 1010 1001 (the copyright sign) is encoded
181 11000010 10101001 = 0xc2 0xa9
184 and character 0x2260 = 0010 0010 0110 0000 (the "not equal" symbol) is
188 11100010 10001001 10100000 = 0xe2 0x89 0xa0
190 .SS Application notes
191 Users have to select a
193 locale, for example with
196 export LANG=en_GB.UTF-8
199 in order to activate the
201 support in applications.
203 Application software that has to be aware of the used character
204 encoding should always set the locale with for example
207 setlocale(LC_CTYPE, "")
210 and programmers can then test the expression
213 strcmp(nl_langinfo(CODESET), "UTF-8") == 0
216 to determine whether a
218 locale has been selected and whether
219 therefore all plaintext standard input and output, terminal
220 communication, plaintext file content, filenames and environment
221 variables are encoded in
224 Programmers accustomed to single-byte encodings such as
228 have to be aware that two assumptions made so far are no longer valid
232 Firstly, a single byte does not necessarily correspond any
233 more to a single character.
234 Secondly, since modern terminal emulators
237 mode also support Chinese, Japanese, and Korean
238 .B double-width characters
239 as well as nonspacing
240 .BR "combining characters" ,
241 outputting a single character does not necessarily advance the cursor
242 by one position as it did in
244 Library functions such as
248 should be used today to count characters and cursor positions.
250 The official ESC sequence to switch from an
252 encoding scheme (as used for instance by VT100 terminals) to
256 The corresponding return sequence from
258 to ISO 2022 is ESC % @ ("\\x1b%@").
259 Other ISO 2022 sequences (such as
260 for switching the G0 and G1 sets) are not applicable in UTF-8 mode.
262 It can be hoped that in the foreseeable future,
268 at all levels as the common character encoding on POSIX systems,
269 leading to a significantly richer environment for handling plain text.
272 .BR Unicode " and " UCS
273 standards require that producers of
275 shall use the shortest form possible, for example, producing a two-byte
276 sequence with first byte 0xc0 is nonconforming.
278 has added the requirement that conforming programs must not accept
279 non-shortest forms in their input.
280 This is for security reasons: if
281 user input is checked for possible security violations, a program
282 might check only for the
284 version of "/../" or ";" or NUL and overlook that there are many
286 ways to represent these things in a non-shortest
290 ISO/IEC 10646-1:2000, Unicode 3.1, RFC\ 3629, Plan 9.
292 .\" Markus Kuhn <mgk25@cl.cam.ac.uk>
299 This page is part of release 3.67 of the Linux
302 A description of the project,
303 information about reporting bugs,
304 and the latest version of this page,
306 \%http://www.kernel.org/doc/man\-pages/.