This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
[patch] :utf8 updates
[perl5.git] / pod / perluniintro.pod
CommitLineData
ba62762e
JH
1=head1 NAME
2
07fcf8ff 3perluniintro - Perl Unicode introduction
ba62762e
JH
4
5=head1 DESCRIPTION
6
7This document gives a general idea of Unicode and how to use Unicode
8in Perl.
9
10=head2 Unicode
11
376d9008 12Unicode is a character set standard which plans to codify all of the
ba62762e
JH
13writing systems of the world, plus many other symbols.
14
15Unicode and ISO/IEC 10646 are coordinated standards that provide code
376d9008 16points for characters in almost all modern character set standards,
ba62762e 17covering more than 30 writing systems and hundreds of languages,
376d9008 18including all commercially-important modern languages. All characters
ba62762e
JH
19in the largest Chinese, Japanese, and Korean dictionaries are also
20encoded. The standards will eventually cover almost all characters in
21more than 250 writing systems and thousands of languages.
4c496f0c 22Unicode 1.0 was released in October 1991, and 4.0 in April 2003.
ba62762e
JH
23
24A Unicode I<character> is an abstract entity. It is not bound to any
376d9008
JB
25particular integer width, especially not to the C language C<char>.
26Unicode is language-neutral and display-neutral: it does not encode the
27language of the text and it does not define fonts or other graphical
ba62762e
JH
28layout details. Unicode operates on characters and on text built from
29those characters.
30
31Unicode defines characters like C<LATIN CAPITAL LETTER A> or C<GREEK
376d9008
JB
32SMALL LETTER ALPHA> and unique numbers for the characters, in this
33case 0x0041 and 0x03B1, respectively. These unique numbers are called
34I<code points>.
ba62762e
JH
35
36The Unicode standard prefers using hexadecimal notation for the code
4c496f0c
JH
37points. If numbers like C<0x0041> are unfamiliar to you, take a peek
38at a later section, L</"Hexadecimal Notation">. The Unicode standard
39uses the notation C<U+0041 LATIN CAPITAL LETTER A>, to give the
40hexadecimal code point and the normative name of the character.
ba62762e
JH
41
42Unicode also defines various I<properties> for the characters, like
376d9008 43"uppercase" or "lowercase", "decimal digit", or "punctuation";
ba62762e
JH
44these properties are independent of the names of the characters.
45Furthermore, various operations on the characters like uppercasing,
376d9008 46lowercasing, and collating (sorting) are defined.
ba62762e
JH
47
48A Unicode character consists either of a single code point, or a
49I<base character> (like C<LATIN CAPITAL LETTER A>), followed by one or
50more I<modifiers> (like C<COMBINING ACUTE ACCENT>). This sequence of
376d9008 51base character and modifiers is called a I<combining character
ba62762e
JH
52sequence>.
53
376d9008
JB
54Whether to call these combining character sequences "characters"
55depends on your point of view. If you are a programmer, you probably
56would tend towards seeing each element in the sequences as one unit,
57or "character". The whole sequence could be seen as one "character",
58however, from the user's point of view, since that's probably what it
59looks like in the context of the user's language.
60
61With this "whole sequence" view of characters, the total number of
62characters is open-ended. But in the programmer's "one unit is one
63character" point of view, the concept of "characters" is more
1bfb14c4
JH
64deterministic. In this document, we take that second point of view:
65one "character" is one Unicode code point, be it a base character or
66a combining character.
376d9008
JB
67
68For some combinations, there are I<precomposed> characters.
69C<LATIN CAPITAL LETTER A WITH ACUTE>, for example, is defined as
ba62762e 70a single code point. These precomposed characters are, however,
376d9008 71only available for some combinations, and are mainly
ba62762e 72meant to support round-trip conversions between Unicode and legacy
376d9008
JB
73standards (like the ISO 8859). In the general case, the composing
74method is more extensible. To support conversion between
ba62762e 75different compositions of the characters, various I<normalization
376d9008 76forms> to standardize representations are also defined.
ba62762e
JH
77
78Because of backward compatibility with legacy encodings, the "a unique
376d9008
JB
79number for every character" idea breaks down a bit: instead, there is
80"at least one number for every character". The same character could
81be represented differently in several legacy encodings. The
82converse is also not true: some code points do not have an assigned
83character. Firstly, there are unallocated code points within
84otherwise used blocks. Secondly, there are special Unicode control
85characters that do not represent true characters.
ba62762e
JH
86
87A common myth about Unicode is that it would be "16-bit", that is,
376d9008 88Unicode is only represented as C<0x10000> (or 65536) characters from
4c496f0c
JH
89C<0x0000> to C<0xFFFF>. B<This is untrue.> Since Unicode 2.0 (July
901996), Unicode has been defined all the way up to 21 bits (C<0x10FFFF>),
91and since Unicode 3.1 (March 2001), characters have been defined
92beyond C<0xFFFF>. The first C<0x10000> characters are called the
93I<Plane 0>, or the I<Basic Multilingual Plane> (BMP). With Unicode
943.1, 17 (yes, seventeen) planes in all were defined--but they are
95nowhere near full of defined characters, yet.
ba62762e 96
1bfb14c4
JH
97Another myth is that the 256-character blocks have something to
98do with languages--that each block would define the characters used
99by a language or a set of languages. B<This is also untrue.>
100The division into blocks exists, but it is almost completely
101accidental--an artifact of how the characters have been and
102still are allocated. Instead, there is a concept called I<scripts>,
103which is more useful: there is C<Latin> script, C<Greek> script, and
104so on. Scripts usually span varied parts of several blocks.
105For further information see L<Unicode::UCD>.
ba62762e
JH
106
107The Unicode code points are just abstract numbers. To input and
4c496f0c
JH
108output these abstract numbers, the numbers must be I<encoded> or
109I<serialised> somehow. Unicode defines several I<character encoding
110forms>, of which I<UTF-8> is perhaps the most popular. UTF-8 is a
111variable length encoding that encodes Unicode characters as 1 to 6
112bytes (only 4 with the currently defined characters). Other encodings
113include UTF-16 and UTF-32 and their big- and little-endian variants
114(UTF-8 is byte-order independent) The ISO/IEC 10646 defines the UCS-2
115and UCS-4 encoding forms.
ba62762e 116
376d9008
JB
117For more information about encodings--for instance, to learn what
118I<surrogates> and I<byte order marks> (BOMs) are--see L<perlunicode>.
ba62762e
JH
119
120=head2 Perl's Unicode Support
121
376d9008
JB
122Starting from Perl 5.6.0, Perl has had the capacity to handle Unicode
123natively. Perl 5.8.0, however, is the first recommended release for
124serious Unicode work. The maintenance release 5.6.1 fixed many of the
125problems of the initial Unicode implementation, but for example
1bfb14c4 126regular expressions still do not work with Unicode in 5.6.1.
ba62762e
JH
127
128B<Starting from Perl 5.8.0, the use of C<use utf8> is no longer
129necessary.> In earlier releases the C<utf8> pragma was used to declare
130that operations in the current block or file would be Unicode-aware.
376d9008 131This model was found to be wrong, or at least clumsy: the "Unicodeness"
1bfb14c4 132is now carried with the data, instead of being attached to the
376d9008
JB
133operations. Only one case remains where an explicit C<use utf8> is
134needed: if your Perl script itself is encoded in UTF-8, you can use
135UTF-8 in your identifier names, and in string and regular expression
136literals, by saying C<use utf8>. This is not the default because
8f8cf39c 137scripts with legacy 8-bit data in them would break. See L<utf8>.
ba62762e
JH
138
139=head2 Perl's Unicode Model
140
376d9008
JB
141Perl supports both pre-5.6 strings of eight-bit native bytes, and
142strings of Unicode characters. The principle is that Perl tries to
143keep its data as eight-bit bytes for as long as possible, but as soon
144as Unicodeness cannot be avoided, the data is transparently upgraded
145to Unicode.
ba62762e 146
4192de81 147Internally, Perl currently uses either whatever the native eight-bit
376d9008
JB
148character set of the platform (for example Latin-1) is, defaulting to
149UTF-8, to encode Unicode strings. Specifically, if all code points in
150the string are C<0xFF> or less, Perl uses the native eight-bit
151character set. Otherwise, it uses UTF-8.
4192de81 152
7ca610e8 153A user of Perl does not normally need to know nor care how Perl
20ba30f4 154happens to encode its internal strings, but it becomes relevant when
fae2c0fb 155outputting Unicode strings to a stream without a PerlIO layer -- one with
376d9008 156the "default" encoding. In such a case, the raw bytes used internally
7ca610e8
JH
157(the native character set or UTF-8, as appropriate for each string)
158will be used, and a "Wide character" warning will be issued if those
159strings contain a character beyond 0x00FF.
4192de81
JH
160
161For example,
162
ae5648b3 163 perl -e 'print "\x{DF}\n", "\x{0100}\x{DF}\n"'
4192de81
JH
164
165produces a fairly useless mixture of native bytes and UTF-8, as well
1bfb14c4 166as a warning:
4192de81 167
1bfb14c4
JH
168 Wide character in print at ...
169
740d4bb2 170To output UTF-8, use the C<:encoding> or C<:utf8> output layer. Prepending
4192de81
JH
171
172 binmode(STDOUT, ":utf8");
173
376d9008 174to this sample program ensures that the output is completely UTF-8,
1bfb14c4 175and removes the program's warning.
ba62762e 176
8aa8f774
JH
177You can enable automatic UTF-8-ification of your standard file
178handles, default C<open()> layer, and C<@ARGV> by using either
179the C<-C> command line switch or the C<PERL_UNICODE> environment
180variable, see L<perlrun> for the documentation of the C<-C> switch.
181
182Note that this means that Perl expects other software to work, too:
183if Perl has been led to believe that STDIN should be UTF-8, but then
184STDIN coming in from another command is not UTF-8, Perl will complain
185about the malformed UTF-8.
b310b053 186
ec71e770
JH
187All features that combine Unicode and I/O also require using the new
188PerlIO feature. Almost all Perl 5.8 platforms do use PerlIO, though:
189you can see whether yours is by running "perl -V" and looking for
190C<useperlio=define>.
191
64c66fb6
JH
192=head2 Unicode and EBCDIC
193
194Perl 5.8.0 also supports Unicode on EBCDIC platforms. There,
376d9008 195Unicode support is somewhat more complex to implement since
64c66fb6 196additional conversions are needed at every step. Some problems
dc4af4bb 197remain, see L<perlebcdic> for details.
64c66fb6
JH
198
199In any case, the Unicode support on EBCDIC platforms is better than
200in the 5.6 series, which didn't work much at all for EBCDIC platform.
201On EBCDIC platforms, the internal Unicode encoding form is UTF-EBCDIC
376d9008
JB
202instead of UTF-8. The difference is that as UTF-8 is "ASCII-safe" in
203that ASCII characters encode to UTF-8 as-is, while UTF-EBCDIC is
204"EBCDIC-safe".
ba62762e
JH
205
206=head2 Creating Unicode
207
376d9008
JB
208To create Unicode characters in literals for code points above C<0xFF>,
209use the C<\x{...}> notation in double-quoted strings:
ba62762e
JH
210
211 my $smiley = "\x{263a}";
212
376d9008 213Similarly, it can be used in regular expression literals
ba62762e
JH
214
215 $smiley =~ /\x{263a}/;
216
217At run-time you can use C<chr()>:
218
219 my $hebrew_alef = chr(0x05d0);
220
376d9008 221See L</"Further Resources"> for how to find all these numeric codes.
ba62762e 222
376d9008
JB
223Naturally, C<ord()> will do the reverse: it turns a character into
224a code point.
ba62762e 225
1bfb14c4
JH
226Note that C<\x..> (no C<{}> and only two hexadecimal digits), C<\x{...}>,
227and C<chr(...)> for arguments less than C<0x100> (decimal 256)
228generate an eight-bit character for backward compatibility with older
229Perls. For arguments of C<0x100> or more, Unicode characters are
a5f0baef
JH
230always produced. If you want to force the production of Unicode
231characters regardless of the numeric value, use C<pack("U", ...)>
232instead of C<\x..>, C<\x{...}>, or C<chr()>.
ba62762e
JH
233
234You can also use the C<charnames> pragma to invoke characters
376d9008 235by name in double-quoted strings:
ba62762e
JH
236
237 use charnames ':full';
238 my $arabic_alef = "\N{ARABIC LETTER ALEF}";
239
240And, as mentioned above, you can also C<pack()> numbers into Unicode
241characters:
242
243 my $georgian_an = pack("U", 0x10a0);
244
8a5e5dd5
JH
245Note that both C<\x{...}> and C<\N{...}> are compile-time string
246constants: you cannot use variables in them. if you want similar
247run-time functionality, use C<chr()> and C<charnames::vianame()>.
248
1eda90df 249If you want to force the result to Unicode characters, use the special
f337b084
TH
250C<"U0"> prefix. It consumes no arguments but causes the following bytes
251to be interpreted as the UTF-8 encoding of Unicode characters:
1eda90df 252
f337b084 253 my $chars = pack("U0W*", 0x80, 0x42);
771cd3b2 254
ae5648b3 255Likewise, you can stop such UTF-8 interpretation by using the special
771cd3b2 256C<"C0"> prefix.
1eda90df 257
ba62762e
JH
258=head2 Handling Unicode
259
260Handling Unicode is for the most part transparent: just use the
261strings as usual. Functions like C<index()>, C<length()>, and
262C<substr()> will work on the Unicode characters; regular expressions
263will work on the Unicode characters (see L<perlunicode> and L<perlretut>).
264
1bfb14c4 265Note that Perl considers combining character sequences to be
c0c50798 266separate characters, so for example
ba62762e
JH
267
268 use charnames ':full';
269 print length("\N{LATIN CAPITAL LETTER A}\N{COMBINING ACUTE ACCENT}"), "\n";
270
271will print 2, not 1. The only exception is that regular expressions
272have C<\X> for matching a combining character sequence.
273
376d9008
JB
274Life is not quite so transparent, however, when working with legacy
275encodings, I/O, and certain special cases:
ba62762e
JH
276
277=head2 Legacy Encodings
278
279When you combine legacy data and Unicode the legacy data needs
280to be upgraded to Unicode. Normally ISO 8859-1 (or EBCDIC, if
ae5648b3 281applicable) is assumed.
ba62762e 282
376d9008 283The C<Encode> module knows about many encodings and has interfaces
ba62762e
JH
284for doing conversions between those encodings:
285
2c9359a2
RGS
286 use Encode 'decode';
287 $data = decode("iso-8859-3", $data); # convert from legacy to utf-8
ba62762e
JH
288
289=head2 Unicode I/O
290
8baee566 291Normally, writing out Unicode data
ba62762e 292
8baee566 293 print FH $some_string_with_unicode, "\n";
ba62762e 294
8baee566 295produces raw bytes that Perl happens to use to internally encode the
376d9008
JB
296Unicode string. Perl's internal encoding depends on the system as
297well as what characters happen to be in the string at the time. If
298any of the characters are at code points C<0x100> or above, you will get
299a warning. To ensure that the output is explicitly rendered in the
300encoding you desire--and to avoid the warning--open the stream with
301the desired encoding. Some examples:
ba62762e 302
ec90690f
ST
303 open FH, ">:utf8", "file";
304
305 open FH, ">:encoding(ucs2)", "file";
306 open FH, ">:encoding(UTF-8)", "file";
307 open FH, ">:encoding(shift_jis)", "file";
1d7919c5 308
376d9008 309and on already open streams, use C<binmode()>:
1d7919c5
JH
310
311 binmode(STDOUT, ":utf8");
312
ec90690f
ST
313 binmode(STDOUT, ":encoding(ucs2)");
314 binmode(STDOUT, ":encoding(UTF-8)");
315 binmode(STDOUT, ":encoding(shift_jis)");
316
b5d8778e 317The matching of encoding names is loose: case does not matter, and
fae2c0fb 318many encodings have several aliases. Note that the C<:utf8> layer
1bfb14c4 319must always be specified exactly like that; it is I<not> subject to
740d4bb2
JW
320the loose matching of encoding names. Also note that C<:utf8> is unsafe for
321input, because it accepts the data without validating that it is indeed valid
322UTF8.
b5d8778e 323
376d9008
JB
324See L<PerlIO> for the C<:utf8> layer, L<PerlIO::encoding> and
325L<Encode::PerlIO> for the C<:encoding()> layer, and
326L<Encode::Supported> for many encodings supported by the C<Encode>
327module.
ba62762e 328
a5f0baef 329Reading in a file that you know happens to be encoded in one of the
1bfb14c4
JH
330Unicode or legacy encodings does not magically turn the data into
331Unicode in Perl's eyes. To do that, specify the appropriate
fae2c0fb 332layer when opening files
ba62762e 333
740d4bb2 334 open(my $fh,'<:encoding(utf8)', 'anything');
8baee566
JF
335 my $line_of_unicode = <$fh>;
336
ec90690f 337 open(my $fh,'<:encoding(Big5)', 'anything');
8baee566 338 my $line_of_unicode = <$fh>;
ba62762e 339
fae2c0fb 340The I/O layers can also be specified more flexibly with
376d9008 341the C<open> pragma. See L<open>, or look at the following example.
ba62762e 342
740d4bb2 343 use open ':encoding(utf8)'; # input/output default encoding will be UTF-8
1d7919c5
JH
344 open X, ">file";
345 print X chr(0x100), "\n";
ba62762e 346 close X;
1d7919c5 347 open Y, "<file";
ba62762e
JH
348 printf "%#x\n", ord(<Y>); # this should print 0x100
349 close Y;
350
fae2c0fb 351With the C<open> pragma you can use the C<:locale> layer
ba62762e 352
12f98225 353 BEGIN { $ENV{LC_ALL} = $ENV{LANG} = 'ru_RU.KOI8-R' }
1ecefa54 354 # the :locale will probe the locale environment variables like LC_ALL
ba62762e
JH
355 use open OUT => ':locale'; # russki parusski
356 open(O, ">koi8");
357 print O chr(0x430); # Unicode CYRILLIC SMALL LETTER A = KOI8-R 0xc1
358 close O;
359 open(I, "<koi8");
360 printf "%#x\n", ord(<I>), "\n"; # this should print 0xc1
361 close I;
362
8baee566
JF
363These methods install a transparent filter on the I/O stream that
364converts data from the specified encoding when it is read in from the
a5f0baef 365stream. The result is always Unicode.
ba62762e
JH
366
367The L<open> pragma affects all the C<open()> calls after the pragma by
fae2c0fb
RGS
368setting default layers. If you want to affect only certain
369streams, use explicit layers directly in the C<open()> call.
ba62762e
JH
370
371You can switch encodings on an already opened stream by using
8baee566 372C<binmode()>; see L<perlfunc/binmode>.
ba62762e 373
1ecefa54
JH
374The C<:locale> does not currently (as of Perl 5.8.0) work with
375C<open()> and C<binmode()>, only with the C<open> pragma. The
8baee566 376C<:utf8> and C<:encoding(...)> methods do work with all of C<open()>,
1ecefa54 377C<binmode()>, and the C<open> pragma.
ba62762e 378
fae2c0fb 379Similarly, you may use these I/O layers on output streams to
a5f0baef
JH
380automatically convert Unicode to the specified encoding when it is
381written to the stream. For example, the following snippet copies the
382contents of the file "text.jis" (encoded as ISO-2022-JP, aka JIS) to
383the file "text.utf8", encoded as UTF-8:
ba62762e 384
c722ef0a
JH
385 open(my $nihongo, '<:encoding(iso-2022-jp)', 'text.jis');
386 open(my $unicode, '>:utf8', 'text.utf8');
0cf8a8d9 387 while (<$nihongo>) { print $unicode $_ }
ba62762e
JH
388
389The naming of encodings, both by the C<open()> and by the C<open>
2575c402
JW
390pragma allows for flexible names: C<koi8-r> and C<KOI8R> will both be
391understood.
ba62762e
JH
392
393Common encodings recognized by ISO, MIME, IANA, and various other
8baee566 394standardisation organisations are recognised; for a more detailed
1bfb14c4 395list see L<Encode::Supported>.
ba62762e
JH
396
397C<read()> reads characters and returns the number of characters.
398C<seek()> and C<tell()> operate on byte counts, as do C<sysread()>
399and C<sysseek()>.
400
8baee566 401Notice that because of the default behaviour of not doing any
fae2c0fb 402conversion upon input if there is no default layer,
ba62762e 403it is easy to mistakenly write code that keeps on expanding a file
1bfb14c4 404by repeatedly encoding the data:
ba62762e
JH
405
406 # BAD CODE WARNING
407 open F, "file";
8baee566 408 local $/; ## read in the whole file of 8-bit characters
ba62762e
JH
409 $t = <F>;
410 close F;
740d4bb2 411 open F, ">:encoding(utf8)", "file";
8baee566 412 print F $t; ## convert to UTF-8 on output
ba62762e
JH
413 close F;
414
415If you run this code twice, the contents of the F<file> will be twice
740d4bb2
JW
416UTF-8 encoded. A C<use open ':encoding(utf8)'> would have avoided the
417bug, or explicitly opening also the F<file> for input as UTF-8.
ba62762e 418
0c901d84 419B<NOTE>: the C<:utf8> and C<:encoding> features work only if your
d0fadae5
JH
420Perl has been built with the new PerlIO feature (which is the default
421on most systems).
0c901d84 422
1ecefa54
JH
423=head2 Displaying Unicode As Text
424
425Sometimes you might want to display Perl scalars containing Unicode as
8baee566 426simple ASCII (or EBCDIC) text. The following subroutine converts
1ecefa54 427its argument so that Unicode characters with code points greater than
1bfb14c4
JH
428255 are displayed as C<\x{...}>, control characters (like C<\n>) are
429displayed as C<\x..>, and the rest of the characters as themselves:
1ecefa54 430
58c274a1
JF
431 sub nice_string {
432 join("",
433 map { $_ > 255 ? # if wide character...
8baee566 434 sprintf("\\x{%04X}", $_) : # \x{...}
58c274a1 435 chr($_) =~ /[[:cntrl:]]/ ? # else if control character ...
8baee566 436 sprintf("\\x%02X", $_) : # \x..
d0551e73 437 quotemeta(chr($_)) # else quoted or as themselves
f337b084 438 } unpack("W*", $_[0])); # unpack Unicode characters
58c274a1
JF
439 }
440
441For example,
442
443 nice_string("foo\x{100}bar\n")
444
d0551e73 445returns the string
58c274a1 446
d0551e73
JH
447 'foo\x{0100}bar\x0A'
448
449which is ready to be printed.
1ecefa54 450
ba62762e
JH
451=head2 Special Cases
452
453=over 4
454
455=item *
456
457Bit Complement Operator ~ And vec()
458
1bfb14c4
JH
459The bit complement operator C<~> may produce surprising results if
460used on strings containing characters with ordinal values above
461255. In such a case, the results are consistent with the internal
462encoding of the characters, but not with much else. So don't do
463that. Similarly for C<vec()>: you will be operating on the
464internally-encoded bit patterns of the Unicode characters, not on
465the code point values, which is very probably not what you want.
ba62762e
JH
466
467=item *
468
8baee566
JF
469Peeking At Perl's Internal Encoding
470
471Normal users of Perl should never care how Perl encodes any particular
a5f0baef 472Unicode string (because the normal ways to get at the contents of a
376d9008 473string with Unicode--via input and output--should always be via
fae2c0fb 474explicitly-defined I/O layers). But if you must, there are two
a5f0baef 475ways of looking behind the scenes.
ba62762e
JH
476
477One way of peeking inside the internal encoding of Unicode characters
f337b084
TH
478is to use C<unpack("C*", ...> to get the bytes of whatever the string
479encoding happens to be, or C<unpack("U0..", ...)> to get the bytes of the
480UTF-8 encoding:
ba62762e 481
8baee566 482 # this prints c4 80 for the UTF-8 bytes 0xc4 0x80
f337b084 483 print join(" ", unpack("U0(H2)*", pack("U", 0x100))), "\n";
ba62762e
JH
484
485Yet another way would be to use the Devel::Peek module:
486
487 perl -MDevel::Peek -e 'Dump(chr(0x100))'
488
1e54db1a 489That shows the C<UTF8> flag in FLAGS and both the UTF-8 bytes
376d9008 490and Unicode characters in C<PV>. See also later in this document
8800c35a 491the discussion about the C<utf8::is_utf8()> function.
ba62762e
JH
492
493=back
494
495=head2 Advanced Topics
496
497=over 4
498
499=item *
500
501String Equivalence
502
503The question of string equivalence turns somewhat complicated
376d9008 504in Unicode: what do you mean by "equal"?
ba62762e 505
07698885
RGS
506(Is C<LATIN CAPITAL LETTER A WITH ACUTE> equal to
507C<LATIN CAPITAL LETTER A>?)
ba62762e 508
a5f0baef
JH
509The short answer is that by default Perl compares equivalence (C<eq>,
510C<ne>) based only on code points of the characters. In the above
376d9008
JB
511case, the answer is no (because 0x00C1 != 0x0041). But sometimes, any
512CAPITAL LETTER As should be considered equal, or even As of any case.
ba62762e
JH
513
514The long answer is that you need to consider character normalization
376d9008 515and casing issues: see L<Unicode::Normalize>, Unicode Technical
ba62762e 516Reports #15 and #21, I<Unicode Normalization Forms> and I<Case
ae5648b3
RGS
517Mappings>, http://www.unicode.org/unicode/reports/tr15/ and
518http://www.unicode.org/unicode/reports/tr21/
ba62762e 519
1bfb14c4
JH
520As of Perl 5.8.0, the "Full" case-folding of I<Case
521Mappings/SpecialCasing> is implemented.
ba62762e
JH
522
523=item *
524
525String Collation
526
376d9008 527People like to see their strings nicely sorted--or as Unicode
ba62762e
JH
528parlance goes, collated. But again, what do you mean by collate?
529
07698885
RGS
530(Does C<LATIN CAPITAL LETTER A WITH ACUTE> come before or after
531C<LATIN CAPITAL LETTER A WITH GRAVE>?)
ba62762e 532
58c274a1 533The short answer is that by default, Perl compares strings (C<lt>,
ba62762e 534C<le>, C<cmp>, C<ge>, C<gt>) based only on the code points of the
1bfb14c4 535characters. In the above case, the answer is "after", since
da76a1f4 536C<0x00C1> > C<0x00C0>.
ba62762e
JH
537
538The long answer is that "it depends", and a good answer cannot be
539given without knowing (at the very least) the language context.
540See L<Unicode::Collate>, and I<Unicode Collation Algorithm>
541http://www.unicode.org/unicode/reports/tr10/
542
543=back
544
545=head2 Miscellaneous
546
547=over 4
548
549=item *
550
3ff56b75 551Character Ranges and Classes
ba62762e
JH
552
553Character ranges in regular expression character classes (C</[a-z]/>)
554and in the C<tr///> (also known as C<y///>) operator are not magically
58c274a1 555Unicode-aware. What this means that C<[A-Za-z]> will not magically start
376d9008
JB
556to mean "all alphabetic letters"; not that it does mean that even for
5578-bit characters, you should be using C</[[:alpha:]]/> in that case.
ba62762e 558
1bfb14c4
JH
559For specifying character classes like that in regular expressions,
560you can use the various Unicode properties--C<\pL>, or perhaps
561C<\p{Alphabetic}>, in this particular case. You can use Unicode
562code points as the end points of character ranges, but there is no
563magic associated with specifying a certain range. For further
564information--there are dozens of Unicode character classes--see
565L<perlunicode>.
ba62762e
JH
566
567=item *
568
569String-To-Number Conversions
570
376d9008
JB
571Unicode does define several other decimal--and numeric--characters
572besides the familiar 0 to 9, such as the Arabic and Indic digits.
ba62762e 573Perl does not support string-to-number conversion for digits other
58c274a1 574than ASCII 0 to 9 (and ASCII a to f for hexadecimal).
ba62762e
JH
575
576=back
577
578=head2 Questions With Answers
579
580=over 4
581
818c4caa 582=item *
5cb3728c
RB
583
584Will My Old Scripts Break?
ba62762e
JH
585
586Very probably not. Unless you are generating Unicode characters
1bfb14c4
JH
587somehow, old behaviour should be preserved. About the only behaviour
588that has changed and which could start generating Unicode is the old
589behaviour of C<chr()> where supplying an argument more than 255
590produced a character modulo 255. C<chr(300)>, for example, was equal
591to C<chr(45)> or "-" (in ASCII), now it is LATIN CAPITAL LETTER I WITH
592BREVE.
ba62762e 593
818c4caa 594=item *
5cb3728c
RB
595
596How Do I Make My Scripts Work With Unicode?
ba62762e
JH
597
598Very little work should be needed since nothing changes until you
1bfb14c4
JH
599generate Unicode data. The most important thing is getting input as
600Unicode; for that, see the earlier I/O discussion.
ba62762e 601
818c4caa 602=item *
5cb3728c
RB
603
604How Do I Know Whether My String Is In Unicode?
ba62762e 605
1bfb14c4
JH
606You shouldn't care. No, you really shouldn't. No, really. If you
607have to care--beyond the cases described above--it means that we
ba62762e
JH
608didn't get the transparency of Unicode quite right.
609
610Okay, if you insist:
611
8800c35a 612 print utf8::is_utf8($string) ? 1 : 0, "\n";
ba62762e
JH
613
614But note that this doesn't mean that any of the characters in the
615string are necessary UTF-8 encoded, or that any of the characters have
616code points greater than 0xFF (255) or even 0x80 (128), or that the
617string has any characters at all. All the C<is_utf8()> does is to
618return the value of the internal "utf8ness" flag attached to the
376d9008 619C<$string>. If the flag is off, the bytes in the scalar are interpreted
3c1c8017 620as a single byte encoding. If the flag is on, the bytes in the scalar
376d9008 621are interpreted as the (multi-byte, variable-length) UTF-8 encoded code
3c1c8017 622points of the characters. Bytes added to an UTF-8 encoded string are
1e54db1a 623automatically upgraded to UTF-8. If mixed non-UTF-8 and UTF-8 scalars
376d9008 624are merged (double-quoted interpolation, explicit concatenation, and
3c1c8017
AT
625printf/sprintf parameter substitution), the result will be UTF-8 encoded
626as if copies of the byte strings were upgraded to UTF-8: for example,
627
628 $a = "ab\x80c";
629 $b = "\x{100}";
630 print "$a = $b\n";
631
a02b5feb
JH
632the output string will be UTF-8-encoded C<ab\x80c = \x{100}\n>, but
633C<$a> will stay byte-encoded.
ba62762e
JH
634
635Sometimes you might really need to know the byte length of a string
ce7675db
A
636instead of the character length. For that use either the
637C<Encode::encode_utf8()> function or the C<bytes> pragma and its only
638defined function C<length()>:
ba62762e
JH
639
640 my $unicode = chr(0x100);
641 print length($unicode), "\n"; # will print 1
ce7675db
A
642 require Encode;
643 print length(Encode::encode_utf8($unicode)), "\n"; # will print 2
ba62762e 644 use bytes;
1bfb14c4
JH
645 print length($unicode), "\n"; # will also print 2
646 # (the 0xC4 0x80 of the UTF-8)
ba62762e 647
818c4caa 648=item *
5cb3728c
RB
649
650How Do I Detect Data That's Not Valid In a Particular Encoding?
ba62762e 651
8baee566
JF
652Use the C<Encode> package to try converting it.
653For example,
ba62762e 654
bb2f379c 655 use Encode 'decode_utf8';
a365f2ce
DR
656 eval { decode_utf8($string, Encode::FB_CROAK) };
657 if ($@) {
658 # $string is valid utf8
ba62762e 659 } else {
a365f2ce 660 # $string is not valid utf8
ba62762e
JH
661 }
662
f337b084 663Or use C<unpack> to try decoding it:
ba62762e
JH
664
665 use warnings;
f337b084 666 @chars = unpack("C0U*", $string_of_bytes_that_I_think_is_utf8);
ba62762e 667
ae5648b3
RGS
668If invalid, a C<Malformed UTF-8 character> warning is produced. The "C0" means
669"process the string character per character". Without that, the
670C<unpack("U*", ...)> would work in C<U0> mode (the default if the format
671string starts with C<U>) and it would return the bytes making up the UTF-8
f337b084 672encoding of the target string, something that will always work.
ba62762e 673
818c4caa 674=item *
5cb3728c
RB
675
676How Do I Convert Binary Data Into a Particular Encoding, Or Vice Versa?
ba62762e 677
8baee566
JF
678This probably isn't as useful as you might think.
679Normally, you shouldn't need to.
ba62762e 680
1bfb14c4 681In one sense, what you are asking doesn't make much sense: encodings
376d9008 682are for characters, and binary data are not "characters", so converting
a5f0baef
JH
683"data" into some encoding isn't meaningful unless you know in what
684character set and encoding the binary data is in, in which case it's
376d9008 685not just binary data, now is it?
8baee566 686
1bfb14c4
JH
687If you have a raw sequence of bytes that you know should be
688interpreted via a particular encoding, you can use C<Encode>:
ba62762e
JH
689
690 use Encode 'from_to';
691 from_to($data, "iso-8859-1", "utf-8"); # from latin-1 to utf-8
692
1bfb14c4
JH
693The call to C<from_to()> changes the bytes in C<$data>, but nothing
694material about the nature of the string has changed as far as Perl is
695concerned. Both before and after the call, the string C<$data>
696contains just a bunch of 8-bit bytes. As far as Perl is concerned,
697the encoding of the string remains as "system-native 8-bit bytes".
8baee566
JF
698
699You might relate this to a fictional 'Translate' module:
700
701 use Translate;
702 my $phrase = "Yes";
703 Translate::from_to($phrase, 'english', 'deutsch');
704 ## phrase now contains "Ja"
ba62762e 705
8baee566 706The contents of the string changes, but not the nature of the string.
1bfb14c4
JH
707Perl doesn't know any more after the call than before that the
708contents of the string indicates the affirmative.
ba62762e 709
376d9008 710Back to converting data. If you have (or want) data in your system's
a5f0baef
JH
711native 8-bit encoding (e.g. Latin-1, EBCDIC, etc.), you can use
712pack/unpack to convert to/from Unicode.
ba62762e 713
f337b084
TH
714 $native_string = pack("W*", unpack("U*", $Unicode_string));
715 $Unicode_string = pack("U*", unpack("W*", $native_string));
ba62762e
JH
716
717If you have a sequence of bytes you B<know> is valid UTF-8,
718but Perl doesn't know it yet, you can make Perl a believer, too:
719
720 use Encode 'decode_utf8';
8baee566 721 $Unicode = decode_utf8($bytes);
ba62762e 722
f337b084
TH
723or:
724
725 $Unicode = pack("U0a*", $bytes);
ae5648b3 726
ba62762e
JH
727You can convert well-formed UTF-8 to a sequence of bytes, but if
728you just want to convert random binary data into UTF-8, you can't.
1bfb14c4 729B<Any random collection of bytes isn't well-formed UTF-8>. You can
ba62762e 730use C<unpack("C*", $string)> for the former, and you can create
8baee566 731well-formed Unicode data by C<pack("U*", 0xff, ...)>.
ba62762e 732
818c4caa 733=item *
5cb3728c
RB
734
735How Do I Display Unicode? How Do I Input Unicode?
ba62762e 736
076d825e 737See http://www.alanwood.net/unicode/ and
ba62762e
JH
738http://www.cl.cam.ac.uk/~mgk25/unicode.html
739
818c4caa 740=item *
5cb3728c
RB
741
742How Does Unicode Work With Traditional Locales?
ba62762e
JH
743
744In Perl, not very well. Avoid using locales through the C<locale>
4c496f0c
JH
745pragma. Use only one or the other. But see L<perlrun> for the
746description of the C<-C> switch and its environment counterpart,
747C<$ENV{PERL_UNICODE}> to see how to enable various Unicode features,
748for example by using locale settings.
ba62762e
JH
749
750=back
751
752=head2 Hexadecimal Notation
753
376d9008
JB
754The Unicode standard prefers using hexadecimal notation because
755that more clearly shows the division of Unicode into blocks of 256 characters.
ba62762e
JH
756Hexadecimal is also simply shorter than decimal. You can use decimal
757notation, too, but learning to use hexadecimal just makes life easier
1bfb14c4 758with the Unicode standard. The C<U+HHHH> notation uses hexadecimal,
076d825e 759for example.
ba62762e
JH
760
761The C<0x> prefix means a hexadecimal number, the digits are 0-9 I<and>
762a-f (or A-F, case doesn't matter). Each hexadecimal digit represents
763four bits, or half a byte. C<print 0x..., "\n"> will show a
764hexadecimal number in decimal, and C<printf "%x\n", $decimal> will
765show a decimal number in hexadecimal. If you have just the
376d9008 766"hex digits" of a hexadecimal number, you can use the C<hex()> function.
ba62762e
JH
767
768 print 0x0009, "\n"; # 9
769 print 0x000a, "\n"; # 10
770 print 0x000f, "\n"; # 15
771 print 0x0010, "\n"; # 16
772 print 0x0011, "\n"; # 17
773 print 0x0100, "\n"; # 256
774
775 print 0x0041, "\n"; # 65
776
777 printf "%x\n", 65; # 41
778 printf "%#x\n", 65; # 0x41
779
780 print hex("41"), "\n"; # 65
781
782=head2 Further Resources
783
784=over 4
785
786=item *
787
788Unicode Consortium
789
ae5648b3 790http://www.unicode.org/
ba62762e
JH
791
792=item *
793
794Unicode FAQ
795
ae5648b3 796http://www.unicode.org/unicode/faq/
ba62762e
JH
797
798=item *
799
800Unicode Glossary
801
ae5648b3 802http://www.unicode.org/glossary/
ba62762e
JH
803
804=item *
805
806Unicode Useful Resources
807
ae5648b3 808http://www.unicode.org/unicode/onlinedat/resources.html
ba62762e
JH
809
810=item *
811
812Unicode and Multilingual Support in HTML, Fonts, Web Browsers and Other Applications
813
ae5648b3 814http://www.alanwood.net/unicode/
ba62762e
JH
815
816=item *
817
818UTF-8 and Unicode FAQ for Unix/Linux
819
ae5648b3 820http://www.cl.cam.ac.uk/~mgk25/unicode.html
ba62762e
JH
821
822=item *
823
824Legacy Character Sets
825
ae5648b3
RGS
826http://www.czyborra.com/
827http://www.eki.ee/letter/
ba62762e
JH
828
829=item *
830
831The Unicode support files live within the Perl installation in the
832directory
833
834 $Config{installprivlib}/unicore
835
ae5648b3 836in Perl 5.8.0 or newer, and
ba62762e
JH
837
838 $Config{installprivlib}/unicode
839
840in the Perl 5.6 series. (The renaming to F<lib/unicore> was done to
841avoid naming conflicts with lib/Unicode in case-insensitive filesystems.)
551b6b6f 842The main Unicode data file is F<UnicodeData.txt> (or F<Unicode.301> in
ba62762e
JH
843Perl 5.6.1.) You can find the C<$Config{installprivlib}> by
844
845 perl "-V:installprivlib"
846
ba62762e
JH
847You can explore various information from the Unicode data files using
848the C<Unicode::UCD> module.
849
850=back
851
f6edf83b
JH
852=head1 UNICODE IN OLDER PERLS
853
854If you cannot upgrade your Perl to 5.8.0 or later, you can still
855do some Unicode processing by using the modules C<Unicode::String>,
856C<Unicode::Map8>, and C<Unicode::Map>, available from CPAN.
857If you have the GNU recode installed, you can also use the
376d9008 858Perl front-end C<Convert::Recode> for character conversions.
f6edf83b 859
aaef10c5 860The following are fast conversions from ISO 8859-1 (Latin-1) bytes
63de3cb2 861to UTF-8 bytes and back, the code works even with older Perl 5 versions.
aaef10c5
JH
862
863 # ISO 8859-1 to UTF-8
864 s/([\x80-\xFF])/chr(0xC0|ord($1)>>6).chr(0x80|ord($1)&0x3F)/eg;
865
866 # UTF-8 to ISO 8859-1
867 s/([\xC2\xC3])([\x80-\xBF])/chr(ord($1)<<6&0xC0|ord($2)&0x3F)/eg;
868
ba62762e
JH
869=head1 SEE ALSO
870
2575c402 871L<perlunitut>, L<perlunicode>, L<Encode>, L<open>, L<utf8>, L<bytes>,
4c496f0c
JH
872L<perlretut>, L<perlrun>, L<Unicode::Collate>, L<Unicode::Normalize>,
873L<Unicode::UCD>
ba62762e 874
376d9008 875=head1 ACKNOWLEDGMENTS
ba62762e
JH
876
877Thanks to the kind readers of the perl5-porters@perl.org,
878perl-unicode@perl.org, linux-utf8@nl.linux.org, and unicore@unicode.org
879mailing lists for their valuable feedback.
880
881=head1 AUTHOR, COPYRIGHT, AND LICENSE
882
0f2f9b7d 883Copyright 2001-2002 Jarkko Hietaniemi E<lt>jhi@iki.fiE<gt>
ba62762e
JH
884
885This document may be distributed under the same terms as Perl itself.