This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Integrate #16731 from //depot/macperl.
[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
12Unicode is a character set standard with plans to cover all of the
13writing systems of the world, plus many other symbols.
14
15Unicode and ISO/IEC 10646 are coordinated standards that provide code
16points for the characters in almost all modern character set standards,
17covering more than 30 writing systems and hundreds of languages,
18including all commercially important modern languages. All characters
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.
22
23A Unicode I<character> is an abstract entity. It is not bound to any
24particular integer width, and especially not to the C language C<char>.
25Unicode is language neutral and display neutral: it doesn't encode the
26language of the text, and it doesn't define fonts or other graphical
27layout details. Unicode operates on characters and on text built from
28those characters.
29
30Unicode defines characters like C<LATIN CAPITAL LETTER A> or C<GREEK
31SMALL LETTER ALPHA>, and then unique numbers for those, hexadecimal
320x0041 or 0x03B1 for those particular characters. Such unique
33numbers are called I<code points>.
34
35The Unicode standard prefers using hexadecimal notation for the code
36points. (In case this notation, numbers like 0x0041, is unfamiliar to
37you, take a peek at a later section, L</"Hexadecimal Notation">.)
38The Unicode standard uses the notation C<U+0041 LATIN CAPITAL LETTER A>,
39which gives the hexadecimal code point, and the normative name of
40the character.
41
42Unicode also defines various I<properties> for the characters, like
43"uppercase" or "lowercase", "decimal digit", or "punctuation":
44these properties are independent of the names of the characters.
45Furthermore, various operations on the characters like uppercasing,
46lowercasing, and collating (sorting), are defined.
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
51a base character and modifiers is called a I<combining character
52sequence>.
53
a5f0baef
JH
54Whether to call these combining character sequences, as a whole,
55"characters" depends on your point of view. If you are a programmer,
56you probably would tend towards seeing each element in the sequences
57as one unit, one "character", but from the user viewpoint, the
58sequence as a whole is probably considered one "character", since
59that's probably what it looks like in the context of the user's
60language.
ba62762e
JH
61
62With this "as a whole" view of characters, the number of characters is
a5f0baef
JH
63open-ended. But in the programmer's "one unit is one character" point
64of view, the concept of "characters" is more deterministic, and so we
65take that point of view in this document: one "character" is one
66Unicode code point, be it a base character or a combining character.
ba62762e
JH
67
68For some of the combinations there are I<precomposed> characters,
69for example C<LATIN CAPITAL LETTER A WITH ACUTE> is defined as
70a single code point. These precomposed characters are, however,
71often available only for some combinations, and mainly they are
72meant to support round-trip conversions between Unicode and legacy
73standards (like the ISO 8859), and in general case the composing
74method is more extensible. To support conversion between the
75different compositions of the characters, various I<normalization
76forms> are also defined.
77
78Because of backward compatibility with legacy encodings, the "a unique
79number for every character" breaks down a bit: "at least one number
80for every character" is closer to truth. (This happens when the same
81character has been encoded in several legacy encodings.) The converse
82is also not true: not every code point has an assigned character.
83Firstly, there are unallocated code points within otherwise used
84blocks. Secondly, there are special Unicode control characters that
85do not represent true characters.
86
87A common myth about Unicode is that it would be "16-bit", that is,
880x10000 (or 65536) characters from 0x0000 to 0xFFFF. B<This is untrue.>
89Since Unicode 2.0 Unicode has been defined all the way up to 21 bits
90(0x10FFFF), and since 3.1 characters have been defined beyond 0xFFFF.
91The first 0x10000 characters are called the I<Plane 0>, or the I<Basic
92Multilingual Plane> (BMP). With the Unicode 3.1, 17 planes in all are
93defined (but nowhere near full of defined characters yet).
94
95Another myth is that the 256-character blocks have something to do
96with languages: a block per language. B<Also this is untrue.>
97The division into the blocks exists but it is almost completely
98accidental, an artifact of how the characters have been historically
99allocated. Instead, there is a concept called I<scripts>, which may
100be more useful: there is C<Latin> script, C<Greek> script, and so on.
101Scripts usually span several parts of several blocks. For further
102information see L<Unicode::UCD>.
103
104The Unicode code points are just abstract numbers. To input and
105output these abstract numbers, the numbers must be I<encoded> somehow.
106Unicode defines several I<character encoding forms>, of which I<UTF-8>
107is perhaps the most popular. UTF-8 is a variable length encoding that
108encodes Unicode characters as 1 to 6 bytes (only 4 with the currently
8baee566 109defined characters). Other encodings include UTF-16 and UTF-32 and their
ba62762e
JH
110big and little endian variants (UTF-8 is byteorder independent).
111The ISO/IEC 10646 defines the UCS-2 and UCS-4 encoding forms.
112
113For more information about encodings, for example to learn what
114I<surrogates> and I<byte order marks> (BOMs) are, see L<perlunicode>.
115
116=head2 Perl's Unicode Support
117
118Starting from Perl 5.6.0, Perl has had the capability of handling
119Unicode natively. The first recommended release for serious Unicode
120work is Perl 5.8.0, however. The maintenance release 5.6.1 fixed many
121of the problems of the initial implementation of Unicode, but for
122example regular expressions didn't really work with Unicode.
123
124B<Starting from Perl 5.8.0, the use of C<use utf8> is no longer
125necessary.> In earlier releases the C<utf8> pragma was used to declare
126that operations in the current block or file would be Unicode-aware.
127This model was found to be wrong, or at least clumsy: the Unicodeness
128is now carried with the data, not attached to the operations. (There
129is one remaining case where an explicit C<use utf8> is needed: if your
a5f0baef 130Perl script itself is encoded in UTF-8, you can use UTF-8 in your
c20e2abd
JH
131identifier names, and in your string and regular expression literals,
132by saying C<use utf8>. This is not the default because that would
133break existing scripts having legacy 8-bit data in them.)
ba62762e
JH
134
135=head2 Perl's Unicode Model
136
137Perl supports both the old, pre-5.6, model of strings of eight-bit
138native bytes, and strings of Unicode characters. The principle is
139that Perl tries to keep its data as eight-bit bytes for as long as
140possible, but as soon as Unicodeness cannot be avoided, the data is
141transparently upgraded to Unicode.
142
4192de81
JH
143Internally, Perl currently uses either whatever the native eight-bit
144character set of the platform (for example Latin-1) or UTF-8 to encode
145Unicode strings. Specifically, if all code points in the string are
a5f0baef
JH
1460xFF or less, Perl uses the native eight-bit character set.
147Otherwise, it uses UTF-8.
4192de81 148
7ca610e8 149A user of Perl does not normally need to know nor care how Perl
20ba30f4 150happens to encode its internal strings, but it becomes relevant when
7ca610e8
JH
151outputting Unicode strings to a stream without a discipline (one with
152the "default default"). In such a case, the raw bytes used internally
153(the native character set or UTF-8, as appropriate for each string)
154will be used, and a "Wide character" warning will be issued if those
155strings contain a character beyond 0x00FF.
4192de81
JH
156
157For example,
158
7ca610e8 159 perl -e 'print "\x{DF}\n", "\x{0100}\x{DF}\n"'
4192de81
JH
160
161produces a fairly useless mixture of native bytes and UTF-8, as well
162as a warning.
163
164To output UTF-8 always, use the ":utf8" output discipline. Prepending
165
166 binmode(STDOUT, ":utf8");
167
168to this sample program ensures the output is completely UTF-8, and
4049dcd4 169of course, removes the warning.
ba62762e 170
b310b053
JH
171If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
172contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
173the default encoding of your STDIN, STDOUT, and STDERR, and of
ac730995
JH
174B<any subsequent file open>, is UTF-8. Note that this means
175that Perl expects other software to work, too: if STDIN coming
176in from another command is not UTF-8, Perl will complain about
177malformed UTF-8.
b310b053 178
64c66fb6
JH
179=head2 Unicode and EBCDIC
180
181Perl 5.8.0 also supports Unicode on EBCDIC platforms. There,
182the Unicode support is somewhat more complex to implement since
183additional conversions are needed at every step. Some problems
dc4af4bb 184remain, see L<perlebcdic> for details.
64c66fb6
JH
185
186In any case, the Unicode support on EBCDIC platforms is better than
187in the 5.6 series, which didn't work much at all for EBCDIC platform.
188On EBCDIC platforms, the internal Unicode encoding form is UTF-EBCDIC
189instead of UTF-8 (the difference is that as UTF-8 is "ASCII-safe" in
190that ASCII characters encode to UTF-8 as-is, UTF-EBCDIC is
191"EBCDIC-safe").
ba62762e
JH
192
193=head2 Creating Unicode
194
a5f0baef
JH
195To create Unicode characters in literals for code points above 0xFF,
196use the C<\x{...}> notation in doublequoted strings:
ba62762e
JH
197
198 my $smiley = "\x{263a}";
199
8baee566 200Similarly in regular expression literals
ba62762e
JH
201
202 $smiley =~ /\x{263a}/;
203
204At run-time you can use C<chr()>:
205
206 my $hebrew_alef = chr(0x05d0);
207
208(See L</"Further Resources"> for how to find all these numeric codes.)
209
210Naturally, C<ord()> will do the reverse: turn a character to a code point.
211
a5f0baef
JH
212Note that C<\x..> (no C<{}> and only two hexadecimal digits),
213C<\x{...}>, and C<chr(...)> for arguments less than 0x100 (decimal
214256) generate an eight-bit character for backward compatibility with
215older Perls. For arguments of 0x100 or more, Unicode characters are
216always produced. If you want to force the production of Unicode
217characters regardless of the numeric value, use C<pack("U", ...)>
218instead of C<\x..>, C<\x{...}>, or C<chr()>.
ba62762e
JH
219
220You can also use the C<charnames> pragma to invoke characters
221by name in doublequoted strings:
222
223 use charnames ':full';
224 my $arabic_alef = "\N{ARABIC LETTER ALEF}";
225
226And, as mentioned above, you can also C<pack()> numbers into Unicode
227characters:
228
229 my $georgian_an = pack("U", 0x10a0);
230
8a5e5dd5
JH
231Note that both C<\x{...}> and C<\N{...}> are compile-time string
232constants: you cannot use variables in them. if you want similar
233run-time functionality, use C<chr()> and C<charnames::vianame()>.
234
1eda90df
JH
235Also note that if all the code points for pack "U" are below 0x100,
236bytes will be generated, just like if you were using C<chr()>.
237
238 my $bytes = pack("U*", 0x80, 0xFF);
239
240If you want to force the result to Unicode characters, use the special
241C<"U0"> prefix. It consumes no arguments but forces the result to be
242in Unicode characters, instead of bytes.
243
244 my $chars = pack("U0U*", 0x80, 0xFF);
245
ba62762e
JH
246=head2 Handling Unicode
247
248Handling Unicode is for the most part transparent: just use the
249strings as usual. Functions like C<index()>, C<length()>, and
250C<substr()> will work on the Unicode characters; regular expressions
251will work on the Unicode characters (see L<perlunicode> and L<perlretut>).
252
253Note that Perl does B<not> consider combining character sequences
254to be characters, such for example
255
256 use charnames ':full';
257 print length("\N{LATIN CAPITAL LETTER A}\N{COMBINING ACUTE ACCENT}"), "\n";
258
259will print 2, not 1. The only exception is that regular expressions
260have C<\X> for matching a combining character sequence.
261
262When life is not quite so transparent is working with legacy
263encodings, and I/O, and certain special cases.
264
265=head2 Legacy Encodings
266
267When you combine legacy data and Unicode the legacy data needs
268to be upgraded to Unicode. Normally ISO 8859-1 (or EBCDIC, if
269applicable) is assumed. You can override this assumption by
270using the C<encoding> pragma, for example
271
272 use encoding 'latin2'; # ISO 8859-2
273
274in which case literals (string or regular expression) and chr/ord
275in your whole script are assumed to produce Unicode characters from
276ISO 8859-2 code points. Note that the matching for the encoding
277names is forgiving: instead of C<latin2> you could have said
278C<Latin 2>, or C<iso8859-2>, and so forth. With just
279
280 use encoding;
281
47b382ca
JH
282the environment variable C<PERL_ENCODING> will be consulted,
283but if that doesn't exist, the encoding pragma fails.
ba62762e
JH
284
285The C<Encode> module knows about many encodings and it has interfaces
286for doing conversions between those encodings:
287
288 use Encode 'from_to';
289 from_to($data, "iso-8859-3", "utf-8"); # from legacy to utf-8
290
291=head2 Unicode I/O
292
8baee566 293Normally, writing out Unicode data
ba62762e 294
8baee566 295 print FH $some_string_with_unicode, "\n";
ba62762e 296
8baee566 297produces raw bytes that Perl happens to use to internally encode the
a5f0baef
JH
298Unicode string (which depends on the system, as well as what
299characters happen to be in the string at the time). If any of the
7ca610e8
JH
300characters are at code points 0x100 or above, you will get a warning.
301To ensure that the output is explicitly rendered in the encoding you
302desire (and to avoid the warning), open the stream with the desired
303encoding. Some examples:
ba62762e 304
ec90690f
ST
305 open FH, ">:utf8", "file";
306
307 open FH, ">:encoding(ucs2)", "file";
308 open FH, ">:encoding(UTF-8)", "file";
309 open FH, ">:encoding(shift_jis)", "file";
1d7919c5
JH
310
311and on already open streams use C<binmode()>:
312
313 binmode(STDOUT, ":utf8");
314
ec90690f
ST
315 binmode(STDOUT, ":encoding(ucs2)");
316 binmode(STDOUT, ":encoding(UTF-8)");
317 binmode(STDOUT, ":encoding(shift_jis)");
318
b5d8778e
JH
319The matching of encoding names is loose: case does not matter, and
320many encodings have several aliases. Note that C<:utf8> discipline
321must always be specified exactly like that, it is not subject to the
322loose matching of encoding names.
323
ec90690f
ST
324See L<PerlIO> for the C<:utf8> layer;
325L<PerlIO::encoding> and L<Encode::PerlIO> for the C<:encoding()> layer;
326L<Encode::Supported> for many encodings supported by the C<Encode> module.
ba62762e 327
a5f0baef
JH
328Reading in a file that you know happens to be encoded in one of the
329Unicode encodings does not magically turn the data into Unicode in
330Perl's eyes. To do that, specify the appropriate discipline when
331opening files
ba62762e
JH
332
333 open(my $fh,'<:utf8', 'anything');
8baee566
JF
334 my $line_of_unicode = <$fh>;
335
ec90690f 336 open(my $fh,'<:encoding(Big5)', 'anything');
8baee566 337 my $line_of_unicode = <$fh>;
ba62762e
JH
338
339The I/O disciplines can also be specified more flexibly with
340the C<open> pragma; see L<open>:
341
1d7919c5
JH
342 use open ':utf8'; # input and output default discipline will be UTF-8
343 open X, ">file";
344 print X chr(0x100), "\n";
ba62762e 345 close X;
1d7919c5 346 open Y, "<file";
ba62762e
JH
347 printf "%#x\n", ord(<Y>); # this should print 0x100
348 close Y;
349
350With the C<open> pragma you can use the C<:locale> discipline
351
1ecefa54
JH
352 $ENV{LC_ALL} = $ENV{LANG} = 'ru_RU.KOI8-R';
353 # the :locale will probe the locale environment variables like LC_ALL
ba62762e
JH
354 use open OUT => ':locale'; # russki parusski
355 open(O, ">koi8");
356 print O chr(0x430); # Unicode CYRILLIC SMALL LETTER A = KOI8-R 0xc1
357 close O;
358 open(I, "<koi8");
359 printf "%#x\n", ord(<I>), "\n"; # this should print 0xc1
360 close I;
361
362or you can also use the C<':encoding(...)'> discipline
363
364 open(my $epic,'<:encoding(iso-8859-7)','iliad.greek');
8baee566 365 my $line_of_unicode = <$epic>;
ba62762e 366
8baee566
JF
367These methods install a transparent filter on the I/O stream that
368converts data from the specified encoding when it is read in from the
a5f0baef 369stream. The result is always Unicode.
ba62762e
JH
370
371The L<open> pragma affects all the C<open()> calls after the pragma by
372setting default disciplines. If you want to affect only certain
373streams, use explicit disciplines directly in the C<open()> call.
374
375You can switch encodings on an already opened stream by using
8baee566 376C<binmode()>; see L<perlfunc/binmode>.
ba62762e 377
1ecefa54
JH
378The C<:locale> does not currently (as of Perl 5.8.0) work with
379C<open()> and C<binmode()>, only with the C<open> pragma. The
8baee566 380C<:utf8> and C<:encoding(...)> methods do work with all of C<open()>,
1ecefa54 381C<binmode()>, and the C<open> pragma.
ba62762e 382
8baee566 383Similarly, you may use these I/O disciplines on output streams to
a5f0baef
JH
384automatically convert Unicode to the specified encoding when it is
385written to the stream. For example, the following snippet copies the
386contents of the file "text.jis" (encoded as ISO-2022-JP, aka JIS) to
387the file "text.utf8", encoded as UTF-8:
ba62762e 388
8baee566
JF
389 open(my $nihongo, '<:encoding(iso2022-jp)', 'text.jis');
390 open(my $unicode, '>:utf8', 'text.utf8');
391 while (<$nihongo>) { print $unicode }
ba62762e
JH
392
393The naming of encodings, both by the C<open()> and by the C<open>
394pragma, is similarly understanding as with the C<encoding> pragma:
395C<koi8-r> and C<KOI8R> will both be understood.
396
397Common encodings recognized by ISO, MIME, IANA, and various other
8baee566 398standardisation organisations are recognised; for a more detailed
ba62762e
JH
399list see L<Encode>.
400
401C<read()> reads characters and returns the number of characters.
402C<seek()> and C<tell()> operate on byte counts, as do C<sysread()>
403and C<sysseek()>.
404
8baee566
JF
405Notice that because of the default behaviour of not doing any
406conversion upon input if there is no default discipline,
ba62762e 407it is easy to mistakenly write code that keeps on expanding a file
8baee566 408by repeatedly encoding:
ba62762e
JH
409
410 # BAD CODE WARNING
411 open F, "file";
8baee566 412 local $/; ## read in the whole file of 8-bit characters
ba62762e
JH
413 $t = <F>;
414 close F;
415 open F, ">:utf8", "file";
8baee566 416 print F $t; ## convert to UTF-8 on output
ba62762e
JH
417 close F;
418
419If you run this code twice, the contents of the F<file> will be twice
1d7919c5
JH
420UTF-8 encoded. A C<use open ':utf8'> would have avoided the bug, or
421explicitly opening also the F<file> for input as UTF-8.
ba62762e 422
0c901d84
JH
423B<NOTE>: the C<:utf8> and C<:encoding> features work only if your
424Perl has been built with the new "perlio" feature. Almost all
425Perl 5.8 platforms do use "perlio", though: you can see whether
426yours is by running "perl -V" and looking for C<useperlio=define>.
427
1ecefa54
JH
428=head2 Displaying Unicode As Text
429
430Sometimes you might want to display Perl scalars containing Unicode as
8baee566 431simple ASCII (or EBCDIC) text. The following subroutine converts
1ecefa54
JH
432its argument so that Unicode characters with code points greater than
433255 are displayed as "\x{...}", control characters (like "\n") are
8baee566 434displayed as "\x..", and the rest of the characters as themselves:
1ecefa54 435
58c274a1
JF
436 sub nice_string {
437 join("",
438 map { $_ > 255 ? # if wide character...
8baee566 439 sprintf("\\x{%04X}", $_) : # \x{...}
58c274a1 440 chr($_) =~ /[[:cntrl:]]/ ? # else if control character ...
8baee566 441 sprintf("\\x%02X", $_) : # \x..
58c274a1
JF
442 chr($_) # else as themselves
443 } unpack("U*", $_[0])); # unpack Unicode characters
444 }
445
446For example,
447
448 nice_string("foo\x{100}bar\n")
449
8baee566 450returns:
58c274a1 451
8baee566 452 "foo\x{0100}bar\x0A"
1ecefa54 453
ba62762e
JH
454=head2 Special Cases
455
456=over 4
457
458=item *
459
460Bit Complement Operator ~ And vec()
461
1d9bf7bb
JF
462The bit complement operator C<~> may produce surprising results if used on
463strings containing characters with ordinal values above 255. In such a
464case, the results are consistent with the internal encoding of the
465characters, but not with much else. So don't do that. Similarly for vec():
466you will be operating on the internally encoded bit patterns of the Unicode
467characters, not on the code point values, which is very probably not what
468you want.
ba62762e
JH
469
470=item *
471
8baee566
JF
472Peeking At Perl's Internal Encoding
473
474Normal users of Perl should never care how Perl encodes any particular
a5f0baef
JH
475Unicode string (because the normal ways to get at the contents of a
476string with Unicode -- via input and output -- should always be via
477explicitly-defined I/O disciplines). But if you must, there are two
478ways of looking behind the scenes.
ba62762e
JH
479
480One way of peeking inside the internal encoding of Unicode characters
481is to use C<unpack("C*", ...> to get the bytes, or C<unpack("H*", ...)>
482to display the bytes:
483
8baee566 484 # this prints c4 80 for the UTF-8 bytes 0xc4 0x80
ba62762e
JH
485 print join(" ", unpack("H*", pack("U", 0x100))), "\n";
486
487Yet another way would be to use the Devel::Peek module:
488
489 perl -MDevel::Peek -e 'Dump(chr(0x100))'
490
8baee566 491That shows the UTF8 flag in FLAGS and both the UTF-8 bytes
ba62762e
JH
492and Unicode characters in PV. See also later in this document
493the discussion about the C<is_utf8> function of the C<Encode> module.
494
495=back
496
497=head2 Advanced Topics
498
499=over 4
500
501=item *
502
503String Equivalence
504
505The question of string equivalence turns somewhat complicated
506in Unicode: what do you mean by equal?
507
07698885
RGS
508(Is C<LATIN CAPITAL LETTER A WITH ACUTE> equal to
509C<LATIN CAPITAL LETTER A>?)
ba62762e 510
a5f0baef
JH
511The short answer is that by default Perl compares equivalence (C<eq>,
512C<ne>) based only on code points of the characters. In the above
513case, the answer is no (because 0x00C1 != 0x0041). But sometimes any
ba62762e
JH
514CAPITAL LETTER As being considered equal, or even any As of any case,
515would be desirable.
516
517The long answer is that you need to consider character normalization
518and casing issues: see L<Unicode::Normalize>, and Unicode Technical
519Reports #15 and #21, I<Unicode Normalization Forms> and I<Case
520Mappings>, http://www.unicode.org/unicode/reports/tr15/
521http://www.unicode.org/unicode/reports/tr21/
522
58c274a1 523As of Perl 5.8.0, regular expression case-ignoring matching
ba62762e
JH
524implements only 1:1 semantics: one character matches one character.
525In I<Case Mappings> both 1:N and N:1 matches are defined.
526
527=item *
528
529String Collation
530
531People like to see their strings nicely sorted, or as Unicode
532parlance goes, collated. But again, what do you mean by collate?
533
07698885
RGS
534(Does C<LATIN CAPITAL LETTER A WITH ACUTE> come before or after
535C<LATIN CAPITAL LETTER A WITH GRAVE>?)
ba62762e 536
58c274a1 537The short answer is that by default, Perl compares strings (C<lt>,
ba62762e 538C<le>, C<cmp>, C<ge>, C<gt>) based only on the code points of the
58c274a1 539characters. In the above case, the answer is "after", since 0x00C1 > 0x00C0.
ba62762e
JH
540
541The long answer is that "it depends", and a good answer cannot be
542given without knowing (at the very least) the language context.
543See L<Unicode::Collate>, and I<Unicode Collation Algorithm>
544http://www.unicode.org/unicode/reports/tr10/
545
546=back
547
548=head2 Miscellaneous
549
550=over 4
551
552=item *
553
3ff56b75 554Character Ranges and Classes
ba62762e
JH
555
556Character ranges in regular expression character classes (C</[a-z]/>)
557and in the C<tr///> (also known as C<y///>) operator are not magically
58c274a1 558Unicode-aware. What this means that C<[A-Za-z]> will not magically start
ba62762e 559to mean "all alphabetic letters" (not that it does mean that even for
3ff56b75 5608-bit characters, you should be using C</[[:alpha:]]/> for that).
ba62762e 561
a5f0baef
JH
562For specifying things like that in regular expressions, you can use
563the various Unicode properties, C<\pL> or perhaps C<\p{Alphabetic}>,
564in this particular case. You can use Unicode code points as the end
565points of character ranges, but that means that particular code point
3ff56b75
JH
566range, nothing more. For further information (there are dozens
567of Unicode character classes), see L<perlunicode>.
ba62762e
JH
568
569=item *
570
571String-To-Number Conversions
572
573Unicode does define several other decimal (and numeric) characters
574than just the familiar 0 to 9, such as the Arabic and Indic digits.
575Perl does not support string-to-number conversion for digits other
58c274a1 576than ASCII 0 to 9 (and ASCII a to f for hexadecimal).
ba62762e
JH
577
578=back
579
580=head2 Questions With Answers
581
582=over 4
583
5cb3728c
RB
584=item
585
586Will My Old Scripts Break?
ba62762e
JH
587
588Very probably not. Unless you are generating Unicode characters
589somehow, any old behaviour should be preserved. About the only
590behaviour that has changed and which could start generating Unicode
591is the old behaviour of C<chr()> where supplying an argument more
592than 255 produced a character modulo 255 (for example, C<chr(300)>
593was equal to C<chr(45)>).
594
5cb3728c
RB
595=item
596
597How Do I Make My Scripts Work With Unicode?
ba62762e
JH
598
599Very little work should be needed since nothing changes until you
600somehow generate Unicode data. The greatest trick will be getting
601input as Unicode, and for that see the earlier I/O discussion.
602
5cb3728c
RB
603=item
604
605How Do I Know Whether My String Is In Unicode?
ba62762e
JH
606
607You shouldn't care. No, you really shouldn't. If you have
608to care (beyond the cases described above), it means that we
609didn't get the transparency of Unicode quite right.
610
611Okay, if you insist:
612
613 use Encode 'is_utf8';
614 print is_utf8($string) ? 1 : 0, "\n";
615
616But note that this doesn't mean that any of the characters in the
617string are necessary UTF-8 encoded, or that any of the characters have
618code points greater than 0xFF (255) or even 0x80 (128), or that the
619string has any characters at all. All the C<is_utf8()> does is to
620return the value of the internal "utf8ness" flag attached to the
3c1c8017
AT
621$string. If the flag is off, the bytes in the scalar are interpreted
622as a single byte encoding. If the flag is on, the bytes in the scalar
623are interpreted as the (multibyte, variable-length) UTF-8 encoded code
624points of the characters. Bytes added to an UTF-8 encoded string are
625automatically upgraded to UTF-8. If mixed non-UTF8 and UTF-8 scalars
626are merged (doublequoted interpolation, explicit concatenation, and
627printf/sprintf parameter substitution), the result will be UTF-8 encoded
628as if copies of the byte strings were upgraded to UTF-8: for example,
629
630 $a = "ab\x80c";
631 $b = "\x{100}";
632 print "$a = $b\n";
633
634the output string will be UTF-8-encoded "ab\x80c\x{100}\n", but note
635that C<$a> will stay single byte encoded.
ba62762e
JH
636
637Sometimes you might really need to know the byte length of a string
ce7675db
A
638instead of the character length. For that use either the
639C<Encode::encode_utf8()> function or the C<bytes> pragma and its only
640defined function C<length()>:
ba62762e
JH
641
642 my $unicode = chr(0x100);
643 print length($unicode), "\n"; # will print 1
ce7675db
A
644 require Encode;
645 print length(Encode::encode_utf8($unicode)), "\n"; # will print 2
ba62762e 646 use bytes;
ce7675db 647 print length($unicode), "\n"; # will also print 2 (the 0xC4 0x80 of the UTF-8)
ba62762e 648
5cb3728c
RB
649=item
650
651How Do I Detect Data That's Not Valid In a Particular Encoding?
ba62762e 652
8baee566
JF
653Use the C<Encode> package to try converting it.
654For example,
ba62762e
JH
655
656 use Encode 'encode_utf8';
8baee566 657 if (encode_utf8($string_of_bytes_that_I_think_is_utf8)) {
ba62762e
JH
658 # valid
659 } else {
660 # invalid
661 }
662
8baee566 663For UTF-8 only, you can use:
ba62762e
JH
664
665 use warnings;
8baee566 666 @chars = unpack("U0U*", $string_of_bytes_that_I_think_is_utf8);
ba62762e 667
8baee566
JF
668If invalid, a C<Malformed UTF-8 character (byte 0x##) in
669unpack> is produced. The "U0" means "expect strictly UTF-8
1eda90df
JH
670encoded Unicode". Without that the C<unpack("U*", ...)>
671would accept also data like C<chr(0xFF>), similarly to the
672C<pack> as we saw earlier.
ba62762e 673
5cb3728c
RB
674=item
675
676How Do I Convert Binary Data Into a Particular Encoding, Or Vice Versa?
ba62762e 677
8baee566
JF
678This probably isn't as useful as you might think.
679Normally, you shouldn't need to.
ba62762e 680
a5f0baef
JH
681In one sense, what you are asking doesn't make much sense: Encodings
682are for characters, and binary data is not "characters", so converting
683"data" into some encoding isn't meaningful unless you know in what
684character set and encoding the binary data is in, in which case it's
685not binary data, now is it?
8baee566
JF
686
687If you have a raw sequence of bytes that you know should be interpreted via
688a particular encoding, you can use C<Encode>:
ba62762e
JH
689
690 use Encode 'from_to';
691 from_to($data, "iso-8859-1", "utf-8"); # from latin-1 to utf-8
692
8baee566
JF
693The call to from_to() changes the bytes in $data, but nothing material
694about the nature of the string has changed as far as Perl is concerned.
695Both before and after the call, the string $data contains just a bunch of
6968-bit bytes. As far as Perl is concerned, the encoding of the string (as
697Perl sees it) remains as "system-native 8-bit bytes".
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
JF
706The contents of the string changes, but not the nature of the string.
707Perl doesn't know any more after the call than before that the contents
708of the string indicates the affirmative.
ba62762e 709
a5f0baef
JH
710Back to converting data, if you have (or want) data in your system's
711native 8-bit encoding (e.g. Latin-1, EBCDIC, etc.), you can use
712pack/unpack to convert to/from Unicode.
ba62762e 713
8baee566
JF
714 $native_string = pack("C*", unpack("U*", $Unicode_string));
715 $Unicode_string = pack("U*", unpack("C*", $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
JH
722
723You can convert well-formed UTF-8 to a sequence of bytes, but if
724you just want to convert random binary data into UTF-8, you can't.
725Any random collection of bytes isn't well-formed UTF-8. You can
726use C<unpack("C*", $string)> for the former, and you can create
8baee566 727well-formed Unicode data by C<pack("U*", 0xff, ...)>.
ba62762e 728
5cb3728c
RB
729=item
730
731How Do I Display Unicode? How Do I Input Unicode?
ba62762e 732
076d825e 733See http://www.alanwood.net/unicode/ and
ba62762e
JH
734http://www.cl.cam.ac.uk/~mgk25/unicode.html
735
5cb3728c
RB
736=item
737
738How Does Unicode Work With Traditional Locales?
ba62762e
JH
739
740In Perl, not very well. Avoid using locales through the C<locale>
741pragma. Use only one or the other.
742
743=back
744
745=head2 Hexadecimal Notation
746
747The Unicode standard prefers using hexadecimal notation because that
748shows better the division of Unicode into blocks of 256 characters.
749Hexadecimal is also simply shorter than decimal. You can use decimal
750notation, too, but learning to use hexadecimal just makes life easier
076d825e
JH
751with the Unicode standard. The "U+HHHH" notation uses hexadecimal,
752for example.
ba62762e
JH
753
754The C<0x> prefix means a hexadecimal number, the digits are 0-9 I<and>
755a-f (or A-F, case doesn't matter). Each hexadecimal digit represents
756four bits, or half a byte. C<print 0x..., "\n"> will show a
757hexadecimal number in decimal, and C<printf "%x\n", $decimal> will
758show a decimal number in hexadecimal. If you have just the
a5f0baef 759"hexdigits" of a hexadecimal number, you can use the C<hex()> function.
ba62762e
JH
760
761 print 0x0009, "\n"; # 9
762 print 0x000a, "\n"; # 10
763 print 0x000f, "\n"; # 15
764 print 0x0010, "\n"; # 16
765 print 0x0011, "\n"; # 17
766 print 0x0100, "\n"; # 256
767
768 print 0x0041, "\n"; # 65
769
770 printf "%x\n", 65; # 41
771 printf "%#x\n", 65; # 0x41
772
773 print hex("41"), "\n"; # 65
774
775=head2 Further Resources
776
777=over 4
778
779=item *
780
781Unicode Consortium
782
783 http://www.unicode.org/
784
785=item *
786
787Unicode FAQ
788
789 http://www.unicode.org/unicode/faq/
790
791=item *
792
793Unicode Glossary
794
795 http://www.unicode.org/glossary/
796
797=item *
798
799Unicode Useful Resources
800
801 http://www.unicode.org/unicode/onlinedat/resources.html
802
803=item *
804
805Unicode and Multilingual Support in HTML, Fonts, Web Browsers and Other Applications
806
076d825e 807 http://www.alanwood.net/unicode/
ba62762e
JH
808
809=item *
810
811UTF-8 and Unicode FAQ for Unix/Linux
812
813 http://www.cl.cam.ac.uk/~mgk25/unicode.html
814
815=item *
816
817Legacy Character Sets
818
819 http://www.czyborra.com/
820 http://www.eki.ee/letter/
821
822=item *
823
824The Unicode support files live within the Perl installation in the
825directory
826
827 $Config{installprivlib}/unicore
828
829in Perl 5.8.0 or newer, and
830
831 $Config{installprivlib}/unicode
832
833in the Perl 5.6 series. (The renaming to F<lib/unicore> was done to
834avoid naming conflicts with lib/Unicode in case-insensitive filesystems.)
551b6b6f 835The main Unicode data file is F<UnicodeData.txt> (or F<Unicode.301> in
ba62762e
JH
836Perl 5.6.1.) You can find the C<$Config{installprivlib}> by
837
838 perl "-V:installprivlib"
839
ba62762e
JH
840You can explore various information from the Unicode data files using
841the C<Unicode::UCD> module.
842
843=back
844
f6edf83b
JH
845=head1 UNICODE IN OLDER PERLS
846
847If you cannot upgrade your Perl to 5.8.0 or later, you can still
848do some Unicode processing by using the modules C<Unicode::String>,
849C<Unicode::Map8>, and C<Unicode::Map>, available from CPAN.
850If you have the GNU recode installed, you can also use the
851Perl frontend C<Convert::Recode> for character conversions.
852
aaef10c5
JH
853The following are fast conversions from ISO 8859-1 (Latin-1) bytes
854to UTF-8 bytes, the code works even with older Perl 5 versions.
855
856 # ISO 8859-1 to UTF-8
857 s/([\x80-\xFF])/chr(0xC0|ord($1)>>6).chr(0x80|ord($1)&0x3F)/eg;
858
859 # UTF-8 to ISO 8859-1
860 s/([\xC2\xC3])([\x80-\xBF])/chr(ord($1)<<6&0xC0|ord($2)&0x3F)/eg;
861
ba62762e
JH
862=head1 SEE ALSO
863
864L<perlunicode>, L<Encode>, L<encoding>, L<open>, L<utf8>, L<bytes>,
865L<perlretut>, L<Unicode::Collate>, L<Unicode::Normalize>, L<Unicode::UCD>
866
867=head1 ACKNOWLEDGEMENTS
868
869Thanks to the kind readers of the perl5-porters@perl.org,
870perl-unicode@perl.org, linux-utf8@nl.linux.org, and unicore@unicode.org
871mailing lists for their valuable feedback.
872
873=head1 AUTHOR, COPYRIGHT, AND LICENSE
874
be3c0a43 875Copyright 2001-2002 Jarkko Hietaniemi <jhi@iki.fi>
ba62762e
JH
876
877This document may be distributed under the same terms as Perl itself.