Commit | Line | Data |
ba62762e |
1 | =head1 NAME |
2 | |
07fcf8ff |
3 | perluniintro - Perl Unicode introduction |
ba62762e |
4 | |
5 | =head1 DESCRIPTION |
6 | |
7 | This document gives a general idea of Unicode and how to use Unicode |
8 | in Perl. |
9 | |
10 | =head2 Unicode |
11 | |
376d9008 |
12 | Unicode is a character set standard which plans to codify all of the |
ba62762e |
13 | writing systems of the world, plus many other symbols. |
14 | |
15 | Unicode and ISO/IEC 10646 are coordinated standards that provide code |
376d9008 |
16 | points for characters in almost all modern character set standards, |
ba62762e |
17 | covering more than 30 writing systems and hundreds of languages, |
376d9008 |
18 | including all commercially-important modern languages. All characters |
ba62762e |
19 | in the largest Chinese, Japanese, and Korean dictionaries are also |
20 | encoded. The standards will eventually cover almost all characters in |
21 | more than 250 writing systems and thousands of languages. |
4c496f0c |
22 | Unicode 1.0 was released in October 1991, and 4.0 in April 2003. |
ba62762e |
23 | |
24 | A Unicode I<character> is an abstract entity. It is not bound to any |
376d9008 |
25 | particular integer width, especially not to the C language C<char>. |
26 | Unicode is language-neutral and display-neutral: it does not encode the |
27 | language of the text and it does not define fonts or other graphical |
ba62762e |
28 | layout details. Unicode operates on characters and on text built from |
29 | those characters. |
30 | |
31 | Unicode defines characters like C<LATIN CAPITAL LETTER A> or C<GREEK |
376d9008 |
32 | SMALL LETTER ALPHA> and unique numbers for the characters, in this |
33 | case 0x0041 and 0x03B1, respectively. These unique numbers are called |
34 | I<code points>. |
ba62762e |
35 | |
36 | The Unicode standard prefers using hexadecimal notation for the code |
4c496f0c |
37 | points. If numbers like C<0x0041> are unfamiliar to you, take a peek |
38 | at a later section, L</"Hexadecimal Notation">. The Unicode standard |
39 | uses the notation C<U+0041 LATIN CAPITAL LETTER A>, to give the |
40 | hexadecimal code point and the normative name of the character. |
ba62762e |
41 | |
42 | Unicode also defines various I<properties> for the characters, like |
376d9008 |
43 | "uppercase" or "lowercase", "decimal digit", or "punctuation"; |
ba62762e |
44 | these properties are independent of the names of the characters. |
45 | Furthermore, various operations on the characters like uppercasing, |
376d9008 |
46 | lowercasing, and collating (sorting) are defined. |
ba62762e |
47 | |
48 | A Unicode character consists either of a single code point, or a |
49 | I<base character> (like C<LATIN CAPITAL LETTER A>), followed by one or |
50 | more I<modifiers> (like C<COMBINING ACUTE ACCENT>). This sequence of |
376d9008 |
51 | base character and modifiers is called a I<combining character |
ba62762e |
52 | sequence>. |
53 | |
376d9008 |
54 | Whether to call these combining character sequences "characters" |
55 | depends on your point of view. If you are a programmer, you probably |
56 | would tend towards seeing each element in the sequences as one unit, |
57 | or "character". The whole sequence could be seen as one "character", |
58 | however, from the user's point of view, since that's probably what it |
59 | looks like in the context of the user's language. |
60 | |
61 | With this "whole sequence" view of characters, the total number of |
62 | characters is open-ended. But in the programmer's "one unit is one |
63 | character" point of view, the concept of "characters" is more |
1bfb14c4 |
64 | deterministic. In this document, we take that second point of view: |
65 | one "character" is one Unicode code point, be it a base character or |
66 | a combining character. |
376d9008 |
67 | |
68 | For some combinations, there are I<precomposed> characters. |
69 | C<LATIN CAPITAL LETTER A WITH ACUTE>, for example, is defined as |
ba62762e |
70 | a single code point. These precomposed characters are, however, |
376d9008 |
71 | only available for some combinations, and are mainly |
ba62762e |
72 | meant to support round-trip conversions between Unicode and legacy |
376d9008 |
73 | standards (like the ISO 8859). In the general case, the composing |
74 | method is more extensible. To support conversion between |
ba62762e |
75 | different compositions of the characters, various I<normalization |
376d9008 |
76 | forms> to standardize representations are also defined. |
ba62762e |
77 | |
78 | Because of backward compatibility with legacy encodings, the "a unique |
376d9008 |
79 | number for every character" idea breaks down a bit: instead, there is |
80 | "at least one number for every character". The same character could |
81 | be represented differently in several legacy encodings. The |
82 | converse is also not true: some code points do not have an assigned |
83 | character. Firstly, there are unallocated code points within |
84 | otherwise used blocks. Secondly, there are special Unicode control |
85 | characters that do not represent true characters. |
ba62762e |
86 | |
87 | A common myth about Unicode is that it would be "16-bit", that is, |
376d9008 |
88 | Unicode is only represented as C<0x10000> (or 65536) characters from |
4c496f0c |
89 | C<0x0000> to C<0xFFFF>. B<This is untrue.> Since Unicode 2.0 (July |
90 | 1996), Unicode has been defined all the way up to 21 bits (C<0x10FFFF>), |
91 | and since Unicode 3.1 (March 2001), characters have been defined |
92 | beyond C<0xFFFF>. The first C<0x10000> characters are called the |
93 | I<Plane 0>, or the I<Basic Multilingual Plane> (BMP). With Unicode |
94 | 3.1, 17 (yes, seventeen) planes in all were defined--but they are |
95 | nowhere near full of defined characters, yet. |
ba62762e |
96 | |
1bfb14c4 |
97 | Another myth is that the 256-character blocks have something to |
98 | do with languages--that each block would define the characters used |
99 | by a language or a set of languages. B<This is also untrue.> |
100 | The division into blocks exists, but it is almost completely |
101 | accidental--an artifact of how the characters have been and |
102 | still are allocated. Instead, there is a concept called I<scripts>, |
103 | which is more useful: there is C<Latin> script, C<Greek> script, and |
104 | so on. Scripts usually span varied parts of several blocks. |
105 | For further information see L<Unicode::UCD>. |
ba62762e |
106 | |
107 | The Unicode code points are just abstract numbers. To input and |
4c496f0c |
108 | output these abstract numbers, the numbers must be I<encoded> or |
109 | I<serialised> somehow. Unicode defines several I<character encoding |
110 | forms>, of which I<UTF-8> is perhaps the most popular. UTF-8 is a |
111 | variable length encoding that encodes Unicode characters as 1 to 6 |
112 | bytes (only 4 with the currently defined characters). Other encodings |
113 | include 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 |
115 | and UCS-4 encoding forms. |
ba62762e |
116 | |
376d9008 |
117 | For more information about encodings--for instance, to learn what |
118 | I<surrogates> and I<byte order marks> (BOMs) are--see L<perlunicode>. |
ba62762e |
119 | |
120 | =head2 Perl's Unicode Support |
121 | |
376d9008 |
122 | Starting from Perl 5.6.0, Perl has had the capacity to handle Unicode |
123 | natively. Perl 5.8.0, however, is the first recommended release for |
124 | serious Unicode work. The maintenance release 5.6.1 fixed many of the |
125 | problems of the initial Unicode implementation, but for example |
1bfb14c4 |
126 | regular expressions still do not work with Unicode in 5.6.1. |
ba62762e |
127 | |
128 | B<Starting from Perl 5.8.0, the use of C<use utf8> is no longer |
129 | necessary.> In earlier releases the C<utf8> pragma was used to declare |
130 | that operations in the current block or file would be Unicode-aware. |
376d9008 |
131 | This model was found to be wrong, or at least clumsy: the "Unicodeness" |
1bfb14c4 |
132 | is now carried with the data, instead of being attached to the |
376d9008 |
133 | operations. Only one case remains where an explicit C<use utf8> is |
134 | needed: if your Perl script itself is encoded in UTF-8, you can use |
135 | UTF-8 in your identifier names, and in string and regular expression |
136 | literals, by saying C<use utf8>. This is not the default because |
8f8cf39c |
137 | scripts with legacy 8-bit data in them would break. See L<utf8>. |
ba62762e |
138 | |
139 | =head2 Perl's Unicode Model |
140 | |
376d9008 |
141 | Perl supports both pre-5.6 strings of eight-bit native bytes, and |
142 | strings of Unicode characters. The principle is that Perl tries to |
143 | keep its data as eight-bit bytes for as long as possible, but as soon |
144 | as Unicodeness cannot be avoided, the data is transparently upgraded |
145 | to Unicode. |
ba62762e |
146 | |
4192de81 |
147 | Internally, Perl currently uses either whatever the native eight-bit |
376d9008 |
148 | character set of the platform (for example Latin-1) is, defaulting to |
149 | UTF-8, to encode Unicode strings. Specifically, if all code points in |
150 | the string are C<0xFF> or less, Perl uses the native eight-bit |
151 | character set. Otherwise, it uses UTF-8. |
4192de81 |
152 | |
7ca610e8 |
153 | A user of Perl does not normally need to know nor care how Perl |
20ba30f4 |
154 | happens to encode its internal strings, but it becomes relevant when |
fae2c0fb |
155 | outputting Unicode strings to a stream without a PerlIO layer -- one with |
376d9008 |
156 | the "default" encoding. In such a case, the raw bytes used internally |
7ca610e8 |
157 | (the native character set or UTF-8, as appropriate for each string) |
158 | will be used, and a "Wide character" warning will be issued if those |
159 | strings contain a character beyond 0x00FF. |
4192de81 |
160 | |
161 | For example, |
162 | |
7ca610e8 |
163 | perl -e 'print "\x{DF}\n", "\x{0100}\x{DF}\n"' |
4192de81 |
164 | |
165 | produces a fairly useless mixture of native bytes and UTF-8, as well |
1bfb14c4 |
166 | as a warning: |
4192de81 |
167 | |
1bfb14c4 |
168 | Wide character in print at ... |
169 | |
fae2c0fb |
170 | To output UTF-8, use the C<:utf8> output layer. Prepending |
4192de81 |
171 | |
172 | binmode(STDOUT, ":utf8"); |
173 | |
376d9008 |
174 | to this sample program ensures that the output is completely UTF-8, |
1bfb14c4 |
175 | and removes the program's warning. |
ba62762e |
176 | |
8aa8f774 |
177 | You can enable automatic UTF-8-ification of your standard file |
178 | handles, default C<open()> layer, and C<@ARGV> by using either |
179 | the C<-C> command line switch or the C<PERL_UNICODE> environment |
180 | variable, see L<perlrun> for the documentation of the C<-C> switch. |
181 | |
182 | Note that this means that Perl expects other software to work, too: |
183 | if Perl has been led to believe that STDIN should be UTF-8, but then |
184 | STDIN coming in from another command is not UTF-8, Perl will complain |
185 | about the malformed UTF-8. |
b310b053 |
186 | |
ec71e770 |
187 | All features that combine Unicode and I/O also require using the new |
188 | PerlIO feature. Almost all Perl 5.8 platforms do use PerlIO, though: |
189 | you can see whether yours is by running "perl -V" and looking for |
190 | C<useperlio=define>. |
191 | |
64c66fb6 |
192 | =head2 Unicode and EBCDIC |
193 | |
194 | Perl 5.8.0 also supports Unicode on EBCDIC platforms. There, |
376d9008 |
195 | Unicode support is somewhat more complex to implement since |
64c66fb6 |
196 | additional conversions are needed at every step. Some problems |
dc4af4bb |
197 | remain, see L<perlebcdic> for details. |
64c66fb6 |
198 | |
199 | In any case, the Unicode support on EBCDIC platforms is better than |
200 | in the 5.6 series, which didn't work much at all for EBCDIC platform. |
201 | On EBCDIC platforms, the internal Unicode encoding form is UTF-EBCDIC |
376d9008 |
202 | instead of UTF-8. The difference is that as UTF-8 is "ASCII-safe" in |
203 | that ASCII characters encode to UTF-8 as-is, while UTF-EBCDIC is |
204 | "EBCDIC-safe". |
ba62762e |
205 | |
206 | =head2 Creating Unicode |
207 | |
376d9008 |
208 | To create Unicode characters in literals for code points above C<0xFF>, |
209 | use the C<\x{...}> notation in double-quoted strings: |
ba62762e |
210 | |
211 | my $smiley = "\x{263a}"; |
212 | |
376d9008 |
213 | Similarly, it can be used in regular expression literals |
ba62762e |
214 | |
215 | $smiley =~ /\x{263a}/; |
216 | |
217 | At run-time you can use C<chr()>: |
218 | |
219 | my $hebrew_alef = chr(0x05d0); |
220 | |
376d9008 |
221 | See L</"Further Resources"> for how to find all these numeric codes. |
ba62762e |
222 | |
376d9008 |
223 | Naturally, C<ord()> will do the reverse: it turns a character into |
224 | a code point. |
ba62762e |
225 | |
1bfb14c4 |
226 | Note that C<\x..> (no C<{}> and only two hexadecimal digits), C<\x{...}>, |
227 | and C<chr(...)> for arguments less than C<0x100> (decimal 256) |
228 | generate an eight-bit character for backward compatibility with older |
229 | Perls. For arguments of C<0x100> or more, Unicode characters are |
a5f0baef |
230 | always produced. If you want to force the production of Unicode |
231 | characters regardless of the numeric value, use C<pack("U", ...)> |
232 | instead of C<\x..>, C<\x{...}>, or C<chr()>. |
ba62762e |
233 | |
234 | You can also use the C<charnames> pragma to invoke characters |
376d9008 |
235 | by name in double-quoted strings: |
ba62762e |
236 | |
237 | use charnames ':full'; |
238 | my $arabic_alef = "\N{ARABIC LETTER ALEF}"; |
239 | |
240 | And, as mentioned above, you can also C<pack()> numbers into Unicode |
241 | characters: |
242 | |
243 | my $georgian_an = pack("U", 0x10a0); |
244 | |
8a5e5dd5 |
245 | Note that both C<\x{...}> and C<\N{...}> are compile-time string |
246 | constants: you cannot use variables in them. if you want similar |
247 | run-time functionality, use C<chr()> and C<charnames::vianame()>. |
248 | |
1eda90df |
249 | If you want to force the result to Unicode characters, use the special |
f337b084 |
250 | C<"U0"> prefix. It consumes no arguments but causes the following bytes |
251 | to be interpreted as the UTF-8 encoding of Unicode characters: |
1eda90df |
252 | |
f337b084 |
253 | my $chars = pack("U0W*", 0x80, 0x42); |
771cd3b2 |
254 | |
f337b084 |
255 | Likewise, you can stop such UTF-8 interpretation by using the special |
771cd3b2 |
256 | C<"C0"> prefix. |
1eda90df |
257 | |
ba62762e |
258 | =head2 Handling Unicode |
259 | |
260 | Handling Unicode is for the most part transparent: just use the |
261 | strings as usual. Functions like C<index()>, C<length()>, and |
262 | C<substr()> will work on the Unicode characters; regular expressions |
263 | will work on the Unicode characters (see L<perlunicode> and L<perlretut>). |
264 | |
1bfb14c4 |
265 | Note that Perl considers combining character sequences to be |
c0c50798 |
266 | separate characters, so for example |
ba62762e |
267 | |
268 | use charnames ':full'; |
269 | print length("\N{LATIN CAPITAL LETTER A}\N{COMBINING ACUTE ACCENT}"), "\n"; |
270 | |
271 | will print 2, not 1. The only exception is that regular expressions |
272 | have C<\X> for matching a combining character sequence. |
273 | |
376d9008 |
274 | Life is not quite so transparent, however, when working with legacy |
275 | encodings, I/O, and certain special cases: |
ba62762e |
276 | |
277 | =head2 Legacy Encodings |
278 | |
279 | When you combine legacy data and Unicode the legacy data needs |
280 | to be upgraded to Unicode. Normally ISO 8859-1 (or EBCDIC, if |
2575c402 |
281 | applicable) is assumed. |
ba62762e |
282 | |
376d9008 |
283 | The C<Encode> module knows about many encodings and has interfaces |
ba62762e |
284 | for doing conversions between those encodings: |
285 | |
2c9359a2 |
286 | use Encode 'decode'; |
287 | $data = decode("iso-8859-3", $data); # convert from legacy to utf-8 |
ba62762e |
288 | |
289 | =head2 Unicode I/O |
290 | |
8baee566 |
291 | Normally, writing out Unicode data |
ba62762e |
292 | |
8baee566 |
293 | print FH $some_string_with_unicode, "\n"; |
ba62762e |
294 | |
8baee566 |
295 | produces raw bytes that Perl happens to use to internally encode the |
376d9008 |
296 | Unicode string. Perl's internal encoding depends on the system as |
297 | well as what characters happen to be in the string at the time. If |
298 | any of the characters are at code points C<0x100> or above, you will get |
299 | a warning. To ensure that the output is explicitly rendered in the |
300 | encoding you desire--and to avoid the warning--open the stream with |
301 | the desired encoding. Some examples: |
ba62762e |
302 | |
ec90690f |
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 |
309 | and on already open streams, use C<binmode()>: |
1d7919c5 |
310 | |
311 | binmode(STDOUT, ":utf8"); |
312 | |
ec90690f |
313 | binmode(STDOUT, ":encoding(ucs2)"); |
314 | binmode(STDOUT, ":encoding(UTF-8)"); |
315 | binmode(STDOUT, ":encoding(shift_jis)"); |
316 | |
b5d8778e |
317 | The matching of encoding names is loose: case does not matter, and |
fae2c0fb |
318 | many encodings have several aliases. Note that the C<:utf8> layer |
1bfb14c4 |
319 | must always be specified exactly like that; it is I<not> subject to |
320 | the loose matching of encoding names. |
b5d8778e |
321 | |
376d9008 |
322 | See L<PerlIO> for the C<:utf8> layer, L<PerlIO::encoding> and |
323 | L<Encode::PerlIO> for the C<:encoding()> layer, and |
324 | L<Encode::Supported> for many encodings supported by the C<Encode> |
325 | module. |
ba62762e |
326 | |
a5f0baef |
327 | Reading in a file that you know happens to be encoded in one of the |
1bfb14c4 |
328 | Unicode or legacy encodings does not magically turn the data into |
329 | Unicode in Perl's eyes. To do that, specify the appropriate |
fae2c0fb |
330 | layer when opening files |
ba62762e |
331 | |
332 | open(my $fh,'<:utf8', 'anything'); |
8baee566 |
333 | my $line_of_unicode = <$fh>; |
334 | |
ec90690f |
335 | open(my $fh,'<:encoding(Big5)', 'anything'); |
8baee566 |
336 | my $line_of_unicode = <$fh>; |
ba62762e |
337 | |
fae2c0fb |
338 | The I/O layers can also be specified more flexibly with |
376d9008 |
339 | the C<open> pragma. See L<open>, or look at the following example. |
ba62762e |
340 | |
fae2c0fb |
341 | use open ':utf8'; # input and output default layer will be UTF-8 |
1d7919c5 |
342 | open X, ">file"; |
343 | print X chr(0x100), "\n"; |
ba62762e |
344 | close X; |
1d7919c5 |
345 | open Y, "<file"; |
ba62762e |
346 | printf "%#x\n", ord(<Y>); # this should print 0x100 |
347 | close Y; |
348 | |
fae2c0fb |
349 | With the C<open> pragma you can use the C<:locale> layer |
ba62762e |
350 | |
12f98225 |
351 | BEGIN { $ENV{LC_ALL} = $ENV{LANG} = 'ru_RU.KOI8-R' } |
1ecefa54 |
352 | # the :locale will probe the locale environment variables like LC_ALL |
ba62762e |
353 | use open OUT => ':locale'; # russki parusski |
354 | open(O, ">koi8"); |
355 | print O chr(0x430); # Unicode CYRILLIC SMALL LETTER A = KOI8-R 0xc1 |
356 | close O; |
357 | open(I, "<koi8"); |
358 | printf "%#x\n", ord(<I>), "\n"; # this should print 0xc1 |
359 | close I; |
360 | |
fae2c0fb |
361 | or you can also use the C<':encoding(...)'> layer |
ba62762e |
362 | |
363 | open(my $epic,'<:encoding(iso-8859-7)','iliad.greek'); |
8baee566 |
364 | my $line_of_unicode = <$epic>; |
ba62762e |
365 | |
8baee566 |
366 | These methods install a transparent filter on the I/O stream that |
367 | converts data from the specified encoding when it is read in from the |
a5f0baef |
368 | stream. The result is always Unicode. |
ba62762e |
369 | |
370 | The L<open> pragma affects all the C<open()> calls after the pragma by |
fae2c0fb |
371 | setting default layers. If you want to affect only certain |
372 | streams, use explicit layers directly in the C<open()> call. |
ba62762e |
373 | |
374 | You can switch encodings on an already opened stream by using |
8baee566 |
375 | C<binmode()>; see L<perlfunc/binmode>. |
ba62762e |
376 | |
1ecefa54 |
377 | The C<:locale> does not currently (as of Perl 5.8.0) work with |
378 | C<open()> and C<binmode()>, only with the C<open> pragma. The |
8baee566 |
379 | C<:utf8> and C<:encoding(...)> methods do work with all of C<open()>, |
1ecefa54 |
380 | C<binmode()>, and the C<open> pragma. |
ba62762e |
381 | |
fae2c0fb |
382 | Similarly, you may use these I/O layers on output streams to |
a5f0baef |
383 | automatically convert Unicode to the specified encoding when it is |
384 | written to the stream. For example, the following snippet copies the |
385 | contents of the file "text.jis" (encoded as ISO-2022-JP, aka JIS) to |
386 | the file "text.utf8", encoded as UTF-8: |
ba62762e |
387 | |
c722ef0a |
388 | open(my $nihongo, '<:encoding(iso-2022-jp)', 'text.jis'); |
389 | open(my $unicode, '>:utf8', 'text.utf8'); |
0cf8a8d9 |
390 | while (<$nihongo>) { print $unicode $_ } |
ba62762e |
391 | |
392 | The naming of encodings, both by the C<open()> and by the C<open> |
2575c402 |
393 | pragma allows for flexible names: C<koi8-r> and C<KOI8R> will both be |
394 | understood. |
ba62762e |
395 | |
396 | Common encodings recognized by ISO, MIME, IANA, and various other |
8baee566 |
397 | standardisation organisations are recognised; for a more detailed |
1bfb14c4 |
398 | list see L<Encode::Supported>. |
ba62762e |
399 | |
400 | C<read()> reads characters and returns the number of characters. |
401 | C<seek()> and C<tell()> operate on byte counts, as do C<sysread()> |
402 | and C<sysseek()>. |
403 | |
8baee566 |
404 | Notice that because of the default behaviour of not doing any |
fae2c0fb |
405 | conversion upon input if there is no default layer, |
ba62762e |
406 | it is easy to mistakenly write code that keeps on expanding a file |
1bfb14c4 |
407 | by repeatedly encoding the data: |
ba62762e |
408 | |
409 | # BAD CODE WARNING |
410 | open F, "file"; |
8baee566 |
411 | local $/; ## read in the whole file of 8-bit characters |
ba62762e |
412 | $t = <F>; |
413 | close F; |
414 | open F, ">:utf8", "file"; |
8baee566 |
415 | print F $t; ## convert to UTF-8 on output |
ba62762e |
416 | close F; |
417 | |
418 | If you run this code twice, the contents of the F<file> will be twice |
1d7919c5 |
419 | UTF-8 encoded. A C<use open ':utf8'> would have avoided the bug, or |
420 | explicitly opening also the F<file> for input as UTF-8. |
ba62762e |
421 | |
0c901d84 |
422 | B<NOTE>: the C<:utf8> and C<:encoding> features work only if your |
d0fadae5 |
423 | Perl has been built with the new PerlIO feature (which is the default |
424 | on most systems). |
0c901d84 |
425 | |
1ecefa54 |
426 | =head2 Displaying Unicode As Text |
427 | |
428 | Sometimes you might want to display Perl scalars containing Unicode as |
8baee566 |
429 | simple ASCII (or EBCDIC) text. The following subroutine converts |
1ecefa54 |
430 | its argument so that Unicode characters with code points greater than |
1bfb14c4 |
431 | 255 are displayed as C<\x{...}>, control characters (like C<\n>) are |
432 | displayed as C<\x..>, and the rest of the characters as themselves: |
1ecefa54 |
433 | |
58c274a1 |
434 | sub nice_string { |
435 | join("", |
436 | map { $_ > 255 ? # if wide character... |
8baee566 |
437 | sprintf("\\x{%04X}", $_) : # \x{...} |
58c274a1 |
438 | chr($_) =~ /[[:cntrl:]]/ ? # else if control character ... |
8baee566 |
439 | sprintf("\\x%02X", $_) : # \x.. |
d0551e73 |
440 | quotemeta(chr($_)) # else quoted or as themselves |
f337b084 |
441 | } unpack("W*", $_[0])); # unpack Unicode characters |
58c274a1 |
442 | } |
443 | |
444 | For example, |
445 | |
446 | nice_string("foo\x{100}bar\n") |
447 | |
d0551e73 |
448 | returns the string |
58c274a1 |
449 | |
d0551e73 |
450 | 'foo\x{0100}bar\x0A' |
451 | |
452 | which is ready to be printed. |
1ecefa54 |
453 | |
ba62762e |
454 | =head2 Special Cases |
455 | |
456 | =over 4 |
457 | |
458 | =item * |
459 | |
460 | Bit Complement Operator ~ And vec() |
461 | |
1bfb14c4 |
462 | The bit complement operator C<~> may produce surprising results if |
463 | used on strings containing characters with ordinal values above |
464 | 255. In such a case, the results are consistent with the internal |
465 | encoding of the characters, but not with much else. So don't do |
466 | that. Similarly for C<vec()>: you will be operating on the |
467 | internally-encoded bit patterns of the Unicode characters, not on |
468 | the code point values, which is very probably not what you want. |
ba62762e |
469 | |
470 | =item * |
471 | |
8baee566 |
472 | Peeking At Perl's Internal Encoding |
473 | |
474 | Normal users of Perl should never care how Perl encodes any particular |
a5f0baef |
475 | Unicode string (because the normal ways to get at the contents of a |
376d9008 |
476 | string with Unicode--via input and output--should always be via |
fae2c0fb |
477 | explicitly-defined I/O layers). But if you must, there are two |
a5f0baef |
478 | ways of looking behind the scenes. |
ba62762e |
479 | |
480 | One way of peeking inside the internal encoding of Unicode characters |
f337b084 |
481 | is to use C<unpack("C*", ...> to get the bytes of whatever the string |
482 | encoding happens to be, or C<unpack("U0..", ...)> to get the bytes of the |
483 | UTF-8 encoding: |
ba62762e |
484 | |
8baee566 |
485 | # this prints c4 80 for the UTF-8 bytes 0xc4 0x80 |
f337b084 |
486 | print join(" ", unpack("U0(H2)*", pack("U", 0x100))), "\n"; |
ba62762e |
487 | |
488 | Yet another way would be to use the Devel::Peek module: |
489 | |
490 | perl -MDevel::Peek -e 'Dump(chr(0x100))' |
491 | |
1e54db1a |
492 | That shows the C<UTF8> flag in FLAGS and both the UTF-8 bytes |
376d9008 |
493 | and Unicode characters in C<PV>. See also later in this document |
8800c35a |
494 | the discussion about the C<utf8::is_utf8()> function. |
ba62762e |
495 | |
496 | =back |
497 | |
498 | =head2 Advanced Topics |
499 | |
500 | =over 4 |
501 | |
502 | =item * |
503 | |
504 | String Equivalence |
505 | |
506 | The question of string equivalence turns somewhat complicated |
376d9008 |
507 | in Unicode: what do you mean by "equal"? |
ba62762e |
508 | |
07698885 |
509 | (Is C<LATIN CAPITAL LETTER A WITH ACUTE> equal to |
510 | C<LATIN CAPITAL LETTER A>?) |
ba62762e |
511 | |
a5f0baef |
512 | The short answer is that by default Perl compares equivalence (C<eq>, |
513 | C<ne>) based only on code points of the characters. In the above |
376d9008 |
514 | case, the answer is no (because 0x00C1 != 0x0041). But sometimes, any |
515 | CAPITAL LETTER As should be considered equal, or even As of any case. |
ba62762e |
516 | |
517 | The long answer is that you need to consider character normalization |
376d9008 |
518 | and casing issues: see L<Unicode::Normalize>, Unicode Technical |
ba62762e |
519 | Reports #15 and #21, I<Unicode Normalization Forms> and I<Case |
376d9008 |
520 | Mappings>, http://www.unicode.org/unicode/reports/tr15/ and |
521 | http://www.unicode.org/unicode/reports/tr21/ |
ba62762e |
522 | |
1bfb14c4 |
523 | As of Perl 5.8.0, the "Full" case-folding of I<Case |
524 | Mappings/SpecialCasing> is implemented. |
ba62762e |
525 | |
526 | =item * |
527 | |
528 | String Collation |
529 | |
376d9008 |
530 | People like to see their strings nicely sorted--or as Unicode |
ba62762e |
531 | parlance goes, collated. But again, what do you mean by collate? |
532 | |
07698885 |
533 | (Does C<LATIN CAPITAL LETTER A WITH ACUTE> come before or after |
534 | C<LATIN CAPITAL LETTER A WITH GRAVE>?) |
ba62762e |
535 | |
58c274a1 |
536 | The short answer is that by default, Perl compares strings (C<lt>, |
ba62762e |
537 | C<le>, C<cmp>, C<ge>, C<gt>) based only on the code points of the |
1bfb14c4 |
538 | characters. In the above case, the answer is "after", since |
da76a1f4 |
539 | C<0x00C1> > C<0x00C0>. |
ba62762e |
540 | |
541 | The long answer is that "it depends", and a good answer cannot be |
542 | given without knowing (at the very least) the language context. |
543 | See L<Unicode::Collate>, and I<Unicode Collation Algorithm> |
544 | http://www.unicode.org/unicode/reports/tr10/ |
545 | |
546 | =back |
547 | |
548 | =head2 Miscellaneous |
549 | |
550 | =over 4 |
551 | |
552 | =item * |
553 | |
3ff56b75 |
554 | Character Ranges and Classes |
ba62762e |
555 | |
556 | Character ranges in regular expression character classes (C</[a-z]/>) |
557 | and in the C<tr///> (also known as C<y///>) operator are not magically |
58c274a1 |
558 | Unicode-aware. What this means that C<[A-Za-z]> will not magically start |
376d9008 |
559 | to mean "all alphabetic letters"; not that it does mean that even for |
560 | 8-bit characters, you should be using C</[[:alpha:]]/> in that case. |
ba62762e |
561 | |
1bfb14c4 |
562 | For specifying character classes like that in regular expressions, |
563 | you can use the various Unicode properties--C<\pL>, or perhaps |
564 | C<\p{Alphabetic}>, in this particular case. You can use Unicode |
565 | code points as the end points of character ranges, but there is no |
566 | magic associated with specifying a certain range. For further |
567 | information--there are dozens of Unicode character classes--see |
568 | L<perlunicode>. |
ba62762e |
569 | |
570 | =item * |
571 | |
572 | String-To-Number Conversions |
573 | |
376d9008 |
574 | Unicode does define several other decimal--and numeric--characters |
575 | besides the familiar 0 to 9, such as the Arabic and Indic digits. |
ba62762e |
576 | Perl does not support string-to-number conversion for digits other |
58c274a1 |
577 | than ASCII 0 to 9 (and ASCII a to f for hexadecimal). |
ba62762e |
578 | |
579 | =back |
580 | |
581 | =head2 Questions With Answers |
582 | |
583 | =over 4 |
584 | |
818c4caa |
585 | =item * |
5cb3728c |
586 | |
587 | Will My Old Scripts Break? |
ba62762e |
588 | |
589 | Very probably not. Unless you are generating Unicode characters |
1bfb14c4 |
590 | somehow, old behaviour should be preserved. About the only behaviour |
591 | that has changed and which could start generating Unicode is the old |
592 | behaviour of C<chr()> where supplying an argument more than 255 |
593 | produced a character modulo 255. C<chr(300)>, for example, was equal |
594 | to C<chr(45)> or "-" (in ASCII), now it is LATIN CAPITAL LETTER I WITH |
595 | BREVE. |
ba62762e |
596 | |
818c4caa |
597 | =item * |
5cb3728c |
598 | |
599 | How Do I Make My Scripts Work With Unicode? |
ba62762e |
600 | |
601 | Very little work should be needed since nothing changes until you |
1bfb14c4 |
602 | generate Unicode data. The most important thing is getting input as |
603 | Unicode; for that, see the earlier I/O discussion. |
ba62762e |
604 | |
818c4caa |
605 | =item * |
5cb3728c |
606 | |
607 | How Do I Know Whether My String Is In Unicode? |
ba62762e |
608 | |
1bfb14c4 |
609 | You shouldn't care. No, you really shouldn't. No, really. If you |
610 | have to care--beyond the cases described above--it means that we |
ba62762e |
611 | didn't get the transparency of Unicode quite right. |
612 | |
613 | Okay, if you insist: |
614 | |
8800c35a |
615 | print utf8::is_utf8($string) ? 1 : 0, "\n"; |
ba62762e |
616 | |
617 | But note that this doesn't mean that any of the characters in the |
618 | string are necessary UTF-8 encoded, or that any of the characters have |
619 | code points greater than 0xFF (255) or even 0x80 (128), or that the |
620 | string has any characters at all. All the C<is_utf8()> does is to |
621 | return the value of the internal "utf8ness" flag attached to the |
376d9008 |
622 | C<$string>. If the flag is off, the bytes in the scalar are interpreted |
3c1c8017 |
623 | as a single byte encoding. If the flag is on, the bytes in the scalar |
376d9008 |
624 | are interpreted as the (multi-byte, variable-length) UTF-8 encoded code |
3c1c8017 |
625 | points of the characters. Bytes added to an UTF-8 encoded string are |
1e54db1a |
626 | automatically upgraded to UTF-8. If mixed non-UTF-8 and UTF-8 scalars |
376d9008 |
627 | are merged (double-quoted interpolation, explicit concatenation, and |
3c1c8017 |
628 | printf/sprintf parameter substitution), the result will be UTF-8 encoded |
629 | as if copies of the byte strings were upgraded to UTF-8: for example, |
630 | |
631 | $a = "ab\x80c"; |
632 | $b = "\x{100}"; |
633 | print "$a = $b\n"; |
634 | |
a02b5feb |
635 | the output string will be UTF-8-encoded C<ab\x80c = \x{100}\n>, but |
636 | C<$a> will stay byte-encoded. |
ba62762e |
637 | |
638 | Sometimes you might really need to know the byte length of a string |
ce7675db |
639 | instead of the character length. For that use either the |
640 | C<Encode::encode_utf8()> function or the C<bytes> pragma and its only |
641 | defined function C<length()>: |
ba62762e |
642 | |
643 | my $unicode = chr(0x100); |
644 | print length($unicode), "\n"; # will print 1 |
ce7675db |
645 | require Encode; |
646 | print length(Encode::encode_utf8($unicode)), "\n"; # will print 2 |
ba62762e |
647 | use bytes; |
1bfb14c4 |
648 | print length($unicode), "\n"; # will also print 2 |
649 | # (the 0xC4 0x80 of the UTF-8) |
ba62762e |
650 | |
818c4caa |
651 | =item * |
5cb3728c |
652 | |
653 | How Do I Detect Data That's Not Valid In a Particular Encoding? |
ba62762e |
654 | |
8baee566 |
655 | Use the C<Encode> package to try converting it. |
656 | For example, |
ba62762e |
657 | |
bb2f379c |
658 | use Encode 'decode_utf8'; |
a365f2ce |
659 | eval { decode_utf8($string, Encode::FB_CROAK) }; |
660 | if ($@) { |
661 | # $string is valid utf8 |
ba62762e |
662 | } else { |
a365f2ce |
663 | # $string is not valid utf8 |
ba62762e |
664 | } |
665 | |
f337b084 |
666 | Or use C<unpack> to try decoding it: |
ba62762e |
667 | |
668 | use warnings; |
f337b084 |
669 | @chars = unpack("C0U*", $string_of_bytes_that_I_think_is_utf8); |
ba62762e |
670 | |
a365f2ce |
671 | If invalid, a C<Malformed UTF-8 character> warning is produced. The "C0" means |
672 | "process the string character per character". Without that, the |
f337b084 |
673 | C<unpack("U*", ...)> would work in C<U0> mode (the default if the format |
674 | string starts with C<U>) and it would return the bytes making up the UTF-8 |
675 | encoding of the target string, something that will always work. |
ba62762e |
676 | |
818c4caa |
677 | =item * |
5cb3728c |
678 | |
679 | How Do I Convert Binary Data Into a Particular Encoding, Or Vice Versa? |
ba62762e |
680 | |
8baee566 |
681 | This probably isn't as useful as you might think. |
682 | Normally, you shouldn't need to. |
ba62762e |
683 | |
1bfb14c4 |
684 | In one sense, what you are asking doesn't make much sense: encodings |
376d9008 |
685 | are for characters, and binary data are not "characters", so converting |
a5f0baef |
686 | "data" into some encoding isn't meaningful unless you know in what |
687 | character set and encoding the binary data is in, in which case it's |
376d9008 |
688 | not just binary data, now is it? |
8baee566 |
689 | |
1bfb14c4 |
690 | If you have a raw sequence of bytes that you know should be |
691 | interpreted via a particular encoding, you can use C<Encode>: |
ba62762e |
692 | |
693 | use Encode 'from_to'; |
694 | from_to($data, "iso-8859-1", "utf-8"); # from latin-1 to utf-8 |
695 | |
1bfb14c4 |
696 | The call to C<from_to()> changes the bytes in C<$data>, but nothing |
697 | material about the nature of the string has changed as far as Perl is |
698 | concerned. Both before and after the call, the string C<$data> |
699 | contains just a bunch of 8-bit bytes. As far as Perl is concerned, |
700 | the encoding of the string remains as "system-native 8-bit bytes". |
8baee566 |
701 | |
702 | You might relate this to a fictional 'Translate' module: |
703 | |
704 | use Translate; |
705 | my $phrase = "Yes"; |
706 | Translate::from_to($phrase, 'english', 'deutsch'); |
707 | ## phrase now contains "Ja" |
ba62762e |
708 | |
8baee566 |
709 | The contents of the string changes, but not the nature of the string. |
1bfb14c4 |
710 | Perl doesn't know any more after the call than before that the |
711 | contents of the string indicates the affirmative. |
ba62762e |
712 | |
376d9008 |
713 | Back to converting data. If you have (or want) data in your system's |
a5f0baef |
714 | native 8-bit encoding (e.g. Latin-1, EBCDIC, etc.), you can use |
715 | pack/unpack to convert to/from Unicode. |
ba62762e |
716 | |
f337b084 |
717 | $native_string = pack("W*", unpack("U*", $Unicode_string)); |
718 | $Unicode_string = pack("U*", unpack("W*", $native_string)); |
ba62762e |
719 | |
720 | If you have a sequence of bytes you B<know> is valid UTF-8, |
721 | but Perl doesn't know it yet, you can make Perl a believer, too: |
722 | |
723 | use Encode 'decode_utf8'; |
8baee566 |
724 | $Unicode = decode_utf8($bytes); |
ba62762e |
725 | |
f337b084 |
726 | or: |
727 | |
728 | $Unicode = pack("U0a*", $bytes); |
729 | |
ba62762e |
730 | You can convert well-formed UTF-8 to a sequence of bytes, but if |
731 | you just want to convert random binary data into UTF-8, you can't. |
1bfb14c4 |
732 | B<Any random collection of bytes isn't well-formed UTF-8>. You can |
ba62762e |
733 | use C<unpack("C*", $string)> for the former, and you can create |
8baee566 |
734 | well-formed Unicode data by C<pack("U*", 0xff, ...)>. |
ba62762e |
735 | |
818c4caa |
736 | =item * |
5cb3728c |
737 | |
738 | How Do I Display Unicode? How Do I Input Unicode? |
ba62762e |
739 | |
076d825e |
740 | See http://www.alanwood.net/unicode/ and |
ba62762e |
741 | http://www.cl.cam.ac.uk/~mgk25/unicode.html |
742 | |
818c4caa |
743 | =item * |
5cb3728c |
744 | |
745 | How Does Unicode Work With Traditional Locales? |
ba62762e |
746 | |
747 | In Perl, not very well. Avoid using locales through the C<locale> |
4c496f0c |
748 | pragma. Use only one or the other. But see L<perlrun> for the |
749 | description of the C<-C> switch and its environment counterpart, |
750 | C<$ENV{PERL_UNICODE}> to see how to enable various Unicode features, |
751 | for example by using locale settings. |
ba62762e |
752 | |
753 | =back |
754 | |
755 | =head2 Hexadecimal Notation |
756 | |
376d9008 |
757 | The Unicode standard prefers using hexadecimal notation because |
758 | that more clearly shows the division of Unicode into blocks of 256 characters. |
ba62762e |
759 | Hexadecimal is also simply shorter than decimal. You can use decimal |
760 | notation, too, but learning to use hexadecimal just makes life easier |
1bfb14c4 |
761 | with the Unicode standard. The C<U+HHHH> notation uses hexadecimal, |
076d825e |
762 | for example. |
ba62762e |
763 | |
764 | The C<0x> prefix means a hexadecimal number, the digits are 0-9 I<and> |
765 | a-f (or A-F, case doesn't matter). Each hexadecimal digit represents |
766 | four bits, or half a byte. C<print 0x..., "\n"> will show a |
767 | hexadecimal number in decimal, and C<printf "%x\n", $decimal> will |
768 | show a decimal number in hexadecimal. If you have just the |
376d9008 |
769 | "hex digits" of a hexadecimal number, you can use the C<hex()> function. |
ba62762e |
770 | |
771 | print 0x0009, "\n"; # 9 |
772 | print 0x000a, "\n"; # 10 |
773 | print 0x000f, "\n"; # 15 |
774 | print 0x0010, "\n"; # 16 |
775 | print 0x0011, "\n"; # 17 |
776 | print 0x0100, "\n"; # 256 |
777 | |
778 | print 0x0041, "\n"; # 65 |
779 | |
780 | printf "%x\n", 65; # 41 |
781 | printf "%#x\n", 65; # 0x41 |
782 | |
783 | print hex("41"), "\n"; # 65 |
784 | |
785 | =head2 Further Resources |
786 | |
787 | =over 4 |
788 | |
789 | =item * |
790 | |
791 | Unicode Consortium |
792 | |
793 | http://www.unicode.org/ |
794 | |
795 | =item * |
796 | |
797 | Unicode FAQ |
798 | |
799 | http://www.unicode.org/unicode/faq/ |
800 | |
801 | =item * |
802 | |
803 | Unicode Glossary |
804 | |
805 | http://www.unicode.org/glossary/ |
806 | |
807 | =item * |
808 | |
809 | Unicode Useful Resources |
810 | |
811 | http://www.unicode.org/unicode/onlinedat/resources.html |
812 | |
813 | =item * |
814 | |
815 | Unicode and Multilingual Support in HTML, Fonts, Web Browsers and Other Applications |
816 | |
076d825e |
817 | http://www.alanwood.net/unicode/ |
ba62762e |
818 | |
819 | =item * |
820 | |
821 | UTF-8 and Unicode FAQ for Unix/Linux |
822 | |
823 | http://www.cl.cam.ac.uk/~mgk25/unicode.html |
824 | |
825 | =item * |
826 | |
827 | Legacy Character Sets |
828 | |
829 | http://www.czyborra.com/ |
830 | http://www.eki.ee/letter/ |
831 | |
832 | =item * |
833 | |
834 | The Unicode support files live within the Perl installation in the |
835 | directory |
836 | |
837 | $Config{installprivlib}/unicore |
838 | |
839 | in Perl 5.8.0 or newer, and |
840 | |
841 | $Config{installprivlib}/unicode |
842 | |
843 | in the Perl 5.6 series. (The renaming to F<lib/unicore> was done to |
844 | avoid naming conflicts with lib/Unicode in case-insensitive filesystems.) |
551b6b6f |
845 | The main Unicode data file is F<UnicodeData.txt> (or F<Unicode.301> in |
ba62762e |
846 | Perl 5.6.1.) You can find the C<$Config{installprivlib}> by |
847 | |
848 | perl "-V:installprivlib" |
849 | |
ba62762e |
850 | You can explore various information from the Unicode data files using |
851 | the C<Unicode::UCD> module. |
852 | |
853 | =back |
854 | |
f6edf83b |
855 | =head1 UNICODE IN OLDER PERLS |
856 | |
857 | If you cannot upgrade your Perl to 5.8.0 or later, you can still |
858 | do some Unicode processing by using the modules C<Unicode::String>, |
859 | C<Unicode::Map8>, and C<Unicode::Map>, available from CPAN. |
860 | If you have the GNU recode installed, you can also use the |
376d9008 |
861 | Perl front-end C<Convert::Recode> for character conversions. |
f6edf83b |
862 | |
aaef10c5 |
863 | The following are fast conversions from ISO 8859-1 (Latin-1) bytes |
63de3cb2 |
864 | to UTF-8 bytes and back, the code works even with older Perl 5 versions. |
aaef10c5 |
865 | |
866 | # ISO 8859-1 to UTF-8 |
867 | s/([\x80-\xFF])/chr(0xC0|ord($1)>>6).chr(0x80|ord($1)&0x3F)/eg; |
868 | |
869 | # UTF-8 to ISO 8859-1 |
870 | s/([\xC2\xC3])([\x80-\xBF])/chr(ord($1)<<6&0xC0|ord($2)&0x3F)/eg; |
871 | |
ba62762e |
872 | =head1 SEE ALSO |
873 | |
2575c402 |
874 | L<perlunitut>, L<perlunicode>, L<Encode>, L<open>, L<utf8>, L<bytes>, |
4c496f0c |
875 | L<perlretut>, L<perlrun>, L<Unicode::Collate>, L<Unicode::Normalize>, |
876 | L<Unicode::UCD> |
ba62762e |
877 | |
376d9008 |
878 | =head1 ACKNOWLEDGMENTS |
ba62762e |
879 | |
880 | Thanks to the kind readers of the perl5-porters@perl.org, |
881 | perl-unicode@perl.org, linux-utf8@nl.linux.org, and unicore@unicode.org |
882 | mailing lists for their valuable feedback. |
883 | |
884 | =head1 AUTHOR, COPYRIGHT, AND LICENSE |
885 | |
0f2f9b7d |
886 | Copyright 2001-2002 Jarkko Hietaniemi E<lt>jhi@iki.fiE<gt> |
ba62762e |
887 | |
888 | This document may be distributed under the same terms as Perl itself. |