Commit | Line | Data |
---|---|---|
2575c402 JW |
1 | =head1 NAME |
2 | ||
3 | perlunifaq - Perl Unicode FAQ | |
4 | ||
740d4bb2 | 5 | =head1 Q and A |
2575c402 JW |
6 | |
7 | This is a list of questions and answers about Unicode in Perl, intended to be | |
8 | read after L<perlunitut>. | |
9 | ||
10 | =head2 perlunitut isn't really a Unicode tutorial, is it? | |
11 | ||
12 | No, and this isn't really a Unicode FAQ. | |
13 | ||
e1b711da | 14 | Perl has an abstracted interface for all supported character encodings, so this |
2575c402 JW |
15 | is actually a generic C<Encode> tutorial and C<Encode> FAQ. But many people |
16 | think that Unicode is special and magical, and I didn't want to disappoint | |
17 | them, so I decided to call the document a Unicode tutorial. | |
18 | ||
740d4bb2 JW |
19 | =head2 What character encodings does Perl support? |
20 | ||
21 | To find out which character encodings your Perl supports, run: | |
22 | ||
23 | perl -MEncode -le "print for Encode->encodings(':all')" | |
24 | ||
25 | =head2 Which version of perl should I use? | |
26 | ||
27 | Well, if you can, upgrade to the most recent, but certainly C<5.8.1> or newer. | |
2a6886e1 | 28 | The tutorial and FAQ assume the latest release. |
740d4bb2 JW |
29 | |
30 | You should also check your modules, and upgrade them if necessary. For example, | |
31 | HTML::Entities requires version >= 1.32 to function correctly, even though the | |
32 | changelog is silent about this. | |
33 | ||
2575c402 JW |
34 | =head2 What about binary data, like images? |
35 | ||
36 | Well, apart from a bare C<binmode $fh>, you shouldn't treat them specially. | |
37 | (The binmode is needed because otherwise Perl may convert line endings on Win32 | |
38 | systems.) | |
39 | ||
40 | Be careful, though, to never combine text strings with binary strings. If you | |
41 | need text in a binary stream, encode your text strings first using the | |
42 | appropriate encoding, then join them with binary strings. See also: "What if I | |
43 | don't encode?". | |
44 | ||
2575c402 JW |
45 | =head2 When should I decode or encode? |
46 | ||
740d4bb2 | 47 | Whenever you're communicating text with anything that is external to your perl |
2575c402 JW |
48 | process, like a database, a text file, a socket, or another program. Even if |
49 | the thing you're communicating with is also written in Perl. | |
50 | ||
51 | =head2 What if I don't decode? | |
52 | ||
53 | Whenever your encoded, binary string is used together with a text string, Perl | |
54 | will assume that your binary string was encoded with ISO-8859-1, also known as | |
55 | latin-1. If it wasn't latin-1, then your data is unpleasantly converted. For | |
56 | example, if it was UTF-8, the individual bytes of multibyte characters are seen | |
57 | as separate characters, and then again converted to UTF-8. Such double encoding | |
58 | can be compared to double HTML encoding (C<&gt;>), or double URI encoding | |
59 | (C<%253E>). | |
60 | ||
61 | This silent implicit decoding is known as "upgrading". That may sound | |
62 | positive, but it's best to avoid it. | |
63 | ||
64 | =head2 What if I don't encode? | |
65 | ||
66 | Your text string will be sent using the bytes in Perl's internal format. In | |
67 | some cases, Perl will warn you that you're doing something wrong, with a | |
68 | friendly warning: | |
69 | ||
70 | Wide character in print at example.pl line 2. | |
71 | ||
72 | Because the internal format is often UTF-8, these bugs are hard to spot, | |
73 | because UTF-8 is usually the encoding you wanted! But don't be lazy, and don't | |
74 | use the fact that Perl's internal format is UTF-8 to your advantage. Encode | |
75 | explicitly to avoid weird bugs, and to show to maintenance programmers that you | |
76 | thought this through. | |
77 | ||
78 | =head2 Is there a way to automatically decode or encode? | |
79 | ||
80 | If all data that comes from a certain handle is encoded in exactly the same | |
81 | way, you can tell the PerlIO system to automatically decode everything, with | |
82 | the C<encoding> layer. If you do this, you can't accidentally forget to decode | |
83 | or encode anymore, on things that use the layered handle. | |
84 | ||
85 | You can provide this layer when C<open>ing the file: | |
86 | ||
9e5bbba0 KW |
87 | open my $fh, '>:encoding(UTF-8)', $filename; # auto encoding on write |
88 | open my $fh, '<:encoding(UTF-8)', $filename; # auto decoding on read | |
2575c402 JW |
89 | |
90 | Or if you already have an open filehandle: | |
91 | ||
9e5bbba0 | 92 | binmode $fh, ':encoding(UTF-8)'; |
2575c402 JW |
93 | |
94 | Some database drivers for DBI can also automatically encode and decode, but | |
740d4bb2 | 95 | that is sometimes limited to the UTF-8 encoding. |
2575c402 JW |
96 | |
97 | =head2 What if I don't know which encoding was used? | |
98 | ||
99 | Do whatever you can to find out, and if you have to: guess. (Don't forget to | |
100 | document your guess with a comment.) | |
101 | ||
102 | You could open the document in a web browser, and change the character set or | |
103 | character encoding until you can visually confirm that all characters look the | |
104 | way they should. | |
105 | ||
106 | There is no way to reliably detect the encoding automatically, so if people | |
107 | keep sending you data without charset indication, you may have to educate them. | |
108 | ||
109 | =head2 Can I use Unicode in my Perl sources? | |
110 | ||
111 | Yes, you can! If your sources are UTF-8 encoded, you can indicate that with the | |
112 | C<use utf8> pragma. | |
113 | ||
114 | use utf8; | |
115 | ||
116 | This doesn't do anything to your input, or to your output. It only influences | |
117 | the way your sources are read. You can use Unicode in string literals, in | |
118 | identifiers (but they still have to be "word characters" according to C<\w>), | |
119 | and even in custom delimiters. | |
120 | ||
121 | =head2 Data::Dumper doesn't restore the UTF8 flag; is it broken? | |
122 | ||
123 | No, Data::Dumper's Unicode abilities are as they should be. There have been | |
124 | some complaints that it should restore the UTF8 flag when the data is read | |
125 | again with C<eval>. However, you should really not look at the flag, and | |
126 | nothing indicates that Data::Dumper should break this rule. | |
127 | ||
128 | Here's what happens: when Perl reads in a string literal, it sticks to 8 bit | |
129 | encoding as long as it can. (But perhaps originally it was internally encoded | |
130 | as UTF-8, when you dumped it.) When it has to give that up because other | |
131 | characters are added to the text string, it silently upgrades the string to | |
132 | UTF-8. | |
133 | ||
134 | If you properly encode your strings for output, none of this is of your | |
135 | concern, and you can just C<eval> dumped data as always. | |
136 | ||
740d4bb2 JW |
137 | =head2 Why do regex character classes sometimes match only in the ASCII range? |
138 | ||
20db7501 KW |
139 | Starting in Perl 5.14 (and partially in Perl 5.12), just put a |
140 | C<use feature 'unicode_strings'> near the beginning of your program. | |
141 | Within its lexical scope you shouldn't have this problem. It also is | |
9c98089c BF |
142 | automatically enabled under C<use feature ':5.12'> or C<use v5.12> or |
143 | using C<-E> on the command line for Perl 5.12 or higher. | |
20db7501 KW |
144 | |
145 | The rationale for requiring this is to not break older programs that | |
146 | rely on the way things worked before Unicode came along. Those older | |
147 | programs knew only about the ASCII character set, and so may not work | |
148 | properly for additional characters. When a string is encoded in UTF-8, | |
149 | Perl assumes that the program is prepared to deal with Unicode, but when | |
4b9734bf KW |
150 | the string isn't, Perl assumes that only ASCII |
151 | is wanted, and so those characters that are not ASCII | |
20db7501 KW |
152 | characters aren't recognized as to what they would be in Unicode. |
153 | C<use feature 'unicode_strings'> tells Perl to treat all characters as | |
154 | Unicode, whether the string is encoded in UTF-8 or not, thus avoiding | |
155 | the problem. | |
156 | ||
157 | However, on earlier Perls, or if you pass strings to subroutines outside | |
850b7ec9 | 158 | the feature's scope, you can force Unicode rules by changing the |
20db7501 | 159 | encoding to UTF-8 by doing C<utf8::upgrade($string)>. This can be used |
2bbc8d55 SP |
160 | safely on any string, as it checks and does not change strings that have |
161 | already been upgraded. | |
740d4bb2 JW |
162 | |
163 | For a more detailed discussion, see L<Unicode::Semantics> on CPAN. | |
164 | ||
3d6c5fec DG |
165 | =head2 Why do some characters not uppercase or lowercase correctly? |
166 | ||
167 | See the answer to the previous question. | |
168 | ||
2575c402 JW |
169 | =head2 How can I determine if a string is a text string or a binary string? |
170 | ||
171 | You can't. Some use the UTF8 flag for this, but that's misuse, and makes well | |
172 | behaved modules like Data::Dumper look bad. The flag is useless for this | |
173 | purpose, because it's off when an 8 bit encoding (by default ISO-8859-1) is | |
174 | used to store the string. | |
175 | ||
176 | This is something you, the programmer, has to keep track of; sorry. You could | |
177 | consider adopting a kind of "Hungarian notation" to help with this. | |
178 | ||
179 | =head2 How do I convert from encoding FOO to encoding BAR? | |
180 | ||
181 | By first converting the FOO-encoded byte string to a text string, and then the | |
182 | text string to a BAR-encoded byte string: | |
183 | ||
184 | my $text_string = decode('FOO', $foo_string); | |
185 | my $bar_string = encode('BAR', $text_string); | |
186 | ||
187 | or by skipping the text string part, and going directly from one binary | |
188 | encoding to the other: | |
189 | ||
190 | use Encode qw(from_to); | |
191 | from_to($string, 'FOO', 'BAR'); # changes contents of $string | |
192 | ||
193 | or by letting automatic decoding and encoding do all the work: | |
194 | ||
195 | open my $foofh, '<:encoding(FOO)', 'example.foo.txt'; | |
196 | open my $barfh, '>:encoding(BAR)', 'example.bar.txt'; | |
197 | print { $barfh } $_ while <$foofh>; | |
198 | ||
740d4bb2 JW |
199 | =head2 What are C<decode_utf8> and C<encode_utf8>? |
200 | ||
201 | These are alternate syntaxes for C<decode('utf8', ...)> and C<encode('utf8', | |
202 | ...)>. | |
203 | ||
204 | =head2 What is a "wide character"? | |
205 | ||
206 | This is a term used both for characters with an ordinal value greater than 127, | |
207 | characters with an ordinal value greater than 255, or any character occupying | |
e1b711da | 208 | more than one byte, depending on the context. |
740d4bb2 JW |
209 | |
210 | The Perl warning "Wide character in ..." is caused by a character with an | |
211 | ordinal value greater than 255. With no specified encoding layer, Perl tries to | |
212 | fit things in ISO-8859-1 for backward compatibility reasons. When it can't, it | |
213 | emits this warning (if warnings are enabled), and outputs UTF-8 encoded data | |
214 | instead. | |
215 | ||
216 | To avoid this warning and to avoid having different output encodings in a single | |
217 | stream, always specify an encoding explicitly, for example with a PerlIO layer: | |
218 | ||
219 | binmode STDOUT, ":encoding(UTF-8)"; | |
220 | ||
221 | =head1 INTERNALS | |
222 | ||
223 | =head2 What is "the UTF8 flag"? | |
224 | ||
225 | Please, unless you're hacking the internals, or debugging weirdness, don't | |
226 | think about the UTF8 flag at all. That means that you very probably shouldn't | |
227 | use C<is_utf8>, C<_utf8_on> or C<_utf8_off> at all. | |
228 | ||
229 | The UTF8 flag, also called SvUTF8, is an internal flag that indicates that the | |
230 | current internal representation is UTF-8. Without the flag, it is assumed to be | |
2a6886e1 KW |
231 | ISO-8859-1. Perl converts between these automatically. (Actually Perl usually |
232 | assumes the representation is ASCII; see L</Why do regex character classes | |
233 | sometimes match only in the ASCII range?> above.) | |
740d4bb2 JW |
234 | |
235 | One of Perl's internal formats happens to be UTF-8. Unfortunately, Perl can't | |
236 | keep a secret, so everyone knows about this. That is the source of much | |
237 | confusion. It's better to pretend that the internal format is some unknown | |
238 | encoding, and that you always have to encode and decode explicitly. | |
239 | ||
2575c402 JW |
240 | =head2 What about the C<use bytes> pragma? |
241 | ||
242 | Don't use it. It makes no sense to deal with bytes in a text string, and it | |
243 | makes no sense to deal with characters in a byte string. Do the proper | |
244 | conversions (by decoding/encoding), and things will work out well: you get | |
245 | character counts for decoded data, and byte counts for encoded data. | |
246 | ||
247 | C<use bytes> is usually a failed attempt to do something useful. Just forget | |
248 | about it. | |
249 | ||
740d4bb2 | 250 | =head2 What about the C<use encoding> pragma? |
2575c402 | 251 | |
740d4bb2 JW |
252 | Don't use it. Unfortunately, it assumes that the programmer's environment and |
253 | that of the user will use the same encoding. It will use the same encoding for | |
254 | the source code and for STDIN and STDOUT. When a program is copied to another | |
255 | machine, the source code does not change, but the STDIO environment might. | |
256 | ||
257 | If you need non-ASCII characters in your source code, make it a UTF-8 encoded | |
258 | file and C<use utf8>. | |
259 | ||
260 | If you need to set the encoding for STDIN, STDOUT, and STDERR, for example | |
261 | based on the user's locale, C<use open>. | |
262 | ||
263 | =head2 What is the difference between C<:encoding> and C<:utf8>? | |
264 | ||
265 | Because UTF-8 is one of Perl's internal formats, you can often just skip the | |
266 | encoding or decoding step, and manipulate the UTF8 flag directly. | |
267 | ||
268 | Instead of C<:encoding(UTF-8)>, you can simply use C<:utf8>, which skips the | |
269 | encoding step if the data was already represented as UTF8 internally. This is | |
270 | widely accepted as good behavior when you're writing, but it can be dangerous | |
271 | when reading, because it causes internal inconsistency when you have invalid | |
272 | byte sequences. Using C<:utf8> for input can sometimes result in security | |
273 | breaches, so please use C<:encoding(UTF-8)> instead. | |
274 | ||
275 | Instead of C<decode> and C<encode>, you could use C<_utf8_on> and C<_utf8_off>, | |
276 | but this is considered bad style. Especially C<_utf8_on> can be dangerous, for | |
277 | the same reason that C<:utf8> can. | |
278 | ||
8da107ac KW |
279 | There are some shortcuts for oneliners; |
280 | see L<-C|perlrun/-C [numberE<sol>list]> in L<perlrun>. | |
2575c402 JW |
281 | |
282 | =head2 What's the difference between C<UTF-8> and C<utf8>? | |
283 | ||
284 | C<UTF-8> is the official standard. C<utf8> is Perl's way of being liberal in | |
285 | what it accepts. If you have to communicate with things that aren't so liberal, | |
286 | you may want to consider using C<UTF-8>. If you have to communicate with things | |
287 | that are too liberal, you may have to use C<utf8>. The full explanation is in | |
288 | L<Encode>. | |
289 | ||
290 | C<UTF-8> is internally known as C<utf-8-strict>. The tutorial uses UTF-8 | |
291 | consistently, even where utf8 is actually used internally, because the | |
292 | distinction can be hard to make, and is mostly irrelevant. | |
293 | ||
294 | For example, utf8 can be used for code points that don't exist in Unicode, like | |
295 | 9999999, but if you encode that to UTF-8, you get a substitution character (by | |
296 | default; see L<Encode/"Handling Malformed Data"> for more ways of dealing with | |
297 | this.) | |
298 | ||
299 | Okay, if you insist: the "internal format" is utf8, not UTF-8. (When it's not | |
300 | some other encoding.) | |
301 | ||
302 | =head2 I lost track; what encoding is the internal format really? | |
303 | ||
304 | It's good that you lost track, because you shouldn't depend on the internal | |
305 | format being any specific encoding. But since you asked: by default, the | |
306 | internal format is either ISO-8859-1 (latin-1), or utf8, depending on the | |
307 | history of the string. On EBCDIC platforms, this may be different even. | |
308 | ||
309 | Perl knows how it stored the string internally, and will use that knowledge | |
310 | when you C<encode>. In other words: don't try to find out what the internal | |
311 | encoding for a certain string is, but instead just encode it into the encoding | |
312 | that you want. | |
313 | ||
2575c402 JW |
314 | =head1 AUTHOR |
315 | ||
740d4bb2 | 316 | Juerd Waalboer <#####@juerd.nl> |
2575c402 JW |
317 | |
318 | =head1 SEE ALSO | |
319 | ||
320 | L<perlunicode>, L<perluniintro>, L<Encode> | |
321 |