Commit | Line | Data |
3ef515df |
1 | package encoding; |
0ab8f81e |
2 | our $VERSION = do { my @r = (q$Revision: 1.33 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r }; |
3ef515df |
3 | |
4 | use Encode; |
046f36bf |
5 | use strict; |
3ef515df |
6 | |
7 | BEGIN { |
8 | if (ord("A") == 193) { |
9 | require Carp; |
10 | Carp::croak "encoding pragma does not support EBCDIC platforms"; |
11 | } |
12 | } |
13 | |
0ab8f81e |
14 | our $HAS_PERLIO = 0; |
15 | eval { require PerlIO::encoding }; |
16 | unless ($@){ |
17 | $HAS_PERLIO = (PerlIO::encoding->VERSION >= 0.02); |
18 | } |
b2704119 |
19 | |
3ef515df |
20 | sub import { |
21 | my $class = shift; |
22 | my $name = shift; |
23 | my %arg = @_; |
24 | $name ||= $ENV{PERL_ENCODING}; |
25 | |
26 | my $enc = find_encoding($name); |
27 | unless (defined $enc) { |
28 | require Carp; |
29 | Carp::croak "Unknown encoding '$name'"; |
30 | } |
aae85ceb |
31 | unless ($arg{Filter}){ |
32 | ${^ENCODING} = $enc; # this is all you need, actually. |
85982a32 |
33 | $HAS_PERLIO or return 1; |
aae85ceb |
34 | for my $h (qw(STDIN STDOUT)){ |
35 | if ($arg{$h}){ |
b2704119 |
36 | unless (defined find_encoding($arg{$h})) { |
aae85ceb |
37 | require Carp; |
38 | Carp::croak "Unknown encoding for $h, '$arg{$h}'"; |
39 | } |
0ab8f81e |
40 | eval { binmode($h, ":encoding($arg{$h})") }; |
aae85ceb |
41 | }else{ |
42 | unless (exists $arg{$h}){ |
0ab8f81e |
43 | eval { |
44 | no warnings 'uninitialized'; |
45 | binmode($h, ":encoding($name)"); |
46 | }; |
aae85ceb |
47 | } |
48 | } |
49 | if ($@){ |
3ef515df |
50 | require Carp; |
aae85ceb |
51 | Carp::croak($@); |
3ef515df |
52 | } |
3ef515df |
53 | } |
aae85ceb |
54 | }else{ |
55 | defined(${^ENCODING}) and undef ${^ENCODING}; |
56 | eval { |
57 | require Filter::Util::Call ; |
58 | Filter::Util::Call->import ; |
b2704119 |
59 | binmode(STDIN); |
60 | binmode(STDOUT); |
aae85ceb |
61 | filter_add(sub{ |
62 | my $status; |
63 | if (($status = filter_read()) > 0){ |
64 | $_ = $enc->decode($_, 1); |
65 | # warn $_; |
66 | } |
67 | $status ; |
68 | }); |
69 | }; |
70 | # warn "Filter installed"; |
3ef515df |
71 | } |
72 | return 1; # I doubt if we need it, though |
73 | } |
74 | |
75 | sub unimport{ |
76 | no warnings; |
77 | undef ${^ENCODING}; |
b2704119 |
78 | binmode(STDIN); |
79 | binmode(STDOUT); |
aae85ceb |
80 | if ($INC{"Filter/Util/Call.pm"}){ |
81 | eval { filter_del() }; |
82 | } |
3ef515df |
83 | } |
84 | |
85 | 1; |
86 | __END__ |
85982a32 |
87 | |
3ef515df |
88 | =pod |
89 | |
90 | =head1 NAME |
91 | |
0ab8f81e |
92 | encoding - allows you to write your script in non-ascii or non-utf8 |
3ef515df |
93 | |
94 | =head1 SYNOPSIS |
95 | |
962111ca |
96 | use encoding "greek"; # Perl like Greek to you? |
3ef515df |
97 | use encoding "euc-jp"; # Jperl! |
98 | |
962111ca |
99 | # or you can even do this if your shell supports your native encoding |
3ef515df |
100 | |
962111ca |
101 | perl -Mencoding=latin2 -e '...' # Feeling centrally European? |
0ab8f81e |
102 | perl -Mencoding=euc-kr -e '...' # Or Korean? |
3ef515df |
103 | |
104 | # or from the shebang line |
105 | |
962111ca |
106 | #!/your/path/to/perl -Mencoding="8859-6" # Arabian Nights |
0ab8f81e |
107 | #!/your/path/to/perl -Mencoding=big5 # Taiwanese |
3ef515df |
108 | |
109 | # more control |
110 | |
962111ca |
111 | # A simple euc-cn => utf-8 converter |
6d1c0808 |
112 | use encoding "euc-cn", STDOUT => "utf8"; while(<>){print}; |
3ef515df |
113 | |
114 | # "no encoding;" supported (but not scoped!) |
115 | no encoding; |
116 | |
aae85ceb |
117 | # an alternate way, Filter |
118 | use encoding "euc-jp", Filter=>1; |
119 | use utf8; |
120 | # now you can use kanji identifiers -- in euc-jp! |
121 | |
3ef515df |
122 | =head1 ABSTRACT |
123 | |
962111ca |
124 | Let's start with a bit of history: Perl 5.6.0 introduced Unicode |
125 | support. You could apply C<substr()> and regexes even to complex CJK |
126 | characters -- so long as the script was written in UTF-8. But back |
0ab8f81e |
127 | then, text editors that supported UTF-8 were still rare and many users |
128 | instead chose to write scripts in legacy encodings, giving up a whole |
129 | new feature of Perl 5.6. |
3ef515df |
130 | |
0ab8f81e |
131 | Rewind to the future: starting from perl 5.8.0 with the B<encoding> |
962111ca |
132 | pragma, you can write your script in any encoding you like (so long |
133 | as the C<Encode> module supports it) and still enjoy Unicode support. |
0ab8f81e |
134 | You can write code in EUC-JP as follows: |
3ef515df |
135 | |
136 | my $Rakuda = "\xF1\xD1\xF1\xCC"; # Camel in Kanji |
137 | #<-char-><-char-> # 4 octets |
138 | s/\bCamel\b/$Rakuda/; |
139 | |
140 | And with C<use encoding "euc-jp"> in effect, it is the same thing as |
962111ca |
141 | the code in UTF-8: |
3ef515df |
142 | |
32b9ed1f |
143 | my $Rakuda = "\x{99F1}\x{99DD}"; # two Unicode Characters |
3ef515df |
144 | s/\bCamel\b/$Rakuda/; |
145 | |
962111ca |
146 | The B<encoding> pragma also modifies the filehandle disciplines of |
3ef515df |
147 | STDIN, STDOUT, and STDERR to the specified encoding. Therefore, |
148 | |
149 | use encoding "euc-jp"; |
150 | my $message = "Camel is the symbol of perl.\n"; |
151 | my $Rakuda = "\xF1\xD1\xF1\xCC"; # Camel in Kanji |
152 | $message =~ s/\bCamel\b/$Rakuda/; |
153 | print $message; |
154 | |
962111ca |
155 | Will print "\xF1\xD1\xF1\xCC is the symbol of perl.\n", |
156 | not "\x{99F1}\x{99DD} is the symbol of perl.\n". |
3ef515df |
157 | |
0ab8f81e |
158 | You can override this by giving extra arguments; see below. |
3ef515df |
159 | |
160 | =head1 USAGE |
161 | |
162 | =over 4 |
163 | |
164 | =item use encoding [I<ENCNAME>] ; |
165 | |
0ab8f81e |
166 | Sets the script encoding to I<ENCNAME>. Filehandle disciplines of |
167 | STDIN and STDOUT are set to ":encoding(I<ENCNAME>)". Note that STDERR |
168 | will not be changed. |
3ef515df |
169 | |
170 | If no encoding is specified, the environment variable L<PERL_ENCODING> |
962111ca |
171 | is consulted. If no encoding can be found, the error C<Unknown encoding |
172 | 'I<ENCNAME>'> will be thrown. |
3ef515df |
173 | |
174 | Note that non-STD file handles remain unaffected. Use C<use open> or |
175 | C<binmode> to change disciplines of those. |
176 | |
aae85ceb |
177 | =item use encoding I<ENCNAME> [ STDIN =E<gt> I<ENCNAME_IN> ...] ; |
3ef515df |
178 | |
0ab8f81e |
179 | You can also individually set encodings of STDIN and STDOUT via the |
32b9ed1f |
180 | C<< STDIN => I<ENCNAME> >> form. In this case, you cannot omit the |
181 | first I<ENCNAME>. C<< STDIN => undef >> turns the IO transcoding |
aae85ceb |
182 | completely off. |
3ef515df |
183 | |
184 | =item no encoding; |
185 | |
0ab8f81e |
186 | Unsets the script encoding. The disciplines of STDIN, STDOUT are |
962111ca |
187 | reset to ":raw" (the default unprocessed raw stream of bytes). |
3ef515df |
188 | |
189 | =back |
190 | |
191 | =head1 CAVEATS |
192 | |
193 | =head2 NOT SCOPED |
194 | |
195 | The pragma is a per script, not a per block lexical. Only the last |
32b9ed1f |
196 | C<use encoding> or C<no encoding> matters, and it affects B<the whole script>. |
0ab8f81e |
197 | However, the <no encoding> pragma is supported and C<use encoding> can |
962111ca |
198 | appear as many times as you want in a given script. The multiple use |
3ef515df |
199 | of this pragma is discouraged. |
200 | |
201 | =head2 DO NOT MIX MULTIPLE ENCODINGS |
202 | |
203 | Notice that only literals (string or regular expression) having only |
204 | legacy code points are affected: if you mix data like this |
205 | |
206 | \xDF\x{100} |
207 | |
208 | the data is assumed to be in (Latin 1 and) Unicode, not in your native |
209 | encoding. In other words, this will match in "greek": |
210 | |
211 | "\xDF" =~ /\x{3af}/ |
212 | |
213 | but this will not |
214 | |
215 | "\xDF\x{100}" =~ /\x{3af}\x{100}/ |
216 | |
962111ca |
217 | since the C<\xDF> (ISO 8859-7 GREEK SMALL LETTER IOTA WITH TONOS) on |
218 | the left will B<not> be upgraded to C<\x{3af}> (Unicode GREEK SMALL |
219 | LETTER IOTA WITH TONOS) because of the C<\x{100}> on the left. You |
220 | should not be mixing your legacy data and Unicode in the same string. |
3ef515df |
221 | |
222 | This pragma also affects encoding of the 0x80..0xFF code point range: |
223 | normally characters in that range are left as eight-bit bytes (unless |
224 | they are combined with characters with code points 0x100 or larger, |
225 | in which case all characters need to become UTF-8 encoded), but if |
226 | the C<encoding> pragma is present, even the 0x80..0xFF range always |
227 | gets UTF-8 encoded. |
228 | |
229 | After all, the best thing about this pragma is that you don't have to |
0ab8f81e |
230 | resort to \x{....} just to spell your name in a native encoding. |
231 | So feel free to put your strings in your encoding in quotes and |
232 | regexes. |
3ef515df |
233 | |
962111ca |
234 | =head1 Non-ASCII Identifiers and Filter option |
aae85ceb |
235 | |
962111ca |
236 | The magic of C<use encoding> is not applied to the names of |
32b9ed1f |
237 | identifiers. In order to make C<${"\x{4eba}"}++> ($human++, where human |
962111ca |
238 | is a single Han ideograph) work, you still need to write your script |
239 | in UTF-8 or use a source filter. |
aae85ceb |
240 | |
0ab8f81e |
241 | In other words, the same restriction as with Jperl applies. |
aae85ceb |
242 | |
0ab8f81e |
243 | If you dare to experiment, however, you can try the Filter option. |
aae85ceb |
244 | |
245 | =over 4 |
246 | |
247 | =item use encoding I<ENCNAME> Filter=E<gt>1; |
248 | |
0ab8f81e |
249 | This turns the encoding pragma into a source filter. While the default |
aae85ceb |
250 | approach just decodes interpolated literals (in qq() and qr()), this |
0ab8f81e |
251 | will apply a source filter to the entire source code. In this case, |
252 | STDIN and STDOUT remain untouched. |
aae85ceb |
253 | |
254 | =back |
255 | |
962111ca |
256 | What does this mean? Your source code behaves as if it is written in |
0ab8f81e |
257 | UTF-8. So even if your editor only supports Shift_JIS, for example, |
258 | you can still try examples in Chapter 15 of C<Programming Perl, 3rd |
259 | Ed.>. For instance, you can use UTF-8 identifiers. |
aae85ceb |
260 | |
261 | This option is significantly slower and (as of this writing) non-ASCII |
262 | identifiers are not very stable WITHOUT this option and with the |
263 | source code written in UTF-8. |
264 | |
962111ca |
265 | To make your script in legacy encoding work with minimum effort, |
266 | do not use Filter=E<gt>1. |
aae85ceb |
267 | |
3ef515df |
268 | =head1 EXAMPLE - Greekperl |
269 | |
270 | use encoding "iso 8859-7"; |
271 | |
0ab8f81e |
272 | # \xDF in ISO 8859-7 (Greek) is \x{3af} in Unicode. |
3ef515df |
273 | |
274 | $a = "\xDF"; |
275 | $b = "\x{100}"; |
276 | |
277 | printf "%#x\n", ord($a); # will print 0x3af, not 0xdf |
278 | |
279 | $c = $a . $b; |
280 | |
281 | # $c will be "\x{3af}\x{100}", not "\x{df}\x{100}". |
282 | |
283 | # chr() is affected, and ... |
284 | |
285 | print "mega\n" if ord(chr(0xdf)) == 0x3af; |
286 | |
287 | # ... ord() is affected by the encoding pragma ... |
288 | |
289 | print "tera\n" if ord(pack("C", 0xdf)) == 0x3af; |
290 | |
291 | # ... as are eq and cmp ... |
292 | |
293 | print "peta\n" if "\x{3af}" eq pack("C", 0xdf); |
294 | print "exa\n" if "\x{3af}" cmp pack("C", 0xdf) == 0; |
295 | |
296 | # ... but pack/unpack C are not affected, in case you still |
0ab8f81e |
297 | # want to go back to your native encoding |
3ef515df |
298 | |
299 | print "zetta\n" if unpack("C", (pack("C", 0xdf))) == 0xdf; |
300 | |
301 | =head1 KNOWN PROBLEMS |
302 | |
0ab8f81e |
303 | For native multibyte encodings (either fixed or variable length), |
3ef515df |
304 | the current implementation of the regular expressions may introduce |
0ab8f81e |
305 | recoding errors for regular expression literals longer than 127 bytes. |
3ef515df |
306 | |
307 | The encoding pragma is not supported on EBCDIC platforms. |
0ab8f81e |
308 | (Porters who are willing and able to remove this limitation are |
309 | welcome.) |
3ef515df |
310 | |
311 | =head1 SEE ALSO |
312 | |
aae85ceb |
313 | L<perlunicode>, L<Encode>, L<open>, L<Filter::Util::Call>, |
314 | |
315 | Ch. 15 of C<Programming Perl (3rd Edition)> |
316 | by Larry Wall, Tom Christiansen, Jon Orwant; |
317 | O'Reilly & Associates; ISBN 0-596-00027-8 |
3ef515df |
318 | |
319 | =cut |