Commit | Line | Data |
10c5ecbb |
1 | # |
56ff7374 |
2 | # $Id: Encode.pm,v 2.11 2005/08/05 10:58:25 dankogai Exp dankogai $ |
10c5ecbb |
3 | # |
2c674647 |
4 | package Encode; |
51ef4e11 |
5 | use strict; |
56ff7374 |
6 | our $VERSION = sprintf "%d.%02d", q$Revision: 2.11 $ =~ /(\d+)/g; |
8f139f4c |
7 | sub DEBUG () { 0 } |
6d1c0808 |
8 | use XSLoader (); |
10c5ecbb |
9 | XSLoader::load(__PACKAGE__, $VERSION); |
2c674647 |
10 | |
2c674647 |
11 | require Exporter; |
7e19fb92 |
12 | use base qw/Exporter/; |
2c674647 |
13 | |
4411f3b6 |
14 | # Public, encouraged API is exported by default |
85982a32 |
15 | |
16 | our @EXPORT = qw( |
17 | decode decode_utf8 encode encode_utf8 |
a0d8a30e |
18 | encodings find_encoding clone_encoding |
4411f3b6 |
19 | ); |
20 | |
b7a5c9de |
21 | our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC |
56ff7374 |
22 | PERLQQ HTMLCREF XMLCREF STOP_AT_PARTIAL); |
b7a5c9de |
23 | our @FB_CONSTS = qw(FB_DEFAULT FB_CROAK FB_QUIET FB_WARN |
af1f55d9 |
24 | FB_PERLQQ FB_HTMLCREF FB_XMLCREF); |
85982a32 |
25 | |
51ef4e11 |
26 | our @EXPORT_OK = |
6d1c0808 |
27 | ( |
85982a32 |
28 | qw( |
29 | _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit |
30 | is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade |
31 | ), |
32 | @FB_FLAGS, @FB_CONSTS, |
33 | ); |
34 | |
6d1c0808 |
35 | our %EXPORT_TAGS = |
85982a32 |
36 | ( |
37 | all => [ @EXPORT, @EXPORT_OK ], |
38 | fallbacks => [ @FB_CONSTS ], |
39 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], |
40 | ); |
41 | |
4411f3b6 |
42 | # Documentation moved after __END__ for speed - NI-S |
2c674647 |
43 | |
a63c962f |
44 | our $ON_EBCDIC = (ord("A") == 193); |
f2a2953c |
45 | |
5d030b67 |
46 | use Encode::Alias; |
47 | |
5129552c |
48 | # Make a %Encoding package variable to allow a certain amount of cheating |
49 | our %Encoding; |
aae85ceb |
50 | our %ExtModule; |
51 | require Encode::Config; |
52 | eval { require Encode::ConfigLocal }; |
5129552c |
53 | |
656753f8 |
54 | sub encodings |
55 | { |
5129552c |
56 | my $class = shift; |
fc17bd48 |
57 | my %enc; |
58 | if (@_ and $_[0] eq ":all"){ |
59 | %enc = ( %Encoding, %ExtModule ); |
60 | }else{ |
61 | %enc = %Encoding; |
62 | for my $mod (map {m/::/o ? $_ : "Encode::$_" } @_){ |
8f139f4c |
63 | DEBUG and warn $mod; |
fc17bd48 |
64 | for my $enc (keys %ExtModule){ |
65 | $ExtModule{$enc} eq $mod and $enc{$enc} = $mod; |
66 | } |
67 | } |
5129552c |
68 | } |
69 | return |
ce912cd4 |
70 | sort { lc $a cmp lc $b } |
fc17bd48 |
71 | grep {!/^(?:Internal|Unicode|Guess)$/o} keys %enc; |
51ef4e11 |
72 | } |
73 | |
85982a32 |
74 | sub perlio_ok{ |
0ab8f81e |
75 | my $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]); |
011b2d2f |
76 | $obj->can("perlio_ok") and return $obj->perlio_ok(); |
0ab8f81e |
77 | return 0; # safety net |
85982a32 |
78 | } |
79 | |
51ef4e11 |
80 | sub define_encoding |
81 | { |
18586f54 |
82 | my $obj = shift; |
83 | my $name = shift; |
5129552c |
84 | $Encoding{$name} = $obj; |
18586f54 |
85 | my $lc = lc($name); |
86 | define_alias($lc => $obj) unless $lc eq $name; |
10c5ecbb |
87 | while (@_){ |
18586f54 |
88 | my $alias = shift; |
10c5ecbb |
89 | define_alias($alias, $obj); |
18586f54 |
90 | } |
91 | return $obj; |
656753f8 |
92 | } |
93 | |
656753f8 |
94 | sub getEncoding |
95 | { |
10c5ecbb |
96 | my ($class, $name, $skip_external) = @_; |
97 | |
a0d8a30e |
98 | ref($name) && $name->can('renew') and return $name; |
10c5ecbb |
99 | exists $Encoding{$name} and return $Encoding{$name}; |
18586f54 |
100 | my $lc = lc $name; |
10c5ecbb |
101 | exists $Encoding{$lc} and return $Encoding{$lc}; |
c50d192e |
102 | |
5129552c |
103 | my $oc = $class->find_alias($name); |
10c5ecbb |
104 | defined($oc) and return $oc; |
105 | $lc ne $name and $oc = $class->find_alias($lc); |
106 | defined($oc) and return $oc; |
c50d192e |
107 | |
c731e18e |
108 | unless ($skip_external) |
d1ed7747 |
109 | { |
c731e18e |
110 | if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){ |
111 | $mod =~ s,::,/,g ; $mod .= '.pm'; |
112 | eval{ require $mod; }; |
10c5ecbb |
113 | exists $Encoding{$name} and return $Encoding{$name}; |
c731e18e |
114 | } |
d1ed7747 |
115 | } |
18586f54 |
116 | return; |
656753f8 |
117 | } |
118 | |
a0d8a30e |
119 | sub find_encoding($;$) |
4411f3b6 |
120 | { |
10c5ecbb |
121 | my ($name, $skip_external) = @_; |
dd9703c9 |
122 | return __PACKAGE__->getEncoding($name,$skip_external); |
4411f3b6 |
123 | } |
124 | |
a0d8a30e |
125 | sub resolve_alias($){ |
fcb875d4 |
126 | my $obj = find_encoding(shift); |
127 | defined $obj and return $obj->name; |
128 | return; |
129 | } |
130 | |
a0d8a30e |
131 | sub clone_encoding($){ |
132 | my $obj = find_encoding(shift); |
133 | ref $obj or return; |
134 | eval { require Storable }; |
135 | $@ and return; |
136 | return Storable::dclone($obj); |
137 | } |
138 | |
b2704119 |
139 | sub encode($$;$) |
4411f3b6 |
140 | { |
e8c86ba6 |
141 | my ($name, $string, $check) = @_; |
0f7c507f |
142 | return undef unless defined $string; |
78589665 |
143 | $string .= '' if ref $string; # stringify; |
b2704119 |
144 | $check ||=0; |
18586f54 |
145 | my $enc = find_encoding($name); |
10c5ecbb |
146 | unless(defined $enc){ |
147 | require Carp; |
148 | Carp::croak("Unknown encoding '$name'"); |
149 | } |
18586f54 |
150 | my $octets = $enc->encode($string,$check); |
7f0d54d7 |
151 | $_[1] = $string if $check and !($check & LEAVE_SRC()); |
18586f54 |
152 | return $octets; |
4411f3b6 |
153 | } |
154 | |
b2704119 |
155 | sub decode($$;$) |
4411f3b6 |
156 | { |
18586f54 |
157 | my ($name,$octets,$check) = @_; |
0f7c507f |
158 | return undef unless defined $octets; |
78589665 |
159 | $octets .= '' if ref $octets; |
b2704119 |
160 | $check ||=0; |
18586f54 |
161 | my $enc = find_encoding($name); |
10c5ecbb |
162 | unless(defined $enc){ |
163 | require Carp; |
164 | Carp::croak("Unknown encoding '$name'"); |
165 | } |
18586f54 |
166 | my $string = $enc->decode($octets,$check); |
7f0d54d7 |
167 | $_[1] = $octets if $check and !($check & LEAVE_SRC()); |
18586f54 |
168 | return $string; |
4411f3b6 |
169 | } |
170 | |
b2704119 |
171 | sub from_to($$$;$) |
4411f3b6 |
172 | { |
18586f54 |
173 | my ($string,$from,$to,$check) = @_; |
0f7c507f |
174 | return undef unless defined $string; |
b2704119 |
175 | $check ||=0; |
18586f54 |
176 | my $f = find_encoding($from); |
10c5ecbb |
177 | unless (defined $f){ |
178 | require Carp; |
179 | Carp::croak("Unknown encoding '$from'"); |
180 | } |
18586f54 |
181 | my $t = find_encoding($to); |
10c5ecbb |
182 | unless (defined $t){ |
183 | require Carp; |
184 | Carp::croak("Unknown encoding '$to'"); |
185 | } |
18586f54 |
186 | my $uni = $f->decode($string,$check); |
187 | return undef if ($check && length($string)); |
a999c27c |
188 | $string = $t->encode($uni,$check); |
18586f54 |
189 | return undef if ($check && length($uni)); |
3ef515df |
190 | return defined($_[0] = $string) ? length($string) : undef ; |
4411f3b6 |
191 | } |
192 | |
b2704119 |
193 | sub encode_utf8($) |
4411f3b6 |
194 | { |
18586f54 |
195 | my ($str) = @_; |
c731e18e |
196 | utf8::encode($str); |
18586f54 |
197 | return $str; |
4411f3b6 |
198 | } |
199 | |
c2cbba7d |
200 | sub decode_utf8($;$) |
4411f3b6 |
201 | { |
c2cbba7d |
202 | my ($str, $check) = @_; |
203 | if ($check){ |
204 | return decode("utf8", $str, $check); |
205 | }else{ |
dc4a2e29 |
206 | return decode("utf8", $str); |
c2cbba7d |
207 | return $str; |
208 | } |
5ad8ef52 |
209 | } |
210 | |
b536bf57 |
211 | predefine_encodings(1); |
f2a2953c |
212 | |
213 | # |
214 | # This is to restore %Encoding if really needed; |
215 | # |
10c5ecbb |
216 | |
f2a2953c |
217 | sub predefine_encodings{ |
10c5ecbb |
218 | use Encode::Encoding; |
b536bf57 |
219 | no warnings 'redefine'; |
220 | my $use_xs = shift; |
6d1c0808 |
221 | if ($ON_EBCDIC) { |
f2a2953c |
222 | # was in Encode::UTF_EBCDIC |
223 | package Encode::UTF_EBCDIC; |
10c5ecbb |
224 | push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; |
f2a2953c |
225 | *decode = sub{ |
226 | my ($obj,$str,$chk) = @_; |
227 | my $res = ''; |
228 | for (my $i = 0; $i < length($str); $i++) { |
6d1c0808 |
229 | $res .= |
f2a2953c |
230 | chr(utf8::unicode_to_native(ord(substr($str,$i,1)))); |
231 | } |
232 | $_[1] = '' if $chk; |
233 | return $res; |
234 | }; |
235 | *encode = sub{ |
236 | my ($obj,$str,$chk) = @_; |
237 | my $res = ''; |
238 | for (my $i = 0; $i < length($str); $i++) { |
6d1c0808 |
239 | $res .= |
f2a2953c |
240 | chr(utf8::native_to_unicode(ord(substr($str,$i,1)))); |
241 | } |
242 | $_[1] = '' if $chk; |
243 | return $res; |
244 | }; |
6d1c0808 |
245 | $Encode::Encoding{Unicode} = |
c731e18e |
246 | bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC"; |
6d1c0808 |
247 | } else { |
f2a2953c |
248 | package Encode::Internal; |
10c5ecbb |
249 | push @Encode::Internal::ISA, 'Encode::Encoding'; |
f2a2953c |
250 | *decode = sub{ |
251 | my ($obj,$str,$chk) = @_; |
252 | utf8::upgrade($str); |
253 | $_[1] = '' if $chk; |
254 | return $str; |
255 | }; |
256 | *encode = \&decode; |
6d1c0808 |
257 | $Encode::Encoding{Unicode} = |
c731e18e |
258 | bless {Name => "Internal"} => "Encode::Internal"; |
f2a2953c |
259 | } |
260 | |
261 | { |
262 | # was in Encode::utf8 |
263 | package Encode::utf8; |
10c5ecbb |
264 | push @Encode::utf8::ISA, 'Encode::Encoding'; |
b536bf57 |
265 | # |
266 | if ($use_xs){ |
8f139f4c |
267 | Encode::DEBUG and warn __PACKAGE__, " XS on"; |
b536bf57 |
268 | *decode = \&decode_xs; |
269 | *encode = \&encode_xs; |
270 | }else{ |
8f139f4c |
271 | Encode::DEBUG and warn __PACKAGE__, " XS off"; |
b536bf57 |
272 | *decode = sub{ |
273 | my ($obj,$octets,$chk) = @_; |
274 | my $str = Encode::decode_utf8($octets); |
275 | if (defined $str) { |
276 | $_[1] = '' if $chk; |
277 | return $str; |
278 | } |
279 | return undef; |
280 | }; |
281 | *encode = sub { |
282 | my ($obj,$string,$chk) = @_; |
283 | my $octets = Encode::encode_utf8($string); |
284 | $_[1] = '' if $chk; |
285 | return $octets; |
286 | }; |
287 | } |
220e2d4e |
288 | *cat_decode = sub{ # ($obj, $dst, $src, $pos, $trm, $chk) |
289 | my ($obj, undef, undef, $pos, $trm) = @_; # currently ignores $chk |
290 | my ($rdst, $rsrc, $rpos) = \@_[1,2,3]; |
291 | use bytes; |
292 | if ((my $npos = index($$rsrc, $trm, $pos)) >= 0) { |
293 | $$rdst .= substr($$rsrc, $pos, $npos - $pos + length($trm)); |
294 | $$rpos = $npos + length($trm); |
295 | return 1; |
296 | } |
297 | $$rdst .= substr($$rsrc, $pos); |
298 | $$rpos = length($$rsrc); |
299 | return ''; |
300 | }; |
b7a5c9de |
301 | $Encode::Encoding{utf8} = |
c731e18e |
302 | bless {Name => "utf8"} => "Encode::utf8"; |
7f0d54d7 |
303 | $Encode::Encoding{"utf-8-strict"} = |
304 | bless {Name => "utf-8-strict", strict_utf8 => 1 } => "Encode::utf8"; |
f2a2953c |
305 | } |
f2a2953c |
306 | } |
307 | |
656753f8 |
308 | 1; |
309 | |
2a936312 |
310 | __END__ |
311 | |
4411f3b6 |
312 | =head1 NAME |
313 | |
314 | Encode - character encodings |
315 | |
316 | =head1 SYNOPSIS |
317 | |
318 | use Encode; |
319 | |
67d7b5ef |
320 | =head2 Table of Contents |
321 | |
0ab8f81e |
322 | Encode consists of a collection of modules whose details are too big |
67d7b5ef |
323 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 |
324 | and general topics at a glance. For other topics and more details, |
0ab8f81e |
325 | see the PODs below: |
67d7b5ef |
326 | |
327 | Name Description |
328 | -------------------------------------------------------- |
6d1c0808 |
329 | Encode::Alias Alias definitions to encodings |
67d7b5ef |
330 | Encode::Encoding Encode Implementation Base Class |
331 | Encode::Supported List of Supported Encodings |
332 | Encode::CN Simplified Chinese Encodings |
333 | Encode::JP Japanese Encodings |
334 | Encode::KR Korean Encodings |
335 | Encode::TW Traditional Chinese Encodings |
336 | -------------------------------------------------------- |
337 | |
4411f3b6 |
338 | =head1 DESCRIPTION |
339 | |
47bfe92f |
340 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef |
341 | and the rest of the system. Perl strings are sequences of |
342 | B<characters>. |
343 | |
344 | The repertoire of characters that Perl can represent is at least that |
345 | defined by the Unicode Consortium. On most platforms the ordinal |
346 | values of the characters (as returned by C<ord(ch)>) is the "Unicode |
347 | codepoint" for the character (the exceptions are those platforms where |
348 | the legacy encoding is some variant of EBCDIC rather than a super-set |
349 | of ASCII - see L<perlebcdic>). |
350 | |
0ab8f81e |
351 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef |
352 | often called "bytes". These chunks are also known as "octets" in |
353 | networking standards. Perl is widely used to manipulate data of many |
354 | types - not only strings of characters representing human or computer |
0ab8f81e |
355 | languages but also "binary" data being the machine's representation of |
67d7b5ef |
356 | numbers, pixels in an image - or just about anything. |
357 | |
0ab8f81e |
358 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef |
359 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e |
360 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef |
361 | "logical character". |
362 | |
363 | =head2 TERMINOLOGY |
4411f3b6 |
364 | |
7e19fb92 |
365 | =over 2 |
21938dfa |
366 | |
67d7b5ef |
367 | =item * |
368 | |
369 | I<character>: a character in the range 0..(2**32-1) (or more). |
370 | (What Perl's strings are made of.) |
371 | |
372 | =item * |
373 | |
374 | I<byte>: a character in the range 0..255 |
375 | (A special case of a Perl character.) |
376 | |
377 | =item * |
378 | |
379 | I<octet>: 8 bits of data, with ordinal values 0..255 |
0ab8f81e |
380 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef |
381 | |
382 | =back |
4411f3b6 |
383 | |
67d7b5ef |
384 | =head1 PERL ENCODING API |
4411f3b6 |
385 | |
7e19fb92 |
386 | =over 2 |
4411f3b6 |
387 | |
b7a5c9de |
388 | =item $octets = encode(ENCODING, $string [, CHECK]) |
4411f3b6 |
389 | |
0ab8f81e |
390 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef |
391 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e |
392 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
393 | For CHECK, see L</"Handling Malformed Data">. |
4411f3b6 |
394 | |
b7a5c9de |
395 | For example, to convert a string from Perl's internal format to |
6d1c0808 |
396 | iso-8859-1 (also known as Latin1), |
681a7c68 |
397 | |
b7a5c9de |
398 | $octets = encode("iso-8859-1", $string); |
7e19fb92 |
399 | |
b7a5c9de |
400 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets |
401 | B<may not be equal to> $string. Though they both contain the same data, the utf8 flag |
7e19fb92 |
402 | for $octets is B<always> off. When you encode anything, utf8 flag of |
403 | the result is always off, even when it contains completely valid utf8 |
404 | string. See L</"The UTF-8 flag"> below. |
681a7c68 |
405 | |
7f0d54d7 |
406 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
407 | |
b7a5c9de |
408 | =item $string = decode(ENCODING, $octets [, CHECK]) |
4411f3b6 |
409 | |
0ab8f81e |
410 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's |
411 | internal form and returns the resulting string. As in encode(), |
412 | ENCODING can be either a canonical name or an alias. For encoding names |
413 | and aliases, see L</"Defining Aliases">. For CHECK, see |
47bfe92f |
414 | L</"Handling Malformed Data">. |
415 | |
b7a5c9de |
416 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: |
681a7c68 |
417 | |
b7a5c9de |
418 | $string = decode("iso-8859-1", $octets); |
681a7c68 |
419 | |
b7a5c9de |
420 | B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string |
421 | B<may not be equal to> $octets. Though they both contain the same data, |
422 | the utf8 flag for $string is on unless $octets entirely consists of |
7e19fb92 |
423 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag"> |
424 | below. |
47bfe92f |
425 | |
7f0d54d7 |
426 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
427 | |
b7a5c9de |
428 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) |
7e19fb92 |
429 | |
b7a5c9de |
430 | Converts B<in-place> data between two encodings. The data in $octets |
431 | must be encoded as octets and not as characters in Perl's internal |
f9d05ba3 |
432 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 |
433 | encoding: |
2b106fbe |
434 | |
b7a5c9de |
435 | from_to($octets, "iso-8859-1", "cp1250"); |
2b106fbe |
436 | |
437 | and to convert it back: |
438 | |
b7a5c9de |
439 | from_to($octets, "cp1250", "iso-8859-1"); |
4411f3b6 |
440 | |
ab97ca19 |
441 | Note that because the conversion happens in place, the data to be |
0ab8f81e |
442 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 |
443 | |
f9d05ba3 |
444 | from_to() returns the length of the converted string in octets on |
445 | success, I<undef> on error. |
3ef515df |
446 | |
b7a5c9de |
447 | B<CAVEAT>: The following operations look the same but are not quite so; |
7e19fb92 |
448 | |
b7a5c9de |
449 | from_to($data, "iso-8859-1", "utf8"); #1 |
7e19fb92 |
450 | $data = decode("iso-8859-1", $data); #2 |
4411f3b6 |
451 | |
b7a5c9de |
452 | Both #1 and #2 make $data consist of a completely valid UTF-8 string |
7e19fb92 |
453 | but only #2 turns utf8 flag on. #1 is equivalent to |
f2a2953c |
454 | |
7e19fb92 |
455 | $data = encode("utf8", decode("iso-8859-1", $data)); |
f2a2953c |
456 | |
7e19fb92 |
457 | See L</"The UTF-8 flag"> below. |
f2a2953c |
458 | |
459 | =item $octets = encode_utf8($string); |
460 | |
7e19fb92 |
461 | Equivalent to C<$octets = encode("utf8", $string);> The characters |
b7a5c9de |
462 | that comprise $string are encoded in Perl's internal format and the |
463 | result is returned as a sequence of octets. All possible |
7e19fb92 |
464 | characters have a UTF-8 representation so this function cannot fail. |
465 | |
f2a2953c |
466 | |
467 | =item $string = decode_utf8($octets [, CHECK]); |
468 | |
7e19fb92 |
469 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. |
b7a5c9de |
470 | The sequence of octets represented by |
7e19fb92 |
471 | $octets is decoded from UTF-8 into a sequence of logical |
472 | characters. Not all sequences of octets form valid UTF-8 encodings, so |
473 | it is possible for this call to fail. For CHECK, see |
474 | L</"Handling Malformed Data">. |
f2a2953c |
475 | |
476 | =back |
477 | |
51ef4e11 |
478 | =head2 Listing available encodings |
479 | |
5129552c |
480 | use Encode; |
481 | @list = Encode->encodings(); |
482 | |
483 | Returns a list of the canonical names of the available encodings that |
484 | are loaded. To get a list of all available encodings including the |
485 | ones that are not loaded yet, say |
486 | |
487 | @all_encodings = Encode->encodings(":all"); |
488 | |
0ab8f81e |
489 | Or you can give the name of a specific module. |
5129552c |
490 | |
c731e18e |
491 | @with_jp = Encode->encodings("Encode::JP"); |
492 | |
493 | When "::" is not in the name, "Encode::" is assumed. |
51ef4e11 |
494 | |
c731e18e |
495 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 |
496 | |
0ab8f81e |
497 | To find out in detail which encodings are supported by this package, |
5d030b67 |
498 | see L<Encode::Supported>. |
51ef4e11 |
499 | |
500 | =head2 Defining Aliases |
501 | |
0ab8f81e |
502 | To add a new alias to a given encoding, use: |
67d7b5ef |
503 | |
5129552c |
504 | use Encode; |
505 | use Encode::Alias; |
a63c962f |
506 | define_alias(newName => ENCODING); |
51ef4e11 |
507 | |
3ef515df |
508 | After that, newName can be used as an alias for ENCODING. |
f2a2953c |
509 | ENCODING may be either the name of an encoding or an |
510 | I<encoding object> |
51ef4e11 |
511 | |
fcb875d4 |
512 | But before you do so, make sure the alias is nonexistent with |
513 | C<resolve_alias()>, which returns the canonical name thereof. |
514 | i.e. |
515 | |
516 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true |
517 | Encode::resolve_alias("iso-8859-12") # false; nonexistent |
518 | Encode::resolve_alias($name) eq $name # true if $name is canonical |
519 | |
0ab8f81e |
520 | resolve_alias() does not need C<use Encode::Alias>; it can be |
521 | exported via C<use Encode qw(resolve_alias)>. |
fcb875d4 |
522 | |
0ab8f81e |
523 | See L<Encode::Alias> for details. |
51ef4e11 |
524 | |
85982a32 |
525 | =head1 Encoding via PerlIO |
4411f3b6 |
526 | |
b7a5c9de |
527 | If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode |
0ab8f81e |
528 | and encode directly via a filehandle. The following two examples |
529 | are totally identical in their functionality. |
4411f3b6 |
530 | |
85982a32 |
531 | # via PerlIO |
532 | open my $in, "<:encoding(shiftjis)", $infile or die; |
533 | open my $out, ">:encoding(euc-jp)", $outfile or die; |
b7a5c9de |
534 | while(<$in>){ print $out $_; } |
8e86646e |
535 | |
85982a32 |
536 | # via from_to |
0ab8f81e |
537 | open my $in, "<", $infile or die; |
538 | open my $out, ">", $outfile or die; |
b7a5c9de |
539 | while(<$in>){ |
0ab8f81e |
540 | from_to($_, "shiftjis", "euc-jp", 1); |
b7a5c9de |
541 | print $out $_; |
85982a32 |
542 | } |
4411f3b6 |
543 | |
b7a5c9de |
544 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check |
0ab8f81e |
545 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
546 | method. |
547 | |
548 | Encode::perlio_ok("hz"); # False |
549 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available |
550 | |
551 | use Encode qw(perlio_ok); # exported upon request |
552 | perlio_ok("euc-jp") |
4411f3b6 |
553 | |
0ab8f81e |
554 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
f9d05ba3 |
555 | except for hz and ISO-2022-kr. For gory details, see |
556 | L<Encode::Encoding> and L<Encode::PerlIO>. |
4411f3b6 |
557 | |
85982a32 |
558 | =head1 Handling Malformed Data |
4411f3b6 |
559 | |
f9d05ba3 |
560 | The optional I<CHECK> argument is used as follows. When you omit it, |
561 | Encode::FB_DEFAULT ( == 0 ) is assumed. |
562 | |
563 | =over 2 |
564 | |
3c4b39be |
565 | =item B<NOTE:> Not all encoding support this feature |
f9d05ba3 |
566 | |
567 | Some encodings ignore I<CHECK> argument. For example, |
568 | L<Encode::Unicode> ignores I<CHECK> and it always croaks on error. |
569 | |
570 | =back |
571 | |
572 | Now here is the list of I<CHECK> values available |
47bfe92f |
573 | |
151b5d36 |
574 | =over 2 |
575 | |
85982a32 |
576 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f |
577 | |
f9d05ba3 |
578 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> in |
78589665 |
579 | place of a malformed character. When you encode, E<lt>subcharE<gt> |
580 | will be used. When you decode the code point C<0xFFFD> is used. If |
581 | the data is supposed to be UTF-8, an optional lexical warning |
582 | (category utf8) is given. |
e9692b5b |
583 | |
7e19fb92 |
584 | =item I<CHECK> = Encode::FB_CROAK ( == 1) |
e9692b5b |
585 | |
b7a5c9de |
586 | If I<CHECK> is 1, methods will die on error immediately with an error |
0ab8f81e |
587 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
f9d05ba3 |
588 | error with eval{} unless you really want to let it die. |
47bfe92f |
589 | |
85982a32 |
590 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f |
591 | |
85982a32 |
592 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
f9d05ba3 |
593 | return the portion of the data that has been processed so far when an |
594 | error occurs. The data argument will be overwritten with everything |
595 | after that point (that is, the unprocessed part of data). This is |
596 | handy when you have to call decode repeatedly in the case where your |
597 | source data may contain partial multi-byte character sequences, |
598 | (i.e. you are reading with a fixed-width buffer). Here is a sample |
599 | code that does exactly this: |
4411f3b6 |
600 | |
78589665 |
601 | my $buffer = ''; my $string = ''; |
602 | while(read $fh, $buffer, 256, length($buffer)){ |
603 | $string .= decode($encoding, $buffer, Encode::FB_QUIET); |
604 | # $buffer now contains the unprocessed partial character |
85982a32 |
605 | } |
1768d7eb |
606 | |
85982a32 |
607 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef |
608 | |
0ab8f81e |
609 | This is the same as above, except that it warns on error. Handy when |
610 | you are debugging the mode above. |
85982a32 |
611 | |
612 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) |
613 | |
af1f55d9 |
614 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) |
615 | |
616 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) |
617 | |
85982a32 |
618 | For encodings that are implemented by Encode::XS, CHECK == |
619 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. |
620 | |
b7a5c9de |
621 | When you decode, C<\xI<HH>> will be inserted for a malformed character, |
622 | where I<HH> is the hex representation of the octet that could not be |
623 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, |
624 | where I<HHHH> is the Unicode ID of the character that cannot be found |
0ab8f81e |
625 | in the character repertoire of the encoding. |
85982a32 |
626 | |
af1f55d9 |
627 | HTML/XML character reference modes are about the same, in place of |
78589665 |
628 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and |
629 | XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number. |
af1f55d9 |
630 | |
7f0d54d7 |
631 | In Encode 2.10 or later, C<LEAVE_SRC> is also implied. |
632 | |
85982a32 |
633 | =item The bitmask |
634 | |
0ab8f81e |
635 | These modes are actually set via a bitmask. Here is how the FB_XX |
636 | constants are laid out. You can import the FB_XX constants via |
637 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask |
638 | constants via C<use Encode qw(:fallback_all)>. |
85982a32 |
639 | |
b0b300a3 |
640 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
641 | DIE_ON_ERR 0x0001 X |
4089adc4 |
642 | WARN_ON_ERR 0x0002 X |
b0b300a3 |
643 | RETURN_ON_ERR 0x0004 X X |
7f0d54d7 |
644 | LEAVE_SRC 0x0008 X |
b0b300a3 |
645 | PERLQQ 0x0100 X |
b7a5c9de |
646 | HTMLCREF 0x0200 |
647 | XMLCREF 0x0400 |
67d7b5ef |
648 | |
151b5d36 |
649 | =back |
650 | |
0ab8f81e |
651 | =head2 Unimplemented fallback schemes |
67d7b5ef |
652 | |
0ab8f81e |
653 | In the future, you will be able to use a code reference to a callback |
f2a2953c |
654 | function for the value of I<CHECK> but its API is still undecided. |
67d7b5ef |
655 | |
982a4085 |
656 | The fallback scheme does not work on EBCDIC platforms. |
657 | |
67d7b5ef |
658 | =head1 Defining Encodings |
659 | |
660 | To define a new encoding, use: |
661 | |
b7a5c9de |
662 | use Encode qw(define_encoding); |
67d7b5ef |
663 | define_encoding($object, 'canonicalName' [, alias...]); |
664 | |
665 | I<canonicalName> will be associated with I<$object>. The object |
0ab8f81e |
666 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef |
667 | If more than two arguments are provided then additional |
b7a5c9de |
668 | arguments are taken as aliases for I<$object>. |
67d7b5ef |
669 | |
f2a2953c |
670 | See L<Encode::Encoding> for more details. |
671 | |
7e19fb92 |
672 | =head1 The UTF-8 flag |
673 | |
674 | Before the introduction of utf8 support in perl, The C<eq> operator |
b7a5c9de |
675 | just compared the strings represented by two scalars. Beginning with |
676 | perl 5.8, C<eq> compares two strings with simultaneous consideration |
677 | of I<the utf8 flag>. To explain why we made it so, I will quote page |
678 | 402 of C<Programming Perl, 3rd ed.> |
7e19fb92 |
679 | |
680 | =over 2 |
681 | |
682 | =item Goal #1: |
683 | |
684 | Old byte-oriented programs should not spontaneously break on the old |
685 | byte-oriented data they used to work on. |
686 | |
687 | =item Goal #2: |
688 | |
689 | Old byte-oriented programs should magically start working on the new |
690 | character-oriented data when appropriate. |
691 | |
692 | =item Goal #3: |
693 | |
694 | Programs should run just as fast in the new character-oriented mode |
695 | as in the old byte-oriented mode. |
696 | |
697 | =item Goal #4: |
698 | |
699 | Perl should remain one language, rather than forking into a |
700 | byte-oriented Perl and a character-oriented Perl. |
701 | |
702 | =back |
703 | |
704 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 |
705 | was born and many features documented in the book remained |
b7a5c9de |
706 | unimplemented for a long time. Perl 5.8 corrected this and the introduction |
707 | of the UTF-8 flag is one of them. You can think of this perl notion as of a |
708 | byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8 |
7e19fb92 |
709 | flag on). |
710 | |
711 | Here is how Encode takes care of the utf8 flag. |
712 | |
4bdf5738 |
713 | =over 2 |
7e19fb92 |
714 | |
715 | =item * |
716 | |
717 | When you encode, the resulting utf8 flag is always off. |
718 | |
151b5d36 |
719 | =item * |
7e19fb92 |
720 | |
b7a5c9de |
721 | When you decode, the resulting utf8 flag is on unless you can |
7e19fb92 |
722 | unambiguously represent data. Here is the definition of |
723 | dis-ambiguity. |
724 | |
b7a5c9de |
725 | After C<$utf8 = decode('foo', $octet);>, |
7e19fb92 |
726 | |
727 | When $octet is... The utf8 flag in $utf8 is |
728 | --------------------------------------------- |
729 | In ASCII only (or EBCDIC only) OFF |
730 | In ISO-8859-1 ON |
731 | In any other Encoding ON |
732 | --------------------------------------------- |
733 | |
3c4b39be |
734 | As you see, there is one exception, In ASCII. That way you can assume |
7e19fb92 |
735 | Goal #1. And with Encode Goal #2 is assumed but you still have to be |
736 | careful in such cases mentioned in B<CAVEAT> paragraphs. |
737 | |
738 | This utf8 flag is not visible in perl scripts, exactly for the same |
739 | reason you cannot (or you I<don't have to>) see if a scalar contains a |
740 | string, integer, or floating point number. But you can still peek |
741 | and poke these if you will. See the section below. |
742 | |
743 | =back |
744 | |
745 | =head2 Messing with Perl's Internals |
4411f3b6 |
746 | |
47bfe92f |
747 | The following API uses parts of Perl's internals in the current |
0ab8f81e |
748 | implementation. As such, they are efficient but may change. |
4411f3b6 |
749 | |
7e19fb92 |
750 | =over 2 |
4411f3b6 |
751 | |
a63c962f |
752 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 |
753 | |
0ab8f81e |
754 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. |
47bfe92f |
755 | If CHECK is true, also checks the data in STRING for being well-formed |
756 | UTF-8. Returns true if successful, false otherwise. |
4411f3b6 |
757 | |
2c246b25 |
758 | As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). |
b5ab1f6f |
759 | |
a63c962f |
760 | =item _utf8_on(STRING) |
4411f3b6 |
761 | |
0ab8f81e |
762 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is |
4411f3b6 |
763 | B<not> checked for being well-formed UTF-8. Do not use unless you |
764 | B<know> that the STRING is well-formed UTF-8. Returns the previous |
0ab8f81e |
765 | state of the UTF-8 flag (so please don't treat the return value as |
766 | indicating success or failure), or C<undef> if STRING is not a string. |
4411f3b6 |
767 | |
a63c962f |
768 | =item _utf8_off(STRING) |
4411f3b6 |
769 | |
0ab8f81e |
770 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. |
771 | Returns the previous state of the UTF-8 flag (so please don't treat the |
772 | return value as indicating success or failure), or C<undef> if STRING is |
4411f3b6 |
773 | not a string. |
774 | |
775 | =back |
776 | |
7f0d54d7 |
777 | =head1 UTF-8 vs. utf8 |
778 | |
779 | ....We now view strings not as sequences of bytes, but as sequences |
780 | of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit |
781 | computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed. |
782 | |
783 | That has been the perl's notion of UTF-8 but official UTF-8 is more |
784 | strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are |
785 | not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al). |
786 | |
787 | Now that is overruled by Larry Wall himself. |
788 | |
789 | From: Larry Wall <larry@wall.org> |
790 | Date: December 04, 2004 11:51:58 JST |
791 | To: perl-unicode@perl.org |
792 | Subject: Re: Make Encode.pm support the real UTF-8 |
793 | Message-Id: <20041204025158.GA28754@wall.org> |
794 | |
795 | On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote: |
796 | : I've no problem with 'utf8' being perl's unrestricted uft8 encoding, |
797 | : but "UTF-8" is the name of the standard and should give the |
798 | : corresponding behaviour. |
799 | |
800 | For what it's worth, that's how I've always kept them straight in my |
801 | head. |
802 | |
803 | Also for what it's worth, Perl 6 will mostly default to strict but |
804 | make it easy to switch back to lax. |
805 | |
806 | Larry |
807 | |
808 | Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8 |
809 | while B<utf8> means liberal, lax, version thereof. And Encode version |
810 | 2.10 or later thus groks the difference between C<UTF-8> and C"utf8". |
811 | |
812 | encode("utf8", "\x{FFFF_FFFF}", 1); # okay |
813 | encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks |
814 | |
815 | C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>. |
816 | Yes, the hyphen between "UTF" and "8" is important. Without it Encode |
817 | goes "liberal" |
818 | |
819 | find_encoding("UTF-8")->name # is 'utf-8-strict' |
820 | find_encoding("utf-8")->name # ditto. names are case insensitive |
821 | find_encoding("utf8")->name # ditto. "_" are treated as "-" |
822 | find_encoding("UTF8")->name # is 'utf8'. |
823 | |
824 | |
4411f3b6 |
825 | =head1 SEE ALSO |
826 | |
5d030b67 |
827 | L<Encode::Encoding>, |
828 | L<Encode::Supported>, |
6d1c0808 |
829 | L<Encode::PerlIO>, |
5d030b67 |
830 | L<encoding>, |
6d1c0808 |
831 | L<perlebcdic>, |
832 | L<perlfunc/open>, |
833 | L<perlunicode>, |
834 | L<utf8>, |
5d030b67 |
835 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 |
836 | |
85982a32 |
837 | =head1 MAINTAINER |
aae85ceb |
838 | |
839 | This project was originated by Nick Ing-Simmons and later maintained |
7e19fb92 |
840 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full |
841 | list of people involved. For any questions, use |
b7a5c9de |
842 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. |
aae85ceb |
843 | |
4411f3b6 |
844 | =cut |