Commit | Line | Data |
10c5ecbb |
1 | # |
40bed538 |
2 | # $Id: Encode.pm,v 2.31 2009/02/16 06:13:11 dankogai Exp $ |
10c5ecbb |
3 | # |
2c674647 |
4 | package Encode; |
51ef4e11 |
5 | use strict; |
656ebd29 |
6 | use warnings; |
40bed538 |
7 | our $VERSION = sprintf "%d.%02d", q$Revision: 2.31 $ =~ /(\d+)/g; |
8f139f4c |
8 | sub DEBUG () { 0 } |
6d1c0808 |
9 | use XSLoader (); |
d1256cb1 |
10 | XSLoader::load( __PACKAGE__, $VERSION ); |
2c674647 |
11 | |
2c674647 |
12 | require Exporter; |
7e19fb92 |
13 | use base qw/Exporter/; |
2c674647 |
14 | |
4411f3b6 |
15 | # Public, encouraged API is exported by default |
85982a32 |
16 | |
17 | our @EXPORT = qw( |
0a8c69ed |
18 | decode decode_utf8 encode encode_utf8 str2bytes bytes2str |
a0d8a30e |
19 | encodings find_encoding clone_encoding |
4411f3b6 |
20 | ); |
d1256cb1 |
21 | our @FB_FLAGS = qw( |
22 | DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC |
23 | PERLQQ HTMLCREF XMLCREF STOP_AT_PARTIAL |
24 | ); |
25 | our @FB_CONSTS = qw( |
26 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN |
27 | FB_PERLQQ FB_HTMLCREF FB_XMLCREF |
28 | ); |
29 | our @EXPORT_OK = ( |
30 | qw( |
31 | _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit |
32 | is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade |
85982a32 |
33 | ), |
d1256cb1 |
34 | @FB_FLAGS, @FB_CONSTS, |
35 | ); |
85982a32 |
36 | |
d1256cb1 |
37 | our %EXPORT_TAGS = ( |
38 | all => [ @EXPORT, @EXPORT_OK ], |
0263186c |
39 | default => [ @EXPORT ], |
40 | fallbacks => [ @FB_CONSTS ], |
d1256cb1 |
41 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], |
42 | ); |
85982a32 |
43 | |
4411f3b6 |
44 | # Documentation moved after __END__ for speed - NI-S |
2c674647 |
45 | |
d1256cb1 |
46 | our $ON_EBCDIC = ( ord("A") == 193 ); |
f2a2953c |
47 | |
5d030b67 |
48 | use Encode::Alias; |
49 | |
5129552c |
50 | # Make a %Encoding package variable to allow a certain amount of cheating |
51 | our %Encoding; |
aae85ceb |
52 | our %ExtModule; |
53 | require Encode::Config; |
2fd0906e |
54 | # See |
55 | # https://bugzilla.redhat.com/show_bug.cgi?id=435505#c2 |
56 | # to find why sig handers inside eval{} are disabled. |
57 | eval { |
58 | local $SIG{__DIE__}; |
59 | local $SIG{__WARN__}; |
60 | require Encode::ConfigLocal; |
61 | }; |
5129552c |
62 | |
d1256cb1 |
63 | sub encodings { |
5129552c |
64 | my $class = shift; |
fc17bd48 |
65 | my %enc; |
d1256cb1 |
66 | if ( @_ and $_[0] eq ":all" ) { |
67 | %enc = ( %Encoding, %ExtModule ); |
5129552c |
68 | } |
d1256cb1 |
69 | else { |
70 | %enc = %Encoding; |
71 | for my $mod ( map { m/::/o ? $_ : "Encode::$_" } @_ ) { |
72 | DEBUG and warn $mod; |
73 | for my $enc ( keys %ExtModule ) { |
74 | $ExtModule{$enc} eq $mod and $enc{$enc} = $mod; |
75 | } |
76 | } |
77 | } |
78 | return sort { lc $a cmp lc $b } |
79 | grep { !/^(?:Internal|Unicode|Guess)$/o } keys %enc; |
51ef4e11 |
80 | } |
81 | |
d1256cb1 |
82 | sub perlio_ok { |
83 | my $obj = ref( $_[0] ) ? $_[0] : find_encoding( $_[0] ); |
011b2d2f |
84 | $obj->can("perlio_ok") and return $obj->perlio_ok(); |
d1256cb1 |
85 | return 0; # safety net |
85982a32 |
86 | } |
87 | |
d1256cb1 |
88 | sub define_encoding { |
18586f54 |
89 | my $obj = shift; |
90 | my $name = shift; |
5129552c |
91 | $Encoding{$name} = $obj; |
18586f54 |
92 | my $lc = lc($name); |
d1256cb1 |
93 | define_alias( $lc => $obj ) unless $lc eq $name; |
94 | while (@_) { |
95 | my $alias = shift; |
96 | define_alias( $alias, $obj ); |
18586f54 |
97 | } |
98 | return $obj; |
656753f8 |
99 | } |
100 | |
d1256cb1 |
101 | sub getEncoding { |
102 | my ( $class, $name, $skip_external ) = @_; |
10c5ecbb |
103 | |
a0d8a30e |
104 | ref($name) && $name->can('renew') and return $name; |
10c5ecbb |
105 | exists $Encoding{$name} and return $Encoding{$name}; |
18586f54 |
106 | my $lc = lc $name; |
10c5ecbb |
107 | exists $Encoding{$lc} and return $Encoding{$lc}; |
c50d192e |
108 | |
5129552c |
109 | my $oc = $class->find_alias($name); |
10c5ecbb |
110 | defined($oc) and return $oc; |
111 | $lc ne $name and $oc = $class->find_alias($lc); |
112 | defined($oc) and return $oc; |
c50d192e |
113 | |
d1256cb1 |
114 | unless ($skip_external) { |
115 | if ( my $mod = $ExtModule{$name} || $ExtModule{$lc} ) { |
116 | $mod =~ s,::,/,g; |
117 | $mod .= '.pm'; |
118 | eval { require $mod; }; |
119 | exists $Encoding{$name} and return $Encoding{$name}; |
120 | } |
d1ed7747 |
121 | } |
18586f54 |
122 | return; |
656753f8 |
123 | } |
124 | |
d1256cb1 |
125 | sub find_encoding($;$) { |
126 | my ( $name, $skip_external ) = @_; |
127 | return __PACKAGE__->getEncoding( $name, $skip_external ); |
4411f3b6 |
128 | } |
129 | |
d1256cb1 |
130 | sub resolve_alias($) { |
fcb875d4 |
131 | my $obj = find_encoding(shift); |
132 | defined $obj and return $obj->name; |
133 | return; |
134 | } |
135 | |
d1256cb1 |
136 | sub clone_encoding($) { |
a0d8a30e |
137 | my $obj = find_encoding(shift); |
138 | ref $obj or return; |
139 | eval { require Storable }; |
140 | $@ and return; |
141 | return Storable::dclone($obj); |
142 | } |
143 | |
d1256cb1 |
144 | sub encode($$;$) { |
145 | my ( $name, $string, $check ) = @_; |
0f7c507f |
146 | return undef unless defined $string; |
d1256cb1 |
147 | $string .= '' if ref $string; # stringify; |
148 | $check ||= 0; |
18586f54 |
149 | my $enc = find_encoding($name); |
d1256cb1 |
150 | unless ( defined $enc ) { |
151 | require Carp; |
152 | Carp::croak("Unknown encoding '$name'"); |
10c5ecbb |
153 | } |
d1256cb1 |
154 | my $octets = $enc->encode( $string, $check ); |
7828f908 |
155 | $_[1] = $string if $check and !ref $check and !( $check & LEAVE_SRC() ); |
18586f54 |
156 | return $octets; |
4411f3b6 |
157 | } |
0a8c69ed |
158 | *str2bytes = \&encode; |
4411f3b6 |
159 | |
d1256cb1 |
160 | sub decode($$;$) { |
161 | my ( $name, $octets, $check ) = @_; |
0f7c507f |
162 | return undef unless defined $octets; |
78589665 |
163 | $octets .= '' if ref $octets; |
d1256cb1 |
164 | $check ||= 0; |
18586f54 |
165 | my $enc = find_encoding($name); |
d1256cb1 |
166 | unless ( defined $enc ) { |
167 | require Carp; |
168 | Carp::croak("Unknown encoding '$name'"); |
10c5ecbb |
169 | } |
d1256cb1 |
170 | my $string = $enc->decode( $octets, $check ); |
7828f908 |
171 | $_[1] = $octets if $check and !ref $check and !( $check & LEAVE_SRC() ); |
18586f54 |
172 | return $string; |
4411f3b6 |
173 | } |
0a8c69ed |
174 | *bytes2str = \&decode; |
4411f3b6 |
175 | |
d1256cb1 |
176 | sub from_to($$$;$) { |
177 | my ( $string, $from, $to, $check ) = @_; |
0f7c507f |
178 | return undef unless defined $string; |
d1256cb1 |
179 | $check ||= 0; |
18586f54 |
180 | my $f = find_encoding($from); |
d1256cb1 |
181 | unless ( defined $f ) { |
182 | require Carp; |
183 | Carp::croak("Unknown encoding '$from'"); |
10c5ecbb |
184 | } |
18586f54 |
185 | my $t = find_encoding($to); |
d1256cb1 |
186 | unless ( defined $t ) { |
187 | require Carp; |
188 | Carp::croak("Unknown encoding '$to'"); |
10c5ecbb |
189 | } |
41c240f5 |
190 | my $uni = $f->decode($string); |
d1256cb1 |
191 | $_[0] = $string = $t->encode( $uni, $check ); |
192 | return undef if ( $check && length($uni) ); |
193 | return defined( $_[0] ) ? length($string) : undef; |
4411f3b6 |
194 | } |
195 | |
d1256cb1 |
196 | sub encode_utf8($) { |
18586f54 |
197 | my ($str) = @_; |
c731e18e |
198 | utf8::encode($str); |
18586f54 |
199 | return $str; |
4411f3b6 |
200 | } |
201 | |
d1256cb1 |
202 | sub decode_utf8($;$) { |
203 | my ( $str, $check ) = @_; |
41c240f5 |
204 | return $str if is_utf8($str); |
d1256cb1 |
205 | if ($check) { |
206 | return decode( "utf8", $str, $check ); |
207 | } |
208 | else { |
209 | return decode( "utf8", $str ); |
210 | return $str; |
c2cbba7d |
211 | } |
5ad8ef52 |
212 | } |
213 | |
b536bf57 |
214 | predefine_encodings(1); |
f2a2953c |
215 | |
216 | # |
217 | # This is to restore %Encoding if really needed; |
218 | # |
10c5ecbb |
219 | |
d1256cb1 |
220 | sub predefine_encodings { |
51e4e64d |
221 | require Encode::Encoding; |
b536bf57 |
222 | no warnings 'redefine'; |
223 | my $use_xs = shift; |
6d1c0808 |
224 | if ($ON_EBCDIC) { |
d1256cb1 |
225 | |
226 | # was in Encode::UTF_EBCDIC |
227 | package Encode::UTF_EBCDIC; |
228 | push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; |
229 | *decode = sub { |
230 | my ( $obj, $str, $chk ) = @_; |
231 | my $res = ''; |
232 | for ( my $i = 0 ; $i < length($str) ; $i++ ) { |
233 | $res .= |
234 | chr( |
235 | utf8::unicode_to_native( ord( substr( $str, $i, 1 ) ) ) |
236 | ); |
237 | } |
238 | $_[1] = '' if $chk; |
239 | return $res; |
240 | }; |
241 | *encode = sub { |
242 | my ( $obj, $str, $chk ) = @_; |
243 | my $res = ''; |
244 | for ( my $i = 0 ; $i < length($str) ; $i++ ) { |
245 | $res .= |
246 | chr( |
247 | utf8::native_to_unicode( ord( substr( $str, $i, 1 ) ) ) |
248 | ); |
249 | } |
250 | $_[1] = '' if $chk; |
251 | return $res; |
252 | }; |
253 | $Encode::Encoding{Unicode} = |
254 | bless { Name => "UTF_EBCDIC" } => "Encode::UTF_EBCDIC"; |
255 | } |
256 | else { |
257 | |
258 | package Encode::Internal; |
259 | push @Encode::Internal::ISA, 'Encode::Encoding'; |
260 | *decode = sub { |
261 | my ( $obj, $str, $chk ) = @_; |
262 | utf8::upgrade($str); |
263 | $_[1] = '' if $chk; |
264 | return $str; |
265 | }; |
266 | *encode = \&decode; |
267 | $Encode::Encoding{Unicode} = |
268 | bless { Name => "Internal" } => "Encode::Internal"; |
f2a2953c |
269 | } |
270 | |
271 | { |
d1256cb1 |
272 | |
273 | # was in Encode::utf8 |
274 | package Encode::utf8; |
275 | push @Encode::utf8::ISA, 'Encode::Encoding'; |
276 | |
277 | # |
278 | if ($use_xs) { |
279 | Encode::DEBUG and warn __PACKAGE__, " XS on"; |
280 | *decode = \&decode_xs; |
281 | *encode = \&encode_xs; |
282 | } |
283 | else { |
284 | Encode::DEBUG and warn __PACKAGE__, " XS off"; |
285 | *decode = sub { |
286 | my ( $obj, $octets, $chk ) = @_; |
287 | my $str = Encode::decode_utf8($octets); |
288 | if ( defined $str ) { |
289 | $_[1] = '' if $chk; |
290 | return $str; |
291 | } |
292 | return undef; |
293 | }; |
294 | *encode = sub { |
295 | my ( $obj, $string, $chk ) = @_; |
296 | my $octets = Encode::encode_utf8($string); |
297 | $_[1] = '' if $chk; |
298 | return $octets; |
299 | }; |
300 | } |
301 | *cat_decode = sub { # ($obj, $dst, $src, $pos, $trm, $chk) |
302 | # currently ignores $chk |
303 | my ( $obj, undef, undef, $pos, $trm ) = @_; |
304 | my ( $rdst, $rsrc, $rpos ) = \@_[ 1, 2, 3 ]; |
305 | use bytes; |
306 | if ( ( my $npos = index( $$rsrc, $trm, $pos ) ) >= 0 ) { |
307 | $$rdst .= |
308 | substr( $$rsrc, $pos, $npos - $pos + length($trm) ); |
309 | $$rpos = $npos + length($trm); |
310 | return 1; |
311 | } |
312 | $$rdst .= substr( $$rsrc, $pos ); |
313 | $$rpos = length($$rsrc); |
314 | return ''; |
315 | }; |
316 | $Encode::Encoding{utf8} = |
317 | bless { Name => "utf8" } => "Encode::utf8"; |
318 | $Encode::Encoding{"utf-8-strict"} = |
319 | bless { Name => "utf-8-strict", strict_utf8 => 1 } => |
320 | "Encode::utf8"; |
f2a2953c |
321 | } |
f2a2953c |
322 | } |
323 | |
656753f8 |
324 | 1; |
325 | |
2a936312 |
326 | __END__ |
327 | |
4411f3b6 |
328 | =head1 NAME |
329 | |
330 | Encode - character encodings |
331 | |
332 | =head1 SYNOPSIS |
333 | |
334 | use Encode; |
335 | |
67d7b5ef |
336 | =head2 Table of Contents |
337 | |
0ab8f81e |
338 | Encode consists of a collection of modules whose details are too big |
67d7b5ef |
339 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 |
340 | and general topics at a glance. For other topics and more details, |
0ab8f81e |
341 | see the PODs below: |
67d7b5ef |
342 | |
343 | Name Description |
344 | -------------------------------------------------------- |
6d1c0808 |
345 | Encode::Alias Alias definitions to encodings |
67d7b5ef |
346 | Encode::Encoding Encode Implementation Base Class |
347 | Encode::Supported List of Supported Encodings |
348 | Encode::CN Simplified Chinese Encodings |
349 | Encode::JP Japanese Encodings |
350 | Encode::KR Korean Encodings |
351 | Encode::TW Traditional Chinese Encodings |
352 | -------------------------------------------------------- |
353 | |
4411f3b6 |
354 | =head1 DESCRIPTION |
355 | |
47bfe92f |
356 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef |
357 | and the rest of the system. Perl strings are sequences of |
358 | B<characters>. |
359 | |
360 | The repertoire of characters that Perl can represent is at least that |
361 | defined by the Unicode Consortium. On most platforms the ordinal |
362 | values of the characters (as returned by C<ord(ch)>) is the "Unicode |
363 | codepoint" for the character (the exceptions are those platforms where |
364 | the legacy encoding is some variant of EBCDIC rather than a super-set |
365 | of ASCII - see L<perlebcdic>). |
366 | |
0ab8f81e |
367 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef |
368 | often called "bytes". These chunks are also known as "octets" in |
369 | networking standards. Perl is widely used to manipulate data of many |
370 | types - not only strings of characters representing human or computer |
0ab8f81e |
371 | languages but also "binary" data being the machine's representation of |
67d7b5ef |
372 | numbers, pixels in an image - or just about anything. |
373 | |
0ab8f81e |
374 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef |
375 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e |
376 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef |
377 | "logical character". |
378 | |
379 | =head2 TERMINOLOGY |
4411f3b6 |
380 | |
7e19fb92 |
381 | =over 2 |
21938dfa |
382 | |
67d7b5ef |
383 | =item * |
384 | |
385 | I<character>: a character in the range 0..(2**32-1) (or more). |
386 | (What Perl's strings are made of.) |
387 | |
388 | =item * |
389 | |
390 | I<byte>: a character in the range 0..255 |
391 | (A special case of a Perl character.) |
392 | |
393 | =item * |
394 | |
395 | I<octet>: 8 bits of data, with ordinal values 0..255 |
0ab8f81e |
396 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef |
397 | |
398 | =back |
4411f3b6 |
399 | |
67d7b5ef |
400 | =head1 PERL ENCODING API |
4411f3b6 |
401 | |
7e19fb92 |
402 | =over 2 |
4411f3b6 |
403 | |
b7a5c9de |
404 | =item $octets = encode(ENCODING, $string [, CHECK]) |
4411f3b6 |
405 | |
0ab8f81e |
406 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef |
407 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e |
408 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
409 | For CHECK, see L</"Handling Malformed Data">. |
4411f3b6 |
410 | |
b7a5c9de |
411 | For example, to convert a string from Perl's internal format to |
6d1c0808 |
412 | iso-8859-1 (also known as Latin1), |
681a7c68 |
413 | |
b7a5c9de |
414 | $octets = encode("iso-8859-1", $string); |
7e19fb92 |
415 | |
44b3b9c7 |
416 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then |
417 | $octets B<may not be equal to> $string. Though they both contain the |
418 | same data, the UTF8 flag for $octets is B<always> off. When you |
419 | encode anything, UTF8 flag of the result is always off, even when it |
420 | contains completely valid utf8 string. See L</"The UTF8 flag"> below. |
681a7c68 |
421 | |
7f0d54d7 |
422 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
423 | |
b7a5c9de |
424 | =item $string = decode(ENCODING, $octets [, CHECK]) |
4411f3b6 |
425 | |
0ab8f81e |
426 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's |
427 | internal form and returns the resulting string. As in encode(), |
428 | ENCODING can be either a canonical name or an alias. For encoding names |
429 | and aliases, see L</"Defining Aliases">. For CHECK, see |
47bfe92f |
430 | L</"Handling Malformed Data">. |
431 | |
b7a5c9de |
432 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: |
681a7c68 |
433 | |
b7a5c9de |
434 | $string = decode("iso-8859-1", $octets); |
681a7c68 |
435 | |
b7a5c9de |
436 | B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string |
437 | B<may not be equal to> $octets. Though they both contain the same data, |
2575c402 |
438 | the UTF8 flag for $string is on unless $octets entirely consists of |
439 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF8 flag"> |
7e19fb92 |
440 | below. |
47bfe92f |
441 | |
7f0d54d7 |
442 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
443 | |
44b3b9c7 |
444 | =item [$obj =] find_encoding(ENCODING) |
445 | |
446 | Returns the I<encoding object> corresponding to ENCODING. Returns |
447 | undef if no matching ENCODING is find. |
448 | |
449 | This object is what actually does the actual (en|de)coding. |
450 | |
451 | $utf8 = decode($name, $bytes); |
452 | |
453 | is in fact |
454 | |
455 | $utf8 = do{ |
456 | $obj = find_encoding($name); |
457 | croak qq(encoding "$name" not found) unless ref $obj; |
458 | $obj->decode($bytes) |
459 | }; |
460 | |
461 | with more error checking. |
462 | |
463 | Therefore you can save time by reusing this object as follows; |
464 | |
465 | my $enc = find_encoding("iso-8859-1"); |
466 | while(<>){ |
467 | my $utf8 = $enc->decode($_); |
468 | # and do someting with $utf8; |
469 | } |
470 | |
471 | Besides C<< ->decode >> and C<< ->encode >>, other methods are |
472 | available as well. For instance, C<< -> name >> returns the canonical |
473 | name of the encoding object. |
474 | |
475 | find_encoding("latin1")->name; # iso-8859-1 |
476 | |
477 | See L<Encode::Encoding> for details. |
478 | |
b7a5c9de |
479 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) |
7e19fb92 |
480 | |
b7a5c9de |
481 | Converts B<in-place> data between two encodings. The data in $octets |
482 | must be encoded as octets and not as characters in Perl's internal |
f9d05ba3 |
483 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 |
484 | encoding: |
2b106fbe |
485 | |
b7a5c9de |
486 | from_to($octets, "iso-8859-1", "cp1250"); |
2b106fbe |
487 | |
488 | and to convert it back: |
489 | |
b7a5c9de |
490 | from_to($octets, "cp1250", "iso-8859-1"); |
4411f3b6 |
491 | |
ab97ca19 |
492 | Note that because the conversion happens in place, the data to be |
0ab8f81e |
493 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 |
494 | |
f9d05ba3 |
495 | from_to() returns the length of the converted string in octets on |
496 | success, I<undef> on error. |
3ef515df |
497 | |
b7a5c9de |
498 | B<CAVEAT>: The following operations look the same but are not quite so; |
7e19fb92 |
499 | |
b7a5c9de |
500 | from_to($data, "iso-8859-1", "utf8"); #1 |
7e19fb92 |
501 | $data = decode("iso-8859-1", $data); #2 |
4411f3b6 |
502 | |
b7a5c9de |
503 | Both #1 and #2 make $data consist of a completely valid UTF-8 string |
2575c402 |
504 | but only #2 turns UTF8 flag on. #1 is equivalent to |
f2a2953c |
505 | |
7e19fb92 |
506 | $data = encode("utf8", decode("iso-8859-1", $data)); |
f2a2953c |
507 | |
2575c402 |
508 | See L</"The UTF8 flag"> below. |
f2a2953c |
509 | |
7828f908 |
510 | Also note that |
511 | |
512 | from_to($octets, $from, $to, $check); |
513 | |
514 | is equivalent to |
515 | |
516 | $octets = encode($to, decode($from, $octets), $check); |
517 | |
518 | Yes, it does not respect the $check during decoding. It is |
519 | deliberately done that way. If you need minute control, C<decode> |
520 | then C<encode> as follows; |
521 | |
522 | $octets = encode($to, decode($from, $octets, $check_from), $check_to); |
523 | |
f2a2953c |
524 | =item $octets = encode_utf8($string); |
525 | |
7e19fb92 |
526 | Equivalent to C<$octets = encode("utf8", $string);> The characters |
b7a5c9de |
527 | that comprise $string are encoded in Perl's internal format and the |
528 | result is returned as a sequence of octets. All possible |
7e19fb92 |
529 | characters have a UTF-8 representation so this function cannot fail. |
530 | |
f2a2953c |
531 | |
532 | =item $string = decode_utf8($octets [, CHECK]); |
533 | |
7e19fb92 |
534 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. |
b7a5c9de |
535 | The sequence of octets represented by |
7e19fb92 |
536 | $octets is decoded from UTF-8 into a sequence of logical |
537 | characters. Not all sequences of octets form valid UTF-8 encodings, so |
538 | it is possible for this call to fail. For CHECK, see |
539 | L</"Handling Malformed Data">. |
f2a2953c |
540 | |
541 | =back |
542 | |
51ef4e11 |
543 | =head2 Listing available encodings |
544 | |
5129552c |
545 | use Encode; |
546 | @list = Encode->encodings(); |
547 | |
548 | Returns a list of the canonical names of the available encodings that |
549 | are loaded. To get a list of all available encodings including the |
550 | ones that are not loaded yet, say |
551 | |
552 | @all_encodings = Encode->encodings(":all"); |
553 | |
0ab8f81e |
554 | Or you can give the name of a specific module. |
5129552c |
555 | |
c731e18e |
556 | @with_jp = Encode->encodings("Encode::JP"); |
557 | |
558 | When "::" is not in the name, "Encode::" is assumed. |
51ef4e11 |
559 | |
c731e18e |
560 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 |
561 | |
0ab8f81e |
562 | To find out in detail which encodings are supported by this package, |
5d030b67 |
563 | see L<Encode::Supported>. |
51ef4e11 |
564 | |
565 | =head2 Defining Aliases |
566 | |
0ab8f81e |
567 | To add a new alias to a given encoding, use: |
67d7b5ef |
568 | |
5129552c |
569 | use Encode; |
570 | use Encode::Alias; |
a63c962f |
571 | define_alias(newName => ENCODING); |
51ef4e11 |
572 | |
3ef515df |
573 | After that, newName can be used as an alias for ENCODING. |
f2a2953c |
574 | ENCODING may be either the name of an encoding or an |
575 | I<encoding object> |
51ef4e11 |
576 | |
fcb875d4 |
577 | But before you do so, make sure the alias is nonexistent with |
578 | C<resolve_alias()>, which returns the canonical name thereof. |
579 | i.e. |
580 | |
581 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true |
582 | Encode::resolve_alias("iso-8859-12") # false; nonexistent |
583 | Encode::resolve_alias($name) eq $name # true if $name is canonical |
584 | |
0ab8f81e |
585 | resolve_alias() does not need C<use Encode::Alias>; it can be |
586 | exported via C<use Encode qw(resolve_alias)>. |
fcb875d4 |
587 | |
0ab8f81e |
588 | See L<Encode::Alias> for details. |
51ef4e11 |
589 | |
742555bd |
590 | =head2 Finding IANA Character Set Registry names |
591 | |
592 | The canonical name of a given encoding does not necessarily agree with |
593 | IANA IANA Character Set Registry, commonly seen as C<< Content-Type: |
594 | text/plain; charset=I<whatever> >>. For most cases canonical names |
595 | work but sometimes it does not (notably 'utf-8-strict'). |
596 | |
597 | Therefore as of Encode version 2.21, a new method C<mime_name()> is added. |
598 | |
599 | use Encode; |
600 | my $enc = find_encoding('UTF-8'); |
601 | warn $enc->name; # utf-8-strict |
602 | warn $enc->mime_name; # UTF-8 |
603 | |
604 | See also: L<Encode::Encoding> |
605 | |
85982a32 |
606 | =head1 Encoding via PerlIO |
4411f3b6 |
607 | |
44b3b9c7 |
608 | If your perl supports I<PerlIO> (which is the default), you can use a |
609 | PerlIO layer to decode and encode directly via a filehandle. The |
610 | following two examples are totally identical in their functionality. |
4411f3b6 |
611 | |
85982a32 |
612 | # via PerlIO |
613 | open my $in, "<:encoding(shiftjis)", $infile or die; |
614 | open my $out, ">:encoding(euc-jp)", $outfile or die; |
b7a5c9de |
615 | while(<$in>){ print $out $_; } |
8e86646e |
616 | |
85982a32 |
617 | # via from_to |
0ab8f81e |
618 | open my $in, "<", $infile or die; |
619 | open my $out, ">", $outfile or die; |
b7a5c9de |
620 | while(<$in>){ |
0ab8f81e |
621 | from_to($_, "shiftjis", "euc-jp", 1); |
b7a5c9de |
622 | print $out $_; |
85982a32 |
623 | } |
4411f3b6 |
624 | |
b7a5c9de |
625 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check |
0ab8f81e |
626 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
627 | method. |
628 | |
629 | Encode::perlio_ok("hz"); # False |
630 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available |
631 | |
632 | use Encode qw(perlio_ok); # exported upon request |
633 | perlio_ok("euc-jp") |
4411f3b6 |
634 | |
0ab8f81e |
635 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
f9d05ba3 |
636 | except for hz and ISO-2022-kr. For gory details, see |
637 | L<Encode::Encoding> and L<Encode::PerlIO>. |
4411f3b6 |
638 | |
85982a32 |
639 | =head1 Handling Malformed Data |
4411f3b6 |
640 | |
8e180e82 |
641 | The optional I<CHECK> argument tells Encode what to do when it |
642 | encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 ) |
643 | is assumed. |
644 | |
645 | As of version 2.12 Encode supports coderef values for CHECK. See below. |
f9d05ba3 |
646 | |
647 | =over 2 |
648 | |
3c4b39be |
649 | =item B<NOTE:> Not all encoding support this feature |
f9d05ba3 |
650 | |
651 | Some encodings ignore I<CHECK> argument. For example, |
652 | L<Encode::Unicode> ignores I<CHECK> and it always croaks on error. |
653 | |
654 | =back |
655 | |
656 | Now here is the list of I<CHECK> values available |
47bfe92f |
657 | |
151b5d36 |
658 | =over 2 |
659 | |
85982a32 |
660 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f |
661 | |
f9d05ba3 |
662 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> in |
78589665 |
663 | place of a malformed character. When you encode, E<lt>subcharE<gt> |
664 | will be used. When you decode the code point C<0xFFFD> is used. If |
665 | the data is supposed to be UTF-8, an optional lexical warning |
666 | (category utf8) is given. |
e9692b5b |
667 | |
7e19fb92 |
668 | =item I<CHECK> = Encode::FB_CROAK ( == 1) |
e9692b5b |
669 | |
b7a5c9de |
670 | If I<CHECK> is 1, methods will die on error immediately with an error |
0ab8f81e |
671 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
f9d05ba3 |
672 | error with eval{} unless you really want to let it die. |
47bfe92f |
673 | |
85982a32 |
674 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f |
675 | |
85982a32 |
676 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
f9d05ba3 |
677 | return the portion of the data that has been processed so far when an |
678 | error occurs. The data argument will be overwritten with everything |
679 | after that point (that is, the unprocessed part of data). This is |
680 | handy when you have to call decode repeatedly in the case where your |
681 | source data may contain partial multi-byte character sequences, |
682 | (i.e. you are reading with a fixed-width buffer). Here is a sample |
683 | code that does exactly this: |
4411f3b6 |
684 | |
78589665 |
685 | my $buffer = ''; my $string = ''; |
686 | while(read $fh, $buffer, 256, length($buffer)){ |
687 | $string .= decode($encoding, $buffer, Encode::FB_QUIET); |
688 | # $buffer now contains the unprocessed partial character |
85982a32 |
689 | } |
1768d7eb |
690 | |
85982a32 |
691 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef |
692 | |
0ab8f81e |
693 | This is the same as above, except that it warns on error. Handy when |
694 | you are debugging the mode above. |
85982a32 |
695 | |
696 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) |
697 | |
af1f55d9 |
698 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) |
699 | |
700 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) |
701 | |
85982a32 |
702 | For encodings that are implemented by Encode::XS, CHECK == |
703 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. |
704 | |
b7a5c9de |
705 | When you decode, C<\xI<HH>> will be inserted for a malformed character, |
706 | where I<HH> is the hex representation of the octet that could not be |
707 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, |
708 | where I<HHHH> is the Unicode ID of the character that cannot be found |
0ab8f81e |
709 | in the character repertoire of the encoding. |
85982a32 |
710 | |
af1f55d9 |
711 | HTML/XML character reference modes are about the same, in place of |
78589665 |
712 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and |
713 | XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number. |
af1f55d9 |
714 | |
7f0d54d7 |
715 | In Encode 2.10 or later, C<LEAVE_SRC> is also implied. |
716 | |
85982a32 |
717 | =item The bitmask |
718 | |
0ab8f81e |
719 | These modes are actually set via a bitmask. Here is how the FB_XX |
720 | constants are laid out. You can import the FB_XX constants via |
721 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask |
722 | constants via C<use Encode qw(:fallback_all)>. |
85982a32 |
723 | |
b0b300a3 |
724 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
725 | DIE_ON_ERR 0x0001 X |
4089adc4 |
726 | WARN_ON_ERR 0x0002 X |
b0b300a3 |
727 | RETURN_ON_ERR 0x0004 X X |
7f0d54d7 |
728 | LEAVE_SRC 0x0008 X |
b0b300a3 |
729 | PERLQQ 0x0100 X |
b7a5c9de |
730 | HTMLCREF 0x0200 |
731 | XMLCREF 0x0400 |
67d7b5ef |
732 | |
151b5d36 |
733 | =back |
734 | |
44b3b9c7 |
735 | =over 2 |
736 | |
51e4e64d |
737 | =item Encode::LEAVE_SRC |
738 | |
739 | If the C<Encode::LEAVE_SRC> bit is not set, but I<CHECK> is, then the second |
740 | argument to C<encode()> or C<decode()> may be assigned to by the functions. If |
741 | you're not interested in this, then bitwise-or the bitmask with it. |
742 | |
44b3b9c7 |
743 | =back |
744 | |
0dbed2e5 |
745 | =head2 coderef for CHECK |
8e180e82 |
746 | |
747 | As of Encode 2.12 CHECK can also be a code reference which takes the |
748 | ord value of unmapped caharacter as an argument and returns a string |
749 | that represents the fallback character. For instance, |
67d7b5ef |
750 | |
8e180e82 |
751 | $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift }); |
67d7b5ef |
752 | |
8e180e82 |
753 | Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of |
754 | \x{I<XXXX>}. |
982a4085 |
755 | |
67d7b5ef |
756 | =head1 Defining Encodings |
757 | |
758 | To define a new encoding, use: |
759 | |
b7a5c9de |
760 | use Encode qw(define_encoding); |
67d7b5ef |
761 | define_encoding($object, 'canonicalName' [, alias...]); |
762 | |
763 | I<canonicalName> will be associated with I<$object>. The object |
0ab8f81e |
764 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef |
765 | If more than two arguments are provided then additional |
b7a5c9de |
766 | arguments are taken as aliases for I<$object>. |
67d7b5ef |
767 | |
f2a2953c |
768 | See L<Encode::Encoding> for more details. |
769 | |
2575c402 |
770 | =head1 The UTF8 flag |
7e19fb92 |
771 | |
2575c402 |
772 | Before the introduction of Unicode support in perl, The C<eq> operator |
b7a5c9de |
773 | just compared the strings represented by two scalars. Beginning with |
2575c402 |
774 | perl 5.8, C<eq> compares two strings with simultaneous consideration of |
775 | I<the UTF8 flag>. To explain why we made it so, I will quote page 402 of |
776 | C<Programming Perl, 3rd ed.> |
7e19fb92 |
777 | |
778 | =over 2 |
779 | |
780 | =item Goal #1: |
781 | |
782 | Old byte-oriented programs should not spontaneously break on the old |
783 | byte-oriented data they used to work on. |
784 | |
785 | =item Goal #2: |
786 | |
787 | Old byte-oriented programs should magically start working on the new |
788 | character-oriented data when appropriate. |
789 | |
790 | =item Goal #3: |
791 | |
792 | Programs should run just as fast in the new character-oriented mode |
793 | as in the old byte-oriented mode. |
794 | |
795 | =item Goal #4: |
796 | |
797 | Perl should remain one language, rather than forking into a |
798 | byte-oriented Perl and a character-oriented Perl. |
799 | |
800 | =back |
801 | |
802 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 |
803 | was born and many features documented in the book remained |
b7a5c9de |
804 | unimplemented for a long time. Perl 5.8 corrected this and the introduction |
2575c402 |
805 | of the UTF8 flag is one of them. You can think of this perl notion as of a |
806 | byte-oriented mode (UTF8 flag off) and a character-oriented mode (UTF8 |
7e19fb92 |
807 | flag on). |
808 | |
2575c402 |
809 | Here is how Encode takes care of the UTF8 flag. |
7e19fb92 |
810 | |
4bdf5738 |
811 | =over 2 |
7e19fb92 |
812 | |
813 | =item * |
814 | |
2575c402 |
815 | When you encode, the resulting UTF8 flag is always off. |
7e19fb92 |
816 | |
151b5d36 |
817 | =item * |
7e19fb92 |
818 | |
2575c402 |
819 | When you decode, the resulting UTF8 flag is on unless you can |
7e19fb92 |
820 | unambiguously represent data. Here is the definition of |
821 | dis-ambiguity. |
822 | |
b7a5c9de |
823 | After C<$utf8 = decode('foo', $octet);>, |
7e19fb92 |
824 | |
2575c402 |
825 | When $octet is... The UTF8 flag in $utf8 is |
7e19fb92 |
826 | --------------------------------------------- |
827 | In ASCII only (or EBCDIC only) OFF |
828 | In ISO-8859-1 ON |
829 | In any other Encoding ON |
830 | --------------------------------------------- |
831 | |
3c4b39be |
832 | As you see, there is one exception, In ASCII. That way you can assume |
7e19fb92 |
833 | Goal #1. And with Encode Goal #2 is assumed but you still have to be |
834 | careful in such cases mentioned in B<CAVEAT> paragraphs. |
835 | |
2575c402 |
836 | This UTF8 flag is not visible in perl scripts, exactly for the same |
7e19fb92 |
837 | reason you cannot (or you I<don't have to>) see if a scalar contains a |
838 | string, integer, or floating point number. But you can still peek |
839 | and poke these if you will. See the section below. |
840 | |
841 | =back |
842 | |
843 | =head2 Messing with Perl's Internals |
4411f3b6 |
844 | |
47bfe92f |
845 | The following API uses parts of Perl's internals in the current |
0ab8f81e |
846 | implementation. As such, they are efficient but may change. |
4411f3b6 |
847 | |
7e19fb92 |
848 | =over 2 |
4411f3b6 |
849 | |
a63c962f |
850 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 |
851 | |
2575c402 |
852 | [INTERNAL] Tests whether the UTF8 flag is turned on in the STRING. |
47bfe92f |
853 | If CHECK is true, also checks the data in STRING for being well-formed |
854 | UTF-8. Returns true if successful, false otherwise. |
4411f3b6 |
855 | |
2c246b25 |
856 | As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). |
b5ab1f6f |
857 | |
a63c962f |
858 | =item _utf8_on(STRING) |
4411f3b6 |
859 | |
2575c402 |
860 | [INTERNAL] Turns on the UTF8 flag in STRING. The data in STRING is |
4411f3b6 |
861 | B<not> checked for being well-formed UTF-8. Do not use unless you |
862 | B<know> that the STRING is well-formed UTF-8. Returns the previous |
2575c402 |
863 | state of the UTF8 flag (so please don't treat the return value as |
0ab8f81e |
864 | indicating success or failure), or C<undef> if STRING is not a string. |
4411f3b6 |
865 | |
64bc6d54 |
866 | This function does not work on tainted values. |
867 | |
a63c962f |
868 | =item _utf8_off(STRING) |
4411f3b6 |
869 | |
2575c402 |
870 | [INTERNAL] Turns off the UTF8 flag in STRING. Do not use frivolously. |
871 | Returns the previous state of the UTF8 flag (so please don't treat the |
0ab8f81e |
872 | return value as indicating success or failure), or C<undef> if STRING is |
4411f3b6 |
873 | not a string. |
874 | |
64bc6d54 |
875 | This function does not work on tainted values. |
876 | |
4411f3b6 |
877 | =back |
878 | |
2575c402 |
879 | =head1 UTF-8 vs. utf8 vs. UTF8 |
7f0d54d7 |
880 | |
881 | ....We now view strings not as sequences of bytes, but as sequences |
882 | of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit |
883 | computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed. |
884 | |
885 | That has been the perl's notion of UTF-8 but official UTF-8 is more |
886 | strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are |
887 | not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al). |
888 | |
889 | Now that is overruled by Larry Wall himself. |
890 | |
891 | From: Larry Wall <larry@wall.org> |
892 | Date: December 04, 2004 11:51:58 JST |
893 | To: perl-unicode@perl.org |
894 | Subject: Re: Make Encode.pm support the real UTF-8 |
895 | Message-Id: <20041204025158.GA28754@wall.org> |
896 | |
897 | On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote: |
898 | : I've no problem with 'utf8' being perl's unrestricted uft8 encoding, |
899 | : but "UTF-8" is the name of the standard and should give the |
900 | : corresponding behaviour. |
901 | |
902 | For what it's worth, that's how I've always kept them straight in my |
903 | head. |
8e180e82 |
904 | |
7f0d54d7 |
905 | Also for what it's worth, Perl 6 will mostly default to strict but |
906 | make it easy to switch back to lax. |
907 | |
908 | Larry |
909 | |
910 | Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8 |
911 | while B<utf8> means liberal, lax, version thereof. And Encode version |
912 | 2.10 or later thus groks the difference between C<UTF-8> and C"utf8". |
913 | |
914 | encode("utf8", "\x{FFFF_FFFF}", 1); # okay |
915 | encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks |
916 | |
917 | C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>. |
918 | Yes, the hyphen between "UTF" and "8" is important. Without it Encode |
919 | goes "liberal" |
920 | |
921 | find_encoding("UTF-8")->name # is 'utf-8-strict' |
922 | find_encoding("utf-8")->name # ditto. names are case insensitive |
50c1ac04 |
923 | find_encoding("utf_8")->name # ditto. "_" are treated as "-" |
7f0d54d7 |
924 | find_encoding("UTF8")->name # is 'utf8'. |
925 | |
2575c402 |
926 | The UTF8 flag is internally called UTF8, without a hyphen. It indicates |
927 | whether a string is internally encoded as utf8, also without a hypen. |
7f0d54d7 |
928 | |
4411f3b6 |
929 | =head1 SEE ALSO |
930 | |
5d030b67 |
931 | L<Encode::Encoding>, |
932 | L<Encode::Supported>, |
6d1c0808 |
933 | L<Encode::PerlIO>, |
5d030b67 |
934 | L<encoding>, |
6d1c0808 |
935 | L<perlebcdic>, |
936 | L<perlfunc/open>, |
370462a2 |
937 | L<perlunicode>, L<perluniintro>, L<perlunifaq>, L<perlunitut> |
6d1c0808 |
938 | L<utf8>, |
5d030b67 |
939 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 |
940 | |
85982a32 |
941 | =head1 MAINTAINER |
aae85ceb |
942 | |
943 | This project was originated by Nick Ing-Simmons and later maintained |
7e19fb92 |
944 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full |
945 | list of people involved. For any questions, use |
b7a5c9de |
946 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. |
aae85ceb |
947 | |
d1256cb1 |
948 | While Dan Kogai retains the copyright as a maintainer, the credit |
949 | should go to all those involoved. See AUTHORS for those submitted |
950 | codes. |
951 | |
952 | =head1 COPYRIGHT |
953 | |
954 | Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt> |
955 | |
956 | This library is free software; you can redistribute it and/or modify |
957 | it under the same terms as Perl itself. |
958 | |
4411f3b6 |
959 | =cut |