3 our $VERSION = do { my @r = (q$Revision: 1.51 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
6 XSLoader::load 'Encode';
9 our @ISA = qw(Exporter);
11 # Public, encouraged API is exported by default
14 decode decode_utf8 encode encode_utf8
15 encodings find_encoding
18 our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC PERLQQ);
19 our @FB_CONSTS = qw(FB_DEFAULT FB_QUIET FB_WARN FB_PERLQQ FB_CROAK);
24 _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit
25 is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade
27 @FB_FLAGS, @FB_CONSTS,
32 all => [ @EXPORT, @EXPORT_OK ],
33 fallbacks => [ @FB_CONSTS ],
34 fallback_all => [ @FB_CONSTS, @FB_FLAGS ],
37 # Documentation moved after __END__ for speed - NI-S
41 our $ON_EBCDIC = (ord("A") == 193);
45 # Make a %Encoding package variable to allow a certain amount of cheating
48 require Encode::Config;
49 eval { require Encode::ConfigLocal };
54 my @modules = (@_ and $_[0] eq ":all") ? values %ExtModule : @_;
55 for my $mod (@modules){
56 $mod =~ s,::,/,g or $mod = "Encode/$mod";
58 $DEBUG and warn "about to require $mod;";
59 eval { require $mod; };
61 my %modules = map {$_ => 1} @modules;
63 sort { lc $a cmp lc $b }
64 grep {!/^(?:Internal|Unicode)$/o} keys %Encoding;
68 exists $INC{"PerlIO/encoding.pm"} or return 0;
69 my $stash = ref($_[0]);
70 $stash ||= ref(find_encoding($_[0]));
71 return ($stash eq "Encode::XS" || $stash eq "Encode::Unicode");
78 $Encoding{$name} = $obj;
80 define_alias($lc => $obj) unless $lc eq $name;
84 define_alias($alias,$obj);
91 my ($class,$name,$skip_external) = @_;
93 if (ref($name) && $name->can('new_sequence'))
98 if (exists $Encoding{$name})
100 return $Encoding{$name};
102 if (exists $Encoding{$lc})
104 return $Encoding{$lc};
107 my $oc = $class->find_alias($name);
108 return $oc if defined $oc;
110 $oc = $class->find_alias($lc) if $lc ne $name;
111 return $oc if defined $oc;
113 unless ($skip_external)
115 if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){
116 $mod =~ s,::,/,g ; $mod .= '.pm';
117 eval{ require $mod; };
118 return $Encoding{$name} if exists $Encoding{$name};
126 my ($name,$skip_external) = @_;
127 return __PACKAGE__->getEncoding($name,$skip_external);
131 my $obj = find_encoding(shift);
132 defined $obj and return $obj->name;
138 my ($name,$string,$check) = @_;
140 my $enc = find_encoding($name);
141 croak("Unknown encoding '$name'") unless defined $enc;
142 my $octets = $enc->encode($string,$check);
143 return undef if ($check && length($string));
149 my ($name,$octets,$check) = @_;
151 my $enc = find_encoding($name);
152 croak("Unknown encoding '$name'") unless defined $enc;
153 my $string = $enc->decode($octets,$check);
154 $_[1] = $octets if $check;
160 my ($string,$from,$to,$check) = @_;
162 my $f = find_encoding($from);
163 croak("Unknown encoding '$from'") unless defined $f;
164 my $t = find_encoding($to);
165 croak("Unknown encoding '$to'") unless defined $t;
166 my $uni = $f->decode($string,$check);
167 return undef if ($check && length($string));
168 $string = $t->encode($uni,$check);
169 return undef if ($check && length($uni));
170 return defined($_[0] = $string) ? length($string) : undef ;
183 return undef unless utf8::decode($str);
187 predefine_encodings();
190 # This is to restore %Encoding if really needed;
192 sub predefine_encodings{
194 # was in Encode::UTF_EBCDIC
195 package Encode::UTF_EBCDIC;
196 *name = sub{ shift->{'Name'} };
197 *new_sequence = sub{ return $_[0] };
199 my ($obj,$str,$chk) = @_;
201 for (my $i = 0; $i < length($str); $i++) {
203 chr(utf8::unicode_to_native(ord(substr($str,$i,1))));
209 my ($obj,$str,$chk) = @_;
211 for (my $i = 0; $i < length($str); $i++) {
213 chr(utf8::native_to_unicode(ord(substr($str,$i,1))));
218 $Encode::Encoding{Unicode} =
219 bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC";
221 # was in Encode::UTF_EBCDIC
222 package Encode::Internal;
223 *name = sub{ shift->{'Name'} };
224 *new_sequence = sub{ return $_[0] };
226 my ($obj,$str,$chk) = @_;
232 $Encode::Encoding{Unicode} =
233 bless {Name => "Internal"} => "Encode::Internal";
237 # was in Encode::utf8
238 package Encode::utf8;
239 *name = sub{ shift->{'Name'} };
240 *new_sequence = sub{ return $_[0] };
242 my ($obj,$octets,$chk) = @_;
243 my $str = Encode::decode_utf8($octets);
251 my ($obj,$string,$chk) = @_;
252 my $octets = Encode::encode_utf8($string);
256 $Encode::Encoding{utf8} =
257 bless {Name => "utf8"} => "Encode::utf8";
261 require Encode::Encoding;
262 @Encode::XS::ISA = qw(Encode::Encoding);
264 # This is very dodgy - PerlIO::encoding does "use Encode" and _BEFORE_ it gets a
265 # chance to set its VERSION we potentially delete it from %INC so it will be re-loaded
268 require PerlIO::encoding;
269 unless (PerlIO::encoding->VERSION >= 0.02){
270 delete $INC{"PerlIO/encoding.pm"};
281 Encode - character encodings
288 =head2 Table of Contents
290 Encode consists of a collection of modules which details are too big
291 to fit in one document. This POD itself explains the top-level APIs
292 and general topics at a glance. For other topics and more details,
296 --------------------------------------------------------
297 Encode::Alias Alias definitions to encodings
298 Encode::Encoding Encode Implementation Base Class
299 Encode::Supported List of Supported Encodings
300 Encode::CN Simplified Chinese Encodings
301 Encode::JP Japanese Encodings
302 Encode::KR Korean Encodings
303 Encode::TW Traditional Chinese Encodings
304 --------------------------------------------------------
308 The C<Encode> module provides the interfaces between Perl's strings
309 and the rest of the system. Perl strings are sequences of
312 The repertoire of characters that Perl can represent is at least that
313 defined by the Unicode Consortium. On most platforms the ordinal
314 values of the characters (as returned by C<ord(ch)>) is the "Unicode
315 codepoint" for the character (the exceptions are those platforms where
316 the legacy encoding is some variant of EBCDIC rather than a super-set
317 of ASCII - see L<perlebcdic>).
319 Traditionally computer data has been moved around in 8-bit chunks
320 often called "bytes". These chunks are also known as "octets" in
321 networking standards. Perl is widely used to manipulate data of many
322 types - not only strings of characters representing human or computer
323 languages but also "binary" data being the machines representation of
324 numbers, pixels in an image - or just about anything.
326 When Perl is processing "binary data" the programmer wants Perl to
327 process "sequences of bytes". This is not a problem for Perl - as a
328 byte has 256 possible values it easily fits in Perl's much larger
337 I<character>: a character in the range 0..(2**32-1) (or more).
338 (What Perl's strings are made of.)
342 I<byte>: a character in the range 0..255
343 (A special case of a Perl character.)
347 I<octet>: 8 bits of data, with ordinal values 0..255
348 (Term for bytes passed to or from a non-Perl context, e.g. disk file.)
352 The marker [INTERNAL] marks Internal Implementation Details, in
353 general meant only for those who think they know what they are doing,
354 and such details may change in future releases.
356 =head1 PERL ENCODING API
360 =item $octets = encode(ENCODING, $string[, CHECK])
362 Encodes string from Perl's internal form into I<ENCODING> and returns
363 a sequence of octets. ENCODING can be either a canonical name or
364 alias. For encoding names and aliases, see L</"Defining Aliases">.
365 For CHECK see L</"Handling Malformed Data">.
367 For example to convert (internally UTF-8 encoded) Unicode string to
368 iso-8859-1 (also known as Latin1),
370 $octets = encode("iso-8859-1", $unicode);
372 =item $string = decode(ENCODING, $octets[, CHECK])
374 Decode sequence of octets assumed to be in I<ENCODING> into Perl's
375 internal form and returns the resulting string. as in encode(),
376 ENCODING can be either a canonical name or alias. For encoding names
377 and aliases, see L</"Defining Aliases">. For CHECK see
378 L</"Handling Malformed Data">.
380 For example to convert ISO-8859-1 data to UTF-8:
382 $utf8 = decode("iso-8859-1", $latin1);
384 =item [$length =] from_to($string, FROM_ENCODING, TO_ENCODING [,CHECK])
386 Convert B<in-place> the data between two encodings.
387 For example to convert ISO-8859-1 data to UTF-8:
389 from_to($data, "iso-8859-1", "utf-8");
391 and to convert it back:
393 from_to($data, "utf-8", "iso-8859-1");
395 Note that because the conversion happens in place, the data to be
396 converted cannot be a string constant, it must be a scalar variable.
398 from_to() return the length of the converted string on success, undef
405 The Unicode consortium defines the UTF-8 standard as a way of encoding
406 the entire Unicode repertoire as sequences of octets. This encoding is
407 expected to become very widespread. Perl can use this form internally
408 to represent strings, so conversions to and from this form are
409 particularly efficient (as octets in memory do not have to change,
410 just the meta-data that tells Perl how to treat them).
414 =item $octets = encode_utf8($string);
416 The characters that comprise string are encoded in Perl's superset of UTF-8
417 and the resulting octets returned as a sequence of bytes. All possible
418 characters have a UTF-8 representation so this function cannot fail.
420 =item $string = decode_utf8($octets [, CHECK]);
422 The sequence of octets represented by $octets is decoded from UTF-8
423 into a sequence of logical characters. Not all sequences of octets
424 form valid UTF-8 encodings, so it is possible for this call to fail.
425 For CHECK see L</"Handling Malformed Data">.
429 =head2 Listing available encodings
432 @list = Encode->encodings();
434 Returns a list of the canonical names of the available encodings that
435 are loaded. To get a list of all available encodings including the
436 ones that are not loaded yet, say
438 @all_encodings = Encode->encodings(":all");
440 Or you can give the name of specific module.
442 @with_jp = Encode->encodings("Encode::JP");
444 When "::" is not in the name, "Encode::" is assumed.
446 @ebcdic = Encode->encodings("EBCDIC");
448 To find which encodings are supported by this package in details,
449 see L<Encode::Supported>.
451 =head2 Defining Aliases
453 To add new alias to a given encoding, Use;
457 define_alias(newName => ENCODING);
459 After that, newName can be used as an alias for ENCODING.
460 ENCODING may be either the name of an encoding or an
463 But before you do so, make sure the alias is nonexistent with
464 C<resolve_alias()>, which returns the canonical name thereof.
467 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
468 Encode::resolve_alias("iso-8859-12") # false; nonexistent
469 Encode::resolve_alias($name) eq $name # true if $name is canonical
471 This resolve_alias() does not need C<use Encode::Alias> and is
472 exported via C<use encode qw(resolve_alias)>.
474 See L<Encode::Alias> on details.
476 =head1 Encoding via PerlIO
478 If your perl supports I<PerlIO>, you can use PerlIO layer to directly
479 decode and encode via filehandle. The following two examples are
480 totally identical by functionality.
483 open my $in, "<:encoding(shiftjis)", $infile or die;
484 open my $out, ">:encoding(euc-jp)", $outfile or die;
488 open my $in, $infile or die;
489 open my $out, $outfile or die;
491 from_to($_, "shiftjis", "euc", 1);
494 Unfortunately, not all encodings are PerlIO-savvy. You can check if
495 your encoding is supported by PerlIO by C<perlio_ok> method.
497 Encode::perlio_ok("iso-20220jp"); # false
498 find_encoding("iso-2022-jp")->perlio_ok; # false
499 use Encode qw(perlio_ok); # exported upon request
500 perlio_ok("euc-jp") # true if PerlIO is enabled
502 For gory details, see L<Encode::PerlIO>;
504 =head1 Handling Malformed Data
508 THE I<CHECK> argument is used as follows. When you omit it, it is
509 identical to I<CHECK> = 0.
511 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
513 If I<CHECK> is 0, (en|de)code will put I<substitution character> in
514 place of the malformed character. for UCM-based encodings,
515 E<lt>subcharE<gt> will be used. For Unicode, \xFFFD is used. If the
516 data is supposed to be UTF-8, an optional lexical warning (category
519 =item I<CHECK> = Encode::DIE_ON_ERROR (== 1)
521 If I<CHECK> is 1, methods will die immediately with an error
522 message. so when I<CHECK> is set, you should trap the fatal error
523 with eval{} unless you really want to let it die on error.
525 =item I<CHECK> = Encode::FB_QUIET
527 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
528 return processed part on error, with data passed via argument
529 overwritten with unprocessed part. This is handy when have to
530 repeatedly call because the source data is chopped in the middle for
531 some reasons, such as fixed-width buffer. Here is a sample code that
535 while(defined(read $fh, $buffer, 256)){
536 # buffer may end in partial character so we append
538 $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET);
539 # $data now contains unprocessed partial character
542 =item I<CHECK> = Encode::FB_WARN
544 This is the same as above, except it warns on error. Handy when you
545 are debugging the mode above.
547 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
549 For encodings that are implemented by Encode::XS, CHECK ==
550 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
552 When you decode, '\xI<XX>' will be placed where I<XX> is the hex
553 representation of the octet that could not be decoded to utf8. And
554 when you encode, '\x{I<xxxx>}' will be placed where I<xxxx> is the
555 Unicode ID of the character that cannot be found in the character
556 repertoire of the encoding.
560 These modes are actually set via bitmask. here is how FB_XX are laid
561 out. for FB_XX you can import via C<use Encode qw(:fallbacks)> for
562 generic bitmask constants, you can import via
563 C<use Encode qw(:fallback_all)>.
565 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
568 RETURN_ON_ERR 0x0004 X X
572 =head2 Unemplemented fallback schemes
574 In future you will be able to use a code reference to a callback
575 function for the value of I<CHECK> but its API is still undecided.
577 =head1 Defining Encodings
579 To define a new encoding, use:
581 use Encode qw(define_alias);
582 define_encoding($object, 'canonicalName' [, alias...]);
584 I<canonicalName> will be associated with I<$object>. The object
585 should provide the interface described in L<Encode::Encoding>
586 If more than two arguments are provided then additional
587 arguments are taken as aliases for I<$object> as for C<define_alias>.
589 See L<Encode::Encoding> for more details.
591 =head1 Messing with Perl's Internals
593 The following API uses parts of Perl's internals in the current
594 implementation. As such they are efficient, but may change.
598 =item is_utf8(STRING [, CHECK])
600 [INTERNAL] Test whether the UTF-8 flag is turned on in the STRING.
601 If CHECK is true, also checks the data in STRING for being well-formed
602 UTF-8. Returns true if successful, false otherwise.
604 =item _utf8_on(STRING)
606 [INTERNAL] Turn on the UTF-8 flag in STRING. The data in STRING is
607 B<not> checked for being well-formed UTF-8. Do not use unless you
608 B<know> that the STRING is well-formed UTF-8. Returns the previous
609 state of the UTF-8 flag (so please don't test the return value as
610 I<not> success or failure), or C<undef> if STRING is not a string.
612 =item _utf8_off(STRING)
614 [INTERNAL] Turn off the UTF-8 flag in STRING. Do not use frivolously.
615 Returns the previous state of the UTF-8 flag (so please don't test the
616 return value as I<not> success or failure), or C<undef> if STRING is
624 L<Encode::Supported>,
631 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
635 This project was originated by Nick Ing-Simmons and later maintained
636 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for full list
637 of people involved. For any questions, use
638 E<lt>perl-unicode@perl.orgE<gt> so others can share.