3 our $VERSION = do { my @r = (q$Revision: 1.51 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
9 our @ISA = qw(Exporter DynaLoader);
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 ],
40 # Documentation moved after __END__ for speed - NI-S
44 our $ON_EBCDIC = (ord("A") == 193);
48 # Make a %Encoding package variable to allow a certain amount of cheating
51 require Encode::Config;
52 eval { require Encode::ConfigLocal };
57 my @modules = (@_ and $_[0] eq ":all") ? values %ExtModule : @_;
58 for my $mod (@modules){
59 $mod =~ s,::,/,g or $mod = "Encode/$mod";
61 $DEBUG and warn "about to require $mod;";
62 eval { require $mod; };
64 my %modules = map {$_ => 1} @modules;
66 sort { lc $a cmp lc $b }
67 grep {!/^(?:Internal|Unicode)$/o} keys %Encoding;
71 exists $INC{"PerlIO/encoding.pm"} or return 0;
72 my $stash = ref($_[0]);
73 $stash ||= ref(find_encoding($_[0]));
74 return ($stash eq "Encode::XS" || $stash eq "Encode::Unicode");
81 $Encoding{$name} = $obj;
83 define_alias($lc => $obj) unless $lc eq $name;
87 define_alias($alias,$obj);
94 my ($class,$name,$skip_external) = @_;
96 if (ref($name) && $name->can('new_sequence'))
101 if (exists $Encoding{$name})
103 return $Encoding{$name};
105 if (exists $Encoding{$lc})
107 return $Encoding{$lc};
110 my $oc = $class->find_alias($name);
111 return $oc if defined $oc;
113 $oc = $class->find_alias($lc) if $lc ne $name;
114 return $oc if defined $oc;
116 unless ($skip_external)
118 if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){
119 $mod =~ s,::,/,g ; $mod .= '.pm';
120 eval{ require $mod; };
121 return $Encoding{$name} if exists $Encoding{$name};
129 my ($name,$skip_external) = @_;
130 return __PACKAGE__->getEncoding($name,$skip_external);
134 my $obj = find_encoding(shift);
135 defined $obj and return $obj->name;
141 my ($name,$string,$check) = @_;
143 my $enc = find_encoding($name);
144 croak("Unknown encoding '$name'") unless defined $enc;
145 my $octets = $enc->encode($string,$check);
146 return undef if ($check && length($string));
152 my ($name,$octets,$check) = @_;
154 my $enc = find_encoding($name);
155 croak("Unknown encoding '$name'") unless defined $enc;
156 my $string = $enc->decode($octets,$check);
157 $_[1] = $octets if $check;
163 my ($string,$from,$to,$check) = @_;
165 my $f = find_encoding($from);
166 croak("Unknown encoding '$from'") unless defined $f;
167 my $t = find_encoding($to);
168 croak("Unknown encoding '$to'") unless defined $t;
169 my $uni = $f->decode($string,$check);
170 return undef if ($check && length($string));
171 $string = $t->encode($uni,$check);
172 return undef if ($check && length($uni));
173 return defined($_[0] = $string) ? length($string) : undef ;
186 return undef unless utf8::decode($str);
190 predefine_encodings();
193 # This is to restore %Encoding if really needed;
195 sub predefine_encodings{
197 # was in Encode::UTF_EBCDIC
198 package Encode::UTF_EBCDIC;
199 *name = sub{ shift->{'Name'} };
200 *new_sequence = sub{ return $_[0] };
202 my ($obj,$str,$chk) = @_;
204 for (my $i = 0; $i < length($str); $i++) {
206 chr(utf8::unicode_to_native(ord(substr($str,$i,1))));
212 my ($obj,$str,$chk) = @_;
214 for (my $i = 0; $i < length($str); $i++) {
216 chr(utf8::native_to_unicode(ord(substr($str,$i,1))));
221 $Encode::Encoding{Unicode} =
222 bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC";
224 # was in Encode::UTF_EBCDIC
225 package Encode::Internal;
226 *name = sub{ shift->{'Name'} };
227 *new_sequence = sub{ return $_[0] };
229 my ($obj,$str,$chk) = @_;
235 $Encode::Encoding{Unicode} =
236 bless {Name => "Internal"} => "Encode::Internal";
240 # was in Encode::utf8
241 package Encode::utf8;
242 *name = sub{ shift->{'Name'} };
243 *new_sequence = sub{ return $_[0] };
245 my ($obj,$octets,$chk) = @_;
246 my $str = Encode::decode_utf8($octets);
254 my ($obj,$string,$chk) = @_;
255 my $octets = Encode::encode_utf8($string);
259 $Encode::Encoding{utf8} =
260 bless {Name => "utf8"} => "Encode::utf8";
264 require Encode::Encoding;
267 require PerlIO::encoding;
268 unless (PerlIO::encoding->VERSION >= 0.02){
269 delete $INC{"PerlIO/encoding.pm"};
280 Encode - character encodings
287 =head2 Table of Contents
289 Encode consists of a collection of modules which details are too big
290 to fit in one document. This POD itself explains the top-level APIs
291 and general topics at a glance. For other topics and more details,
295 --------------------------------------------------------
296 Encode::Alias Alias defintions to encodings
297 Encode::Encoding Encode Implementation Base Class
298 Encode::Supported List of Supported Encodings
299 Encode::CN Simplified Chinese Encodings
300 Encode::JP Japanese Encodings
301 Encode::KR Korean Encodings
302 Encode::TW Traditional Chinese Encodings
303 --------------------------------------------------------
307 The C<Encode> module provides the interfaces between Perl's strings
308 and the rest of the system. Perl strings are sequences of
311 The repertoire of characters that Perl can represent is at least that
312 defined by the Unicode Consortium. On most platforms the ordinal
313 values of the characters (as returned by C<ord(ch)>) is the "Unicode
314 codepoint" for the character (the exceptions are those platforms where
315 the legacy encoding is some variant of EBCDIC rather than a super-set
316 of ASCII - see L<perlebcdic>).
318 Traditionally computer data has been moved around in 8-bit chunks
319 often called "bytes". These chunks are also known as "octets" in
320 networking standards. Perl is widely used to manipulate data of many
321 types - not only strings of characters representing human or computer
322 languages but also "binary" data being the machines representation of
323 numbers, pixels in an image - or just about anything.
325 When Perl is processing "binary data" the programmer wants Perl to
326 process "sequences of bytes". This is not a problem for Perl - as a
327 byte has 256 possible values it easily fits in Perl's much larger
336 I<character>: a character in the range 0..(2**32-1) (or more).
337 (What Perl's strings are made of.)
341 I<byte>: a character in the range 0..255
342 (A special case of a Perl character.)
346 I<octet>: 8 bits of data, with ordinal values 0..255
347 (Term for bytes passed to or from a non-Perl context, e.g. disk file.)
351 The marker [INTERNAL] marks Internal Implementation Details, in
352 general meant only for those who think they know what they are doing,
353 and such details may change in future releases.
355 =head1 PERL ENCODING API
359 =item $octets = encode(ENCODING, $string[, CHECK])
361 Encodes string from Perl's internal form into I<ENCODING> and returns
362 a sequence of octets. ENCODING can be either a canonical name or
363 alias. For encoding names and aliases, see L</"Defining Aliases">.
364 For CHECK see L</"Handling Malformed Data">.
366 For example to convert (internally UTF-8 encoded) Unicode string to
367 iso-8859-1 (also known as Latin1),
369 $octets = encode("iso-8859-1", $unicode);
371 =item $string = decode(ENCODING, $octets[, CHECK])
373 Decode sequence of octets assumed to be in I<ENCODING> into Perl's
374 internal form and returns the resulting string. as in encode(),
375 ENCODING can be either a canonical name or alias. For encoding names
376 and aliases, see L</"Defining Aliases">. For CHECK see
377 L</"Handling Malformed Data">.
379 For example to convert ISO-8859-1 data to UTF-8:
381 $utf8 = decode("iso-8859-1", $latin1);
383 =item [$length =] from_to($string, FROM_ENCODING, TO_ENCODING [,CHECK])
385 Convert B<in-place> the data between two encodings.
386 For example to convert ISO-8859-1 data to UTF-8:
388 from_to($data, "iso-8859-1", "utf-8");
390 and to convert it back:
392 from_to($data, "utf-8", "iso-8859-1");
394 Note that because the conversion happens in place, the data to be
395 converted cannot be a string constant, it must be a scalar variable.
397 from_to() return the length of the converted string on success, undef
404 The Unicode consortium defines the UTF-8 standard as a way of encoding
405 the entire Unicode repertoire as sequences of octets. This encoding is
406 expected to become very widespread. Perl can use this form internally
407 to represent strings, so conversions to and from this form are
408 particularly efficient (as octets in memory do not have to change,
409 just the meta-data that tells Perl how to treat them).
413 =item $octets = encode_utf8($string);
415 The characters that comprise string are encoded in Perl's superset of UTF-8
416 and the resulting octets returned as a sequence of bytes. All possible
417 characters have a UTF-8 representation so this function cannot fail.
419 =item $string = decode_utf8($octets [, CHECK]);
421 The sequence of octets represented by $octets is decoded from UTF-8
422 into a sequence of logical characters. Not all sequences of octets
423 form valid UTF-8 encodings, so it is possible for this call to fail.
424 For CHECK see L</"Handling Malformed Data">.
428 =head2 Listing available encodings
431 @list = Encode->encodings();
433 Returns a list of the canonical names of the available encodings that
434 are loaded. To get a list of all available encodings including the
435 ones that are not loaded yet, say
437 @all_encodings = Encode->encodings(":all");
439 Or you can give the name of specific module.
441 @with_jp = Encode->encodings("Encode::JP");
443 When "::" is not in the name, "Encode::" is assumed.
445 @ebcdic = Encode->encodings("EBCDIC");
447 To find which encodings are supported by this package in details,
448 see L<Encode::Supported>.
450 =head2 Defining Aliases
452 To add new alias to a given encoding, Use;
456 define_alias(newName => ENCODING);
458 After that, newName can be used as an alias for ENCODING.
459 ENCODING may be either the name of an encoding or an
462 But before you do so, make sure the alias is nonexistent with
463 C<resolve_alias()>, which returns the canonical name thereof.
466 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
467 Encode::resolve_alias("iso-8859-12") # false; nonexistent
468 Encode::resolve_alias($name) eq $name # true if $name is canonical
470 This resolve_alias() does not need C<use Encode::Alias> and is
471 exported via C<use encode qw(resolve_alias)>.
473 See L<Encode::Alias> on details.
475 =head1 Encoding via PerlIO
477 If your perl supports I<PerlIO>, you can use PerlIO layer to directly
478 decode and encode via filehandle. The following two examples are
479 totally identical by functionality.
482 open my $in, "<:encoding(shiftjis)", $infile or die;
483 open my $out, ">:encoding(euc-jp)", $outfile or die;
487 open my $in, $infile or die;
488 open my $out, $outfile or die;
490 from_to($_, "shiftjis", "euc", 1);
493 Unfortunately, not all encodings are PerlIO-savvy. You can check if
494 your encoding is supported by PerlIO by C<perlio_ok> method.
496 Encode::perlio_ok("iso-20220jp"); # false
497 find_encoding("iso-2022-jp")->perlio_ok; # false
498 use Encode qw(perlio_ok); # exported upon request
499 perlio_ok("euc-jp") # true if PerlIO is enabled
501 For gory details, see L<Encode::PerlIO>;
503 =head1 Handling Malformed Data
507 THE I<CHECK> argument is used as follows. When you omit it, it is
508 identical to I<CHECK> = 0.
510 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
512 If I<CHECK> is 0, (en|de)code will put I<substitution character> in
513 place of the malformed character. for UCM-based encodings,
514 E<lt>subcharE<gt> will be used. For Unicode, \xFFFD is used. If the
515 data is supposed to be UTF-8, an optional lexical warning (category
518 =item I<CHECK> = Encode::DIE_ON_ERROR (== 1)
520 If I<CHECK> is 1, methods will die immediately with an error
521 message. so when I<CHECK> is set, you should trap the fatal error
522 with eval{} unless you really want to let it die on error.
524 =item I<CHECK> = Encode::FB_QUIET
526 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
527 return proccessed part on error, with data passed via argument
528 overwritten with unproccessed part. This is handy when have to
529 repeatedly call because the source data is chopped in the middle for
530 some reasons, such as fixed-width buffer. Here is a sample code that
534 while(defined(read $fh, $buffer, 256)){
535 # buffer may end in partial character so we append
537 $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET);
538 # $data now contains unprocessed partial character
541 =item I<CHECK> = Encode::FB_WARN
543 This is the same as above, except it warns on error. Handy when you
544 are debugging the mode above.
546 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
548 For encodings that are implemented by Encode::XS, CHECK ==
549 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
551 When you decode, '\xI<XX>' will be placed where I<XX> is the hex
552 representation of the octet that could not be decoded to utf8. And
553 when you encode, '\x{I<xxxx>}' will be placed where I<xxxx> is the
554 Unicode ID of the character that cannot be found in the character
555 repartoire of the encoding.
559 These modes are actually set via bitmask. here is how FB_XX are laid
560 out. for FB_XX you can import via C<use Encode qw(:fallbacks)> for
561 generic bitmask constants, you can import via
562 C<use Encode qw(:fallback_all)>.
564 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
567 RETURN_ON_ERR 0x0004 X X
571 =head2 Unemplemented fallback schemes
573 In future you will be able to use a code reference to a callback
574 function for the value of I<CHECK> but its API is still undecided.
576 =head1 Defining Encodings
578 To define a new encoding, use:
580 use Encode qw(define_alias);
581 define_encoding($object, 'canonicalName' [, alias...]);
583 I<canonicalName> will be associated with I<$object>. The object
584 should provide the interface described in L<Encode::Encoding>
585 If more than two arguments are provided then additional
586 arguments are taken as aliases for I<$object> as for C<define_alias>.
588 See L<Encode::Encoding> for more details.
590 =head1 Messing with Perl's Internals
592 The following API uses parts of Perl's internals in the current
593 implementation. As such they are efficient, but may change.
597 =item is_utf8(STRING [, CHECK])
599 [INTERNAL] Test whether the UTF-8 flag is turned on in the STRING.
600 If CHECK is true, also checks the data in STRING for being well-formed
601 UTF-8. Returns true if successful, false otherwise.
603 =item _utf8_on(STRING)
605 [INTERNAL] Turn on the UTF-8 flag in STRING. The data in STRING is
606 B<not> checked for being well-formed UTF-8. Do not use unless you
607 B<know> that the STRING is well-formed UTF-8. Returns the previous
608 state of the UTF-8 flag (so please don't test the return value as
609 I<not> success or failure), or C<undef> if STRING is not a string.
611 =item _utf8_off(STRING)
613 [INTERNAL] Turn off the UTF-8 flag in STRING. Do not use frivolously.
614 Returns the previous state of the UTF-8 flag (so please don't test the
615 return value as I<not> success or failure), or C<undef> if STRING is
623 L<Encode::Supported>,
630 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
634 This project was originated by Nick Ing-Simmons and later maintained
635 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for full list
636 of people involved. For any questions, use
637 E<lt>perl-unicode@perl.orgE<gt> so others can share.