3 our $VERSION = do { my @r = (q$Revision: 1.56 $ =~ /\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 $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]);
70 $obj->can("perlio_ok") and return $obj->perlio_ok() unless $@;
71 return 0; # safety net
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";
267 Encode - character encodings
273 =head2 Table of Contents
275 Encode consists of a collection of modules whose details are too big
276 to fit in one document. This POD itself explains the top-level APIs
277 and general topics at a glance. For other topics and more details,
281 --------------------------------------------------------
282 Encode::Alias Alias definitions to encodings
283 Encode::Encoding Encode Implementation Base Class
284 Encode::Supported List of Supported Encodings
285 Encode::CN Simplified Chinese Encodings
286 Encode::JP Japanese Encodings
287 Encode::KR Korean Encodings
288 Encode::TW Traditional Chinese Encodings
289 --------------------------------------------------------
293 The C<Encode> module provides the interfaces between Perl's strings
294 and the rest of the system. Perl strings are sequences of
297 The repertoire of characters that Perl can represent is at least that
298 defined by the Unicode Consortium. On most platforms the ordinal
299 values of the characters (as returned by C<ord(ch)>) is the "Unicode
300 codepoint" for the character (the exceptions are those platforms where
301 the legacy encoding is some variant of EBCDIC rather than a super-set
302 of ASCII - see L<perlebcdic>).
304 Traditionally, computer data has been moved around in 8-bit chunks
305 often called "bytes". These chunks are also known as "octets" in
306 networking standards. Perl is widely used to manipulate data of many
307 types - not only strings of characters representing human or computer
308 languages but also "binary" data being the machine's representation of
309 numbers, pixels in an image - or just about anything.
311 When Perl is processing "binary data", the programmer wants Perl to
312 process "sequences of bytes". This is not a problem for Perl - as a
313 byte has 256 possible values, it easily fits in Perl's much larger
322 I<character>: a character in the range 0..(2**32-1) (or more).
323 (What Perl's strings are made of.)
327 I<byte>: a character in the range 0..255
328 (A special case of a Perl character.)
332 I<octet>: 8 bits of data, with ordinal values 0..255
333 (Term for bytes passed to or from a non-Perl context, e.g. a disk file.)
337 The marker [INTERNAL] marks Internal Implementation Details, in
338 general meant only for those who think they know what they are doing,
339 and such details may change in future releases.
341 =head1 PERL ENCODING API
345 =item $octets = encode(ENCODING, $string[, CHECK])
347 Encodes a string from Perl's internal form into I<ENCODING> and returns
348 a sequence of octets. ENCODING can be either a canonical name or
349 an alias. For encoding names and aliases, see L</"Defining Aliases">.
350 For CHECK, see L</"Handling Malformed Data">.
352 For example, to convert (internally UTF-8 encoded) Unicode string to
353 iso-8859-1 (also known as Latin1),
355 $octets = encode("iso-8859-1", $unicode);
357 =item $string = decode(ENCODING, $octets[, CHECK])
359 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
360 internal form and returns the resulting string. As in encode(),
361 ENCODING can be either a canonical name or an alias. For encoding names
362 and aliases, see L</"Defining Aliases">. For CHECK, see
363 L</"Handling Malformed Data">.
365 For example, to convert ISO-8859-1 data to UTF-8:
367 $utf8 = decode("iso-8859-1", $latin1);
369 =item [$length =] from_to($string, FROM_ENCODING, TO_ENCODING [,CHECK])
371 Converts B<in-place> data between two encodings.
372 For example, to convert ISO-8859-1 data to UTF-8:
374 from_to($data, "iso-8859-1", "utf-8");
376 and to convert it back:
378 from_to($data, "utf-8", "iso-8859-1");
380 Note that because the conversion happens in place, the data to be
381 converted cannot be a string constant; it must be a scalar variable.
383 from_to() returns the length of the converted string on success, undef
390 The Unicode Consortium defines the UTF-8 transformation format as a
391 way of encoding the entire Unicode repertoire as sequences of octets.
392 This encoding is expected to become very widespread. Perl can use this
393 form internally to represent strings, so conversions to and from this
394 form are particularly efficient (as octets in memory do not have to
395 change, just the meta-data that tells Perl how to treat them).
399 =item $octets = encode_utf8($string);
401 The characters that comprise $string are encoded in Perl's superset of
402 UTF-8 and the resulting octets are returned as a sequence of bytes. All
403 possible characters have a UTF-8 representation so this function cannot
406 =item $string = decode_utf8($octets [, CHECK]);
408 The sequence of octets represented by $octets is decoded from UTF-8
409 into a sequence of logical characters. Not all sequences of octets
410 form valid UTF-8 encodings, so it is possible for this call to fail.
411 For CHECK, see L</"Handling Malformed Data">.
415 =head2 Listing available encodings
418 @list = Encode->encodings();
420 Returns a list of the canonical names of the available encodings that
421 are loaded. To get a list of all available encodings including the
422 ones that are not loaded yet, say
424 @all_encodings = Encode->encodings(":all");
426 Or you can give the name of a specific module.
428 @with_jp = Encode->encodings("Encode::JP");
430 When "::" is not in the name, "Encode::" is assumed.
432 @ebcdic = Encode->encodings("EBCDIC");
434 To find out in detail which encodings are supported by this package,
435 see L<Encode::Supported>.
437 =head2 Defining Aliases
439 To add a new alias to a given encoding, use:
443 define_alias(newName => ENCODING);
445 After that, newName can be used as an alias for ENCODING.
446 ENCODING may be either the name of an encoding or an
449 But before you do so, make sure the alias is nonexistent with
450 C<resolve_alias()>, which returns the canonical name thereof.
453 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
454 Encode::resolve_alias("iso-8859-12") # false; nonexistent
455 Encode::resolve_alias($name) eq $name # true if $name is canonical
457 resolve_alias() does not need C<use Encode::Alias>; it can be
458 exported via C<use Encode qw(resolve_alias)>.
460 See L<Encode::Alias> for details.
462 =head1 Encoding via PerlIO
464 If your perl supports I<PerlIO>, you can use a PerlIO layer to decode
465 and encode directly via a filehandle. The following two examples
466 are totally identical in their functionality.
469 open my $in, "<:encoding(shiftjis)", $infile or die;
470 open my $out, ">:encoding(euc-jp)", $outfile or die;
474 open my $in, "<", $infile or die;
475 open my $out, ">", $outfile or die;
477 from_to($_, "shiftjis", "euc-jp", 1);
480 Unfortunately, there may be encodings are PerlIO-savvy. You can check
481 if your encoding is supported by PerlIO by calling the C<perlio_ok>
484 Encode::perlio_ok("hz"); # False
485 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
487 use Encode qw(perlio_ok); # exported upon request
490 Fortunately, all encodings that come with Encode core are PerlIO-savvy
491 except for hz and ISO-2022-kr. See L<Encode::Encoding> for details.
493 For gory details, see L<Encode::PerlIO>.
495 =head1 Handling Malformed Data
499 The I<CHECK> argument is used as follows. When you omit it,
500 the behaviour is the same as if you had passed a value of 0 for
503 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
505 If I<CHECK> is 0, (en|de)code will put a I<substitution character>
506 in place of a malformed character. For UCM-based encodings,
507 E<lt>subcharE<gt> will be used. For Unicode, "\x{FFFD}" is used.
508 If the data is supposed to be UTF-8, an optional lexical warning
509 (category utf8) is given.
511 =item I<CHECK> = Encode::DIE_ON_ERROR (== 1)
513 If I<CHECK> is 1, methods will die immediately with an error
514 message. Therefore, when I<CHECK> is set to 1, you should trap the
515 fatal error with eval{} unless you really want to let it die on error.
517 =item I<CHECK> = Encode::FB_QUIET
519 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
520 return the portion of the data that has been processed so far when
521 an error occurs. The data argument will be overwritten with
522 everything after that point (that is, the unprocessed part of data).
523 This is handy when you have to call decode repeatedly in the case
524 where your source data may contain partial multi-byte character
525 sequences, for example because you are reading with a fixed-width
526 buffer. Here is some sample code that does exactly this:
529 while(defined(read $fh, $buffer, 256)){
530 # buffer may end in a partial character so we append
532 $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET);
533 # $data now contains the unprocessed partial character
536 =item I<CHECK> = Encode::FB_WARN
538 This is the same as above, except that it warns on error. Handy when
539 you are debugging the mode above.
541 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
543 For encodings that are implemented by Encode::XS, CHECK ==
544 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
546 When you decode, '\xI<XX>' will be inserted for a malformed character,
547 where I<XX> is the hex representation of the octet that could not be
548 decoded to utf8. And when you encode, '\x{I<xxxx>}' will be inserted,
549 where I<xxxx> is the Unicode ID of the character that cannot be found
550 in the character repertoire of the encoding.
554 These modes are actually set via a bitmask. Here is how the FB_XX
555 constants are laid out. You can import the FB_XX constants via
556 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
557 constants via C<use Encode qw(:fallback_all)>.
559 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
562 RETURN_ON_ERR 0x0004 X X
566 =head2 Unimplemented fallback schemes
568 In the future, you will be able to use a code reference to a callback
569 function for the value of I<CHECK> but its API is still undecided.
571 =head1 Defining Encodings
573 To define a new encoding, use:
575 use Encode qw(define_alias);
576 define_encoding($object, 'canonicalName' [, alias...]);
578 I<canonicalName> will be associated with I<$object>. The object
579 should provide the interface described in L<Encode::Encoding>.
580 If more than two arguments are provided then additional
581 arguments are taken as aliases for I<$object>, as for C<define_alias>.
583 See L<Encode::Encoding> for more details.
585 =head1 Messing with Perl's Internals
587 The following API uses parts of Perl's internals in the current
588 implementation. As such, they are efficient but may change.
592 =item is_utf8(STRING [, CHECK])
594 [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING.
595 If CHECK is true, also checks the data in STRING for being well-formed
596 UTF-8. Returns true if successful, false otherwise.
598 =item _utf8_on(STRING)
600 [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is
601 B<not> checked for being well-formed UTF-8. Do not use unless you
602 B<know> that the STRING is well-formed UTF-8. Returns the previous
603 state of the UTF-8 flag (so please don't treat the return value as
604 indicating success or failure), or C<undef> if STRING is not a string.
606 =item _utf8_off(STRING)
608 [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously.
609 Returns the previous state of the UTF-8 flag (so please don't treat the
610 return value as indicating success or failure), or C<undef> if STRING is
618 L<Encode::Supported>,
625 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
629 This project was originated by Nick Ing-Simmons and later maintained
630 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full list
631 of people involved. For any questions, use
632 E<lt>perl-unicode@perl.orgE<gt> so others can share.