3 our $VERSION = do { my @r = (q$Revision: 1.60 $ =~ /\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
19 PERLQQ HTMLCREF XMLCREF);
20 our @FB_CONSTS = qw(FB_DEFAULT FB_CROAK FB_QUIET FB_WARN
21 FB_PERLQQ FB_HTMLCREF FB_XMLCREF);
26 _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit
27 is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade
29 @FB_FLAGS, @FB_CONSTS,
34 all => [ @EXPORT, @EXPORT_OK ],
35 fallbacks => [ @FB_CONSTS ],
36 fallback_all => [ @FB_CONSTS, @FB_FLAGS ],
39 # Documentation moved after __END__ for speed - NI-S
43 our $ON_EBCDIC = (ord("A") == 193);
47 # Make a %Encoding package variable to allow a certain amount of cheating
50 require Encode::Config;
51 eval { require Encode::ConfigLocal };
56 my @modules = (@_ and $_[0] eq ":all") ? values %ExtModule : @_;
57 for my $mod (@modules){
58 $mod =~ s,::,/,g or $mod = "Encode/$mod";
60 $DEBUG and warn "about to require $mod;";
61 eval { require $mod; };
63 my %modules = map {$_ => 1} @modules;
65 sort { lc $a cmp lc $b }
66 grep {!/^(?:Internal|Unicode)$/o} keys %Encoding;
70 my $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]);
71 $obj->can("perlio_ok") and return $obj->perlio_ok();
72 return 0; # safety net
79 $Encoding{$name} = $obj;
81 define_alias($lc => $obj) unless $lc eq $name;
85 define_alias($alias,$obj);
92 my ($class,$name,$skip_external) = @_;
94 if (ref($name) && $name->can('new_sequence'))
99 if (exists $Encoding{$name})
101 return $Encoding{$name};
103 if (exists $Encoding{$lc})
105 return $Encoding{$lc};
108 my $oc = $class->find_alias($name);
109 return $oc if defined $oc;
111 $oc = $class->find_alias($lc) if $lc ne $name;
112 return $oc if defined $oc;
114 unless ($skip_external)
116 if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){
117 $mod =~ s,::,/,g ; $mod .= '.pm';
118 eval{ require $mod; };
119 return $Encoding{$name} if exists $Encoding{$name};
127 my ($name,$skip_external) = @_;
128 return __PACKAGE__->getEncoding($name,$skip_external);
132 my $obj = find_encoding(shift);
133 defined $obj and return $obj->name;
139 my ($name,$string,$check) = @_;
141 my $enc = find_encoding($name);
142 croak("Unknown encoding '$name'") unless defined $enc;
143 my $octets = $enc->encode($string,$check);
144 return undef if ($check && length($string));
150 my ($name,$octets,$check) = @_;
152 my $enc = find_encoding($name);
153 croak("Unknown encoding '$name'") unless defined $enc;
154 my $string = $enc->decode($octets,$check);
155 $_[1] = $octets if $check;
161 my ($string,$from,$to,$check) = @_;
163 my $f = find_encoding($from);
164 croak("Unknown encoding '$from'") unless defined $f;
165 my $t = find_encoding($to);
166 croak("Unknown encoding '$to'") unless defined $t;
167 my $uni = $f->decode($string,$check);
168 return undef if ($check && length($string));
169 $string = $t->encode($uni,$check);
170 return undef if ($check && length($uni));
171 return defined($_[0] = $string) ? length($string) : undef ;
184 return undef unless utf8::decode($str);
188 predefine_encodings();
191 # This is to restore %Encoding if really needed;
193 sub predefine_encodings{
195 # was in Encode::UTF_EBCDIC
196 package Encode::UTF_EBCDIC;
197 *name = sub{ shift->{'Name'} };
198 *new_sequence = sub{ return $_[0] };
199 *needs_lines = sub{ 0 };
201 eval{ require PerlIO::encoding };
205 my ($obj,$str,$chk) = @_;
207 for (my $i = 0; $i < length($str); $i++) {
209 chr(utf8::unicode_to_native(ord(substr($str,$i,1))));
215 my ($obj,$str,$chk) = @_;
217 for (my $i = 0; $i < length($str); $i++) {
219 chr(utf8::native_to_unicode(ord(substr($str,$i,1))));
224 $Encode::Encoding{Unicode} =
225 bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC";
227 # was in Encode::UTF_EBCDIC
228 package Encode::Internal;
229 *name = sub{ shift->{'Name'} };
230 *new_sequence = sub{ return $_[0] };
231 *needs_lines = sub{ 0 };
233 eval{ require PerlIO::encoding };
237 my ($obj,$str,$chk) = @_;
243 $Encode::Encoding{Unicode} =
244 bless {Name => "Internal"} => "Encode::Internal";
248 # was in Encode::utf8
249 package Encode::utf8;
250 *name = sub{ shift->{'Name'} };
251 *new_sequence = sub{ return $_[0] };
252 *needs_lines = sub{ 0 };
254 eval{ require PerlIO::encoding };
258 my ($obj,$octets,$chk) = @_;
259 my $str = Encode::decode_utf8($octets);
267 my ($obj,$string,$chk) = @_;
268 my $octets = Encode::encode_utf8($string);
272 $Encode::Encoding{utf8} =
273 bless {Name => "utf8"} => "Encode::utf8";
283 Encode - character encodings
289 =head2 Table of Contents
291 Encode consists of a collection of modules whose details are too big
292 to fit in one document. This POD itself explains the top-level APIs
293 and general topics at a glance. For other topics and more details,
297 --------------------------------------------------------
298 Encode::Alias Alias definitions to encodings
299 Encode::Encoding Encode Implementation Base Class
300 Encode::Supported List of Supported Encodings
301 Encode::CN Simplified Chinese Encodings
302 Encode::JP Japanese Encodings
303 Encode::KR Korean Encodings
304 Encode::TW Traditional Chinese Encodings
305 --------------------------------------------------------
309 The C<Encode> module provides the interfaces between Perl's strings
310 and the rest of the system. Perl strings are sequences of
313 The repertoire of characters that Perl can represent is at least that
314 defined by the Unicode Consortium. On most platforms the ordinal
315 values of the characters (as returned by C<ord(ch)>) is the "Unicode
316 codepoint" for the character (the exceptions are those platforms where
317 the legacy encoding is some variant of EBCDIC rather than a super-set
318 of ASCII - see L<perlebcdic>).
320 Traditionally, computer data has been moved around in 8-bit chunks
321 often called "bytes". These chunks are also known as "octets" in
322 networking standards. Perl is widely used to manipulate data of many
323 types - not only strings of characters representing human or computer
324 languages but also "binary" data being the machine's representation of
325 numbers, pixels in an image - or just about anything.
327 When Perl is processing "binary data", the programmer wants Perl to
328 process "sequences of bytes". This is not a problem for Perl - as a
329 byte has 256 possible values, it easily fits in Perl's much larger
338 I<character>: a character in the range 0..(2**32-1) (or more).
339 (What Perl's strings are made of.)
343 I<byte>: a character in the range 0..255
344 (A special case of a Perl character.)
348 I<octet>: 8 bits of data, with ordinal values 0..255
349 (Term for bytes passed to or from a non-Perl context, e.g. a disk file.)
353 The marker [INTERNAL] marks Internal Implementation Details, in
354 general meant only for those who think they know what they are doing,
355 and such details may change in future releases.
357 =head1 PERL ENCODING API
361 =item $octets = encode(ENCODING, $string[, CHECK])
363 Encodes a string from Perl's internal form into I<ENCODING> and returns
364 a sequence of octets. ENCODING can be either a canonical name or
365 an alias. For encoding names and aliases, see L</"Defining Aliases">.
366 For CHECK, see L</"Handling Malformed Data">.
368 For example, to convert (internally UTF-8 encoded) Unicode string to
369 iso-8859-1 (also known as Latin1),
371 $octets = encode("iso-8859-1", $unicode);
373 =item $string = decode(ENCODING, $octets[, CHECK])
375 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
376 internal form and returns the resulting string. As in encode(),
377 ENCODING can be either a canonical name or an alias. For encoding names
378 and aliases, see L</"Defining Aliases">. For CHECK, see
379 L</"Handling Malformed Data">.
381 For example, to convert ISO-8859-1 data to UTF-8:
383 $utf8 = decode("iso-8859-1", $latin1);
385 =item [$length =] from_to($string, FROM_ENCODING, TO_ENCODING [,CHECK])
387 Converts B<in-place> data between two encodings.
388 For example, to convert ISO-8859-1 data to UTF-8:
390 from_to($data, "iso-8859-1", "utf-8");
392 and to convert it back:
394 from_to($data, "utf-8", "iso-8859-1");
396 Note that because the conversion happens in place, the data to be
397 converted cannot be a string constant; it must be a scalar variable.
399 from_to() returns the length of the converted string on success, undef
406 The Unicode Consortium defines the UTF-8 transformation format as a
407 way of encoding the entire Unicode repertoire as sequences of octets.
408 This encoding is expected to become very widespread. Perl can use this
409 form internally to represent strings, so conversions to and from this
410 form are particularly efficient (as octets in memory do not have to
411 change, just the meta-data that tells Perl how to treat them).
415 =item $octets = encode_utf8($string);
417 The characters that comprise $string are encoded in Perl's superset of
418 UTF-8 and the resulting octets are returned as a sequence of bytes. All
419 possible characters have a UTF-8 representation so this function cannot
422 =item $string = decode_utf8($octets [, CHECK]);
424 The sequence of octets represented by $octets is decoded from UTF-8
425 into a sequence of logical characters. Not all sequences of octets
426 form valid UTF-8 encodings, so it is possible for this call to fail.
427 For CHECK, see L</"Handling Malformed Data">.
431 =head2 Listing available encodings
434 @list = Encode->encodings();
436 Returns a list of the canonical names of the available encodings that
437 are loaded. To get a list of all available encodings including the
438 ones that are not loaded yet, say
440 @all_encodings = Encode->encodings(":all");
442 Or you can give the name of a specific module.
444 @with_jp = Encode->encodings("Encode::JP");
446 When "::" is not in the name, "Encode::" is assumed.
448 @ebcdic = Encode->encodings("EBCDIC");
450 To find out in detail which encodings are supported by this package,
451 see L<Encode::Supported>.
453 =head2 Defining Aliases
455 To add a new alias to a given encoding, use:
459 define_alias(newName => ENCODING);
461 After that, newName can be used as an alias for ENCODING.
462 ENCODING may be either the name of an encoding or an
465 But before you do so, make sure the alias is nonexistent with
466 C<resolve_alias()>, which returns the canonical name thereof.
469 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
470 Encode::resolve_alias("iso-8859-12") # false; nonexistent
471 Encode::resolve_alias($name) eq $name # true if $name is canonical
473 resolve_alias() does not need C<use Encode::Alias>; it can be
474 exported via C<use Encode qw(resolve_alias)>.
476 See L<Encode::Alias> for details.
478 =head1 Encoding via PerlIO
480 If your perl supports I<PerlIO>, you can use a PerlIO layer to decode
481 and encode directly via a filehandle. The following two examples
482 are totally identical in their functionality.
485 open my $in, "<:encoding(shiftjis)", $infile or die;
486 open my $out, ">:encoding(euc-jp)", $outfile or die;
490 open my $in, "<", $infile or die;
491 open my $out, ">", $outfile or die;
493 from_to($_, "shiftjis", "euc-jp", 1);
496 Unfortunately, there may be encodings are PerlIO-savvy. You can check
497 if your encoding is supported by PerlIO by calling the C<perlio_ok>
500 Encode::perlio_ok("hz"); # False
501 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
503 use Encode qw(perlio_ok); # exported upon request
506 Fortunately, all encodings that come with Encode core are PerlIO-savvy
507 except for hz and ISO-2022-kr. See L<Encode::Encoding> for details.
509 For gory details, see L<Encode::PerlIO>.
511 =head1 Handling Malformed Data
515 The I<CHECK> argument is used as follows. When you omit it,
516 the behaviour is the same as if you had passed a value of 0 for
519 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
521 If I<CHECK> is 0, (en|de)code will put a I<substitution character>
522 in place of a malformed character. For UCM-based encodings,
523 E<lt>subcharE<gt> will be used. For Unicode, "\x{FFFD}" is used.
524 If the data is supposed to be UTF-8, an optional lexical warning
525 (category utf8) is given.
527 =item I<CHECK> = Encode::DIE_ON_ERROR (== 1)
529 If I<CHECK> is 1, methods will die immediately with an error
530 message. Therefore, when I<CHECK> is set to 1, you should trap the
531 fatal error with eval{} unless you really want to let it die on error.
533 =item I<CHECK> = Encode::FB_QUIET
535 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
536 return the portion of the data that has been processed so far when
537 an error occurs. The data argument will be overwritten with
538 everything after that point (that is, the unprocessed part of data).
539 This is handy when you have to call decode repeatedly in the case
540 where your source data may contain partial multi-byte character
541 sequences, for example because you are reading with a fixed-width
542 buffer. Here is some sample code that does exactly this:
545 while(defined(read $fh, $buffer, 256)){
546 # buffer may end in a partial character so we append
548 $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET);
549 # $data now contains the unprocessed partial character
552 =item I<CHECK> = Encode::FB_WARN
554 This is the same as above, except that it warns on error. Handy when
555 you are debugging the mode above.
557 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
559 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF)
561 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF)
563 For encodings that are implemented by Encode::XS, CHECK ==
564 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
566 When you decode, '\xI<XX>' will be inserted for a malformed character,
567 where I<XX> is the hex representation of the octet that could not be
568 decoded to utf8. And when you encode, '\x{I<xxxx>}' will be inserted,
569 where I<xxxx> is the Unicode ID of the character that cannot be found
570 in the character repertoire of the encoding.
572 HTML/XML character reference modes are about the same, in place of
573 \x{I<xxxx>}, HTML uses &#I<1234>; where I<1234> is a decimal digit and
574 XML uses &#xI<abcd>; where I<abcd> is the hexadecimal digit.
578 These modes are actually set via a bitmask. Here is how the FB_XX
579 constants are laid out. You can import the FB_XX constants via
580 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
581 constants via C<use Encode qw(:fallback_all)>.
583 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
586 RETURN_ON_ERR 0x0004 X X
592 =head2 Unimplemented fallback schemes
594 In the future, you will be able to use a code reference to a callback
595 function for the value of I<CHECK> but its API is still undecided.
597 =head1 Defining Encodings
599 To define a new encoding, use:
601 use Encode qw(define_alias);
602 define_encoding($object, 'canonicalName' [, alias...]);
604 I<canonicalName> will be associated with I<$object>. The object
605 should provide the interface described in L<Encode::Encoding>.
606 If more than two arguments are provided then additional
607 arguments are taken as aliases for I<$object>, as for C<define_alias>.
609 See L<Encode::Encoding> for more details.
611 =head1 Messing with Perl's Internals
613 The following API uses parts of Perl's internals in the current
614 implementation. As such, they are efficient but may change.
618 =item is_utf8(STRING [, CHECK])
620 [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING.
621 If CHECK is true, also checks the data in STRING for being well-formed
622 UTF-8. Returns true if successful, false otherwise.
624 =item _utf8_on(STRING)
626 [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is
627 B<not> checked for being well-formed UTF-8. Do not use unless you
628 B<know> that the STRING is well-formed UTF-8. Returns the previous
629 state of the UTF-8 flag (so please don't treat the return value as
630 indicating success or failure), or C<undef> if STRING is not a string.
632 =item _utf8_off(STRING)
634 [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously.
635 Returns the previous state of the UTF-8 flag (so please don't treat the
636 return value as indicating success or failure), or C<undef> if STRING is
644 L<Encode::Supported>,
651 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
655 This project was originated by Nick Ing-Simmons and later maintained
656 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full list
657 of people involved. For any questions, use
658 E<lt>perl-unicode@perl.orgE<gt> so others can share.