3 our $VERSION = do { my @r = (q$Revision: 1.61 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
6 XSLoader::load 'Encode';
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", $utf8);
373 B<CAVEAT>: When you C<$octets = encode("utf8", $utf8)>, then $octets
374 B<ne> $utf8. Though they both contain the same data, the utf8 flag
375 for $octets is B<always> off. When you encode anything, utf8 flag of
376 the result is always off, even when it contains completely valid utf8
377 string. See L</"The UTF-8 flag"> below.
379 =item $string = decode(ENCODING, $octets[, CHECK])
381 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
382 internal form and returns the resulting string. As in encode(),
383 ENCODING can be either a canonical name or an alias. For encoding names
384 and aliases, see L</"Defining Aliases">. For CHECK, see
385 L</"Handling Malformed Data">.
387 For example, to convert ISO-8859-1 data to UTF-8:
389 $utf8 = decode("iso-8859-1", $latin1);
391 B<CAVEAT>: When you C<$utf8 = encode("utf8", $octets)>, then $utf8
392 B<may not be equal to> $utf8. Though they both contain the same data,
393 the utf8 flag for $utf8 is on unless $octets entirely conststs of
394 ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag">
397 =item [$length =] from_to($string, FROM_ENC, TO_ENC [, CHECK])
399 Converts B<in-place> data between two encodings. For example, to
400 convert ISO-8859-1 data to UTF-8:
402 from_to($data, "iso-8859-1", "utf8");
404 and to convert it back:
406 from_to($data, "utf8", "iso-8859-1");
408 Note that because the conversion happens in place, the data to be
409 converted cannot be a string constant; it must be a scalar variable.
411 from_to() returns the length of the converted string on success, undef
414 B<CAVEAT>: The following operations look the same but not quite so;
416 from_to($data, "iso-8859-1", "utf8"); #1
417 $data = decode("iso-8859-1", $data); #2
419 Both #1 and #2 makes $data consists of completely valid UTF-8 string
420 but only #2 turns utf8 flag on. #1 is equivalent to
422 $data = encode("utf8", decode("iso-8859-1", $data));
424 See L</"The UTF-8 flag"> below.
426 =item $octets = encode_utf8($string);
428 Equivalent to C<$octets = encode("utf8", $string);> The characters
429 that comprise $string are encoded in Perl's superset of UTF-8 and the
430 resulting octets are returned as a sequence of bytes. All possible
431 characters have a UTF-8 representation so this function cannot fail.
434 =item $string = decode_utf8($octets [, CHECK]);
436 equivalent to C<$string = decode("utf8", $octets [, CHECK])>.
437 decode_utf8($octets [, CHECK]); The sequence of octets represented by
438 $octets is decoded from UTF-8 into a sequence of logical
439 characters. Not all sequences of octets form valid UTF-8 encodings, so
440 it is possible for this call to fail. For CHECK, see
441 L</"Handling Malformed Data">.
445 =head2 Listing available encodings
448 @list = Encode->encodings();
450 Returns a list of the canonical names of the available encodings that
451 are loaded. To get a list of all available encodings including the
452 ones that are not loaded yet, say
454 @all_encodings = Encode->encodings(":all");
456 Or you can give the name of a specific module.
458 @with_jp = Encode->encodings("Encode::JP");
460 When "::" is not in the name, "Encode::" is assumed.
462 @ebcdic = Encode->encodings("EBCDIC");
464 To find out in detail which encodings are supported by this package,
465 see L<Encode::Supported>.
467 =head2 Defining Aliases
469 To add a new alias to a given encoding, use:
473 define_alias(newName => ENCODING);
475 After that, newName can be used as an alias for ENCODING.
476 ENCODING may be either the name of an encoding or an
479 But before you do so, make sure the alias is nonexistent with
480 C<resolve_alias()>, which returns the canonical name thereof.
483 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
484 Encode::resolve_alias("iso-8859-12") # false; nonexistent
485 Encode::resolve_alias($name) eq $name # true if $name is canonical
487 resolve_alias() does not need C<use Encode::Alias>; it can be
488 exported via C<use Encode qw(resolve_alias)>.
490 See L<Encode::Alias> for details.
492 =head1 Encoding via PerlIO
494 If your perl supports I<PerlIO>, you can use a PerlIO layer to decode
495 and encode directly via a filehandle. The following two examples
496 are totally identical in their functionality.
499 open my $in, "<:encoding(shiftjis)", $infile or die;
500 open my $out, ">:encoding(euc-jp)", $outfile or die;
504 open my $in, "<", $infile or die;
505 open my $out, ">", $outfile or die;
507 from_to($_, "shiftjis", "euc-jp", 1);
510 Unfortunately, there may be encodings are PerlIO-savvy. You can check
511 if your encoding is supported by PerlIO by calling the C<perlio_ok>
514 Encode::perlio_ok("hz"); # False
515 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
517 use Encode qw(perlio_ok); # exported upon request
520 Fortunately, all encodings that come with Encode core are PerlIO-savvy
521 except for hz and ISO-2022-kr. See L<Encode::Encoding> for details.
523 For gory details, see L<Encode::PerlIO>.
525 =head1 Handling Malformed Data
529 The I<CHECK> argument is used as follows. When you omit it,
530 the behaviour is the same as if you had passed a value of 0 for
533 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
535 If I<CHECK> is 0, (en|de)code will put a I<substitution character>
536 in place of a malformed character. For UCM-based encodings,
537 E<lt>subcharE<gt> will be used. For Unicode, "\x{FFFD}" is used.
538 If the data is supposed to be UTF-8, an optional lexical warning
539 (category utf8) is given.
541 =item I<CHECK> = Encode::FB_CROAK ( == 1)
543 If I<CHECK> is 1, methods will die immediately with an error
544 message. Therefore, when I<CHECK> is set to 1, you should trap the
545 fatal error with eval{} unless you really want to let it die on error.
547 =item I<CHECK> = Encode::FB_QUIET
549 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
550 return the portion of the data that has been processed so far when
551 an error occurs. The data argument will be overwritten with
552 everything after that point (that is, the unprocessed part of data).
553 This is handy when you have to call decode repeatedly in the case
554 where your source data may contain partial multi-byte character
555 sequences, for example because you are reading with a fixed-width
556 buffer. Here is some sample code that does exactly this:
559 while(defined(read $fh, $buffer, 256)){
560 # buffer may end in a partial character so we append
562 $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET);
563 # $data now contains the unprocessed partial character
566 =item I<CHECK> = Encode::FB_WARN
568 This is the same as above, except that it warns on error. Handy when
569 you are debugging the mode above.
571 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
573 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF)
575 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF)
577 For encodings that are implemented by Encode::XS, CHECK ==
578 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
580 When you decode, '\xI<XX>' will be inserted for a malformed character,
581 where I<XX> is the hex representation of the octet that could not be
582 decoded to utf8. And when you encode, '\x{I<xxxx>}' will be inserted,
583 where I<xxxx> is the Unicode ID of the character that cannot be found
584 in the character repertoire of the encoding.
586 HTML/XML character reference modes are about the same, in place of
587 \x{I<xxxx>}, HTML uses &#I<1234>; where I<1234> is a decimal digit and
588 XML uses &#xI<abcd>; where I<abcd> is the hexadecimal digit.
592 These modes are actually set via a bitmask. Here is how the FB_XX
593 constants are laid out. You can import the FB_XX constants via
594 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
595 constants via C<use Encode qw(:fallback_all)>.
597 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
600 RETURN_ON_ERR 0x0004 X X
606 =head2 Unimplemented fallback schemes
608 In the future, you will be able to use a code reference to a callback
609 function for the value of I<CHECK> but its API is still undecided.
611 =head1 Defining Encodings
613 To define a new encoding, use:
615 use Encode qw(define_alias);
616 define_encoding($object, 'canonicalName' [, alias...]);
618 I<canonicalName> will be associated with I<$object>. The object
619 should provide the interface described in L<Encode::Encoding>.
620 If more than two arguments are provided then additional
621 arguments are taken as aliases for I<$object>, as for C<define_alias>.
623 See L<Encode::Encoding> for more details.
625 =head1 The UTF-8 flag
627 Before the introduction of utf8 support in perl, The C<eq> operator
628 just compares internal data of the scalars. Now C<eq> means internal
629 data equality AND I<the utf8 flag>. To explain why we made it so, I
630 will quote page 402 of C<Programming Perl, 3rd ed.>
636 Old byte-oriented programs should not spontaneously break on the old
637 byte-oriented data they used to work on.
641 Old byte-oriented programs should magically start working on the new
642 character-oriented data when appropriate.
646 Programs should run just as fast in the new character-oriented mode
647 as in the old byte-oriented mode.
651 Perl should remain one language, rather than forking into a
652 byte-oriented Perl and a character-oriented Perl.
656 Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0
657 was born and many features documented in the book remained
658 unimplemented. Perl 5.8 hopefully correct this and the introduction
659 of UTF-8 flag is one of them. You can think this perl notion of
660 byte-oriented mode (utf8 flag off) and character-oriented mode (utf8
663 Here is how Encode takes care of the utf8 flag.
669 When you encode, the resulting utf8 flag is always off.
673 When you decode, the resuting utf8 flag is on unless you can
674 unambiguously represent data. Here is the definition of
677 After C<$utf8 = decode('foo', $octet);>,
679 When $octet is... The utf8 flag in $utf8 is
680 ---------------------------------------------
681 In ASCII only (or EBCDIC only) OFF
683 In any other Encoding ON
684 ---------------------------------------------
686 As you see, there is one exception, In ASCII. That way you can assue
687 Goal #1. And with Encode Goal #2 is assumed but you still have to be
688 careful in such cases mentioned in B<CAVEAT> paragraphs.
690 This utf8 flag is not visible in perl scripts, exactly for the same
691 reason you cannot (or you I<don't have to>) see if a scalar contains a
692 string, integer, or floating point number. But you can still peek
693 and poke these if you will. See the section below.
697 =head2 Messing with Perl's Internals
699 The following API uses parts of Perl's internals in the current
700 implementation. As such, they are efficient but may change.
704 =item is_utf8(STRING [, CHECK])
706 [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING.
707 If CHECK is true, also checks the data in STRING for being well-formed
708 UTF-8. Returns true if successful, false otherwise.
710 =item _utf8_on(STRING)
712 [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is
713 B<not> checked for being well-formed UTF-8. Do not use unless you
714 B<know> that the STRING is well-formed UTF-8. Returns the previous
715 state of the UTF-8 flag (so please don't treat the return value as
716 indicating success or failure), or C<undef> if STRING is not a string.
718 =item _utf8_off(STRING)
720 [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously.
721 Returns the previous state of the UTF-8 flag (so please don't treat the
722 return value as indicating success or failure), or C<undef> if STRING is
730 L<Encode::Supported>,
737 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
741 This project was originated by Nick Ing-Simmons and later maintained
742 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full
743 list of people involved. For any questions, use
744 E<lt>perl-unicode@perl.orgE<gt> so we can all share share.