2 # $Id: Encode.pm,v 1.65 2002/04/30 16:13:37 dankogai Exp dankogai $
6 our $VERSION = do { my @r = (q$Revision: 1.65 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
9 XSLoader::load(__PACKAGE__, $VERSION);
12 use base qw/Exporter/;
14 # Public, encouraged API is exported by default
17 decode decode_utf8 encode encode_utf8
18 encodings find_encoding
21 our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC
22 PERLQQ HTMLCREF XMLCREF);
23 our @FB_CONSTS = qw(FB_DEFAULT FB_CROAK FB_QUIET FB_WARN
24 FB_PERLQQ FB_HTMLCREF FB_XMLCREF);
29 _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit
30 is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade
32 @FB_FLAGS, @FB_CONSTS,
37 all => [ @EXPORT, @EXPORT_OK ],
38 fallbacks => [ @FB_CONSTS ],
39 fallback_all => [ @FB_CONSTS, @FB_FLAGS ],
42 # 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 };
58 if (@_ and $_[0] eq ":all"){
59 %enc = ( %Encoding, %ExtModule );
62 for my $mod (map {m/::/o ? $_ : "Encode::$_" } @_){
64 for my $enc (keys %ExtModule){
65 $ExtModule{$enc} eq $mod and $enc{$enc} = $mod;
70 sort { lc $a cmp lc $b }
71 grep {!/^(?:Internal|Unicode|Guess)$/o} keys %enc;
75 my $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]);
76 $obj->can("perlio_ok") and return $obj->perlio_ok();
77 return 0; # safety net
84 $Encoding{$name} = $obj;
86 define_alias($lc => $obj) unless $lc eq $name;
89 define_alias($alias, $obj);
96 my ($class, $name, $skip_external) = @_;
98 ref($name) && $name->can('new_sequence') and return $name;
99 exists $Encoding{$name} and return $Encoding{$name};
101 exists $Encoding{$lc} and return $Encoding{$lc};
103 my $oc = $class->find_alias($name);
104 defined($oc) and return $oc;
105 $lc ne $name and $oc = $class->find_alias($lc);
106 defined($oc) and return $oc;
108 unless ($skip_external)
110 if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){
111 $mod =~ s,::,/,g ; $mod .= '.pm';
112 eval{ require $mod; };
113 exists $Encoding{$name} and return $Encoding{$name};
121 my ($name, $skip_external) = @_;
122 return __PACKAGE__->getEncoding($name,$skip_external);
126 my $obj = find_encoding(shift);
127 defined $obj and return $obj->name;
133 my ($name, $string, $check) = @_;
134 defined $string or return;
136 my $enc = find_encoding($name);
137 unless(defined $enc){
139 Carp::croak("Unknown encoding '$name'");
141 my $octets = $enc->encode($string,$check);
142 return undef if ($check && length($string));
148 my ($name,$octets,$check) = @_;
149 defined $octets or return;
151 my $enc = find_encoding($name);
152 unless(defined $enc){
154 Carp::croak("Unknown encoding '$name'");
156 my $string = $enc->decode($octets,$check);
157 $_[1] = $octets if $check;
163 my ($string,$from,$to,$check) = @_;
164 defined $string or return;
166 my $f = find_encoding($from);
169 Carp::croak("Unknown encoding '$from'");
171 my $t = find_encoding($to);
174 Carp::croak("Unknown encoding '$to'");
176 my $uni = $f->decode($string,$check);
177 return undef if ($check && length($string));
178 $string = $t->encode($uni,$check);
179 return undef if ($check && length($uni));
180 return defined($_[0] = $string) ? length($string) : undef ;
186 defined $str or return;
194 defined $str or return;
195 return undef unless utf8::decode($str);
199 predefine_encodings();
202 # This is to restore %Encoding if really needed;
205 sub predefine_encodings{
206 use Encode::Encoding;
208 # was in Encode::UTF_EBCDIC
209 package Encode::UTF_EBCDIC;
210 push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding';
212 my ($obj,$str,$chk) = @_;
214 for (my $i = 0; $i < length($str); $i++) {
216 chr(utf8::unicode_to_native(ord(substr($str,$i,1))));
222 my ($obj,$str,$chk) = @_;
224 for (my $i = 0; $i < length($str); $i++) {
226 chr(utf8::native_to_unicode(ord(substr($str,$i,1))));
231 $Encode::Encoding{Unicode} =
232 bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC";
234 package Encode::Internal;
235 push @Encode::Internal::ISA, 'Encode::Encoding';
237 my ($obj,$str,$chk) = @_;
243 $Encode::Encoding{Unicode} =
244 bless {Name => "Internal"} => "Encode::Internal";
248 # was in Encode::utf8
249 package Encode::utf8;
250 push @Encode::utf8::ISA, 'Encode::Encoding';
252 my ($obj,$octets,$chk) = @_;
253 my $str = Encode::decode_utf8($octets);
261 my ($obj,$string,$chk) = @_;
262 my $octets = Encode::encode_utf8($string);
266 $Encode::Encoding{utf8} =
267 bless {Name => "utf8"} => "Encode::utf8";
277 Encode - character encodings
283 =head2 Table of Contents
285 Encode consists of a collection of modules whose details are too big
286 to fit in one document. This POD itself explains the top-level APIs
287 and general topics at a glance. For other topics and more details,
291 --------------------------------------------------------
292 Encode::Alias Alias definitions to encodings
293 Encode::Encoding Encode Implementation Base Class
294 Encode::Supported List of Supported Encodings
295 Encode::CN Simplified Chinese Encodings
296 Encode::JP Japanese Encodings
297 Encode::KR Korean Encodings
298 Encode::TW Traditional Chinese Encodings
299 --------------------------------------------------------
303 The C<Encode> module provides the interfaces between Perl's strings
304 and the rest of the system. Perl strings are sequences of
307 The repertoire of characters that Perl can represent is at least that
308 defined by the Unicode Consortium. On most platforms the ordinal
309 values of the characters (as returned by C<ord(ch)>) is the "Unicode
310 codepoint" for the character (the exceptions are those platforms where
311 the legacy encoding is some variant of EBCDIC rather than a super-set
312 of ASCII - see L<perlebcdic>).
314 Traditionally, computer data has been moved around in 8-bit chunks
315 often called "bytes". These chunks are also known as "octets" in
316 networking standards. Perl is widely used to manipulate data of many
317 types - not only strings of characters representing human or computer
318 languages but also "binary" data being the machine's representation of
319 numbers, pixels in an image - or just about anything.
321 When Perl is processing "binary data", the programmer wants Perl to
322 process "sequences of bytes". This is not a problem for Perl - as a
323 byte has 256 possible values, it easily fits in Perl's much larger
332 I<character>: a character in the range 0..(2**32-1) (or more).
333 (What Perl's strings are made of.)
337 I<byte>: a character in the range 0..255
338 (A special case of a Perl character.)
342 I<octet>: 8 bits of data, with ordinal values 0..255
343 (Term for bytes passed to or from a non-Perl context, e.g. a disk file.)
347 =head1 PERL ENCODING API
351 =item $octets = encode(ENCODING, $string [, CHECK])
353 Encodes a string from Perl's internal form into I<ENCODING> and returns
354 a sequence of octets. ENCODING can be either a canonical name or
355 an alias. For encoding names and aliases, see L</"Defining Aliases">.
356 For CHECK, see L</"Handling Malformed Data">.
358 For example, to convert a string from Perl's internal format to
359 iso-8859-1 (also known as Latin1),
361 $octets = encode("iso-8859-1", $string);
363 B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets
364 B<may not be equal to> $string. Though they both contain the same data, the utf8 flag
365 for $octets is B<always> off. When you encode anything, utf8 flag of
366 the result is always off, even when it contains completely valid utf8
367 string. See L</"The UTF-8 flag"> below.
369 =item $string = decode(ENCODING, $octets [, CHECK])
371 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
372 internal form and returns the resulting string. As in encode(),
373 ENCODING can be either a canonical name or an alias. For encoding names
374 and aliases, see L</"Defining Aliases">. For CHECK, see
375 L</"Handling Malformed Data">.
377 For example, to convert ISO-8859-1 data to a string in Perl's internal format:
379 $string = decode("iso-8859-1", $octets);
381 B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string
382 B<may not be equal to> $octets. Though they both contain the same data,
383 the utf8 flag for $string is on unless $octets entirely consists of
384 ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag">
387 =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK])
389 Converts B<in-place> data between two encodings. The data in $octets
390 must be encoded as octets and not as characters in Perl's internal
391 format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 encoding:
393 from_to($octets, "iso-8859-1", "cp1250");
395 and to convert it back:
397 from_to($octets, "cp1250", "iso-8859-1");
399 Note that because the conversion happens in place, the data to be
400 converted cannot be a string constant; it must be a scalar variable.
402 from_to() returns the length of the converted string in octets on success, undef
405 B<CAVEAT>: The following operations look the same but are not quite so;
407 from_to($data, "iso-8859-1", "utf8"); #1
408 $data = decode("iso-8859-1", $data); #2
410 Both #1 and #2 make $data consist of a completely valid UTF-8 string
411 but only #2 turns utf8 flag on. #1 is equivalent to
413 $data = encode("utf8", decode("iso-8859-1", $data));
415 See L</"The UTF-8 flag"> below.
417 =item $octets = encode_utf8($string);
419 Equivalent to C<$octets = encode("utf8", $string);> The characters
420 that comprise $string are encoded in Perl's internal format and the
421 result is returned as a sequence of octets. All possible
422 characters have a UTF-8 representation so this function cannot fail.
425 =item $string = decode_utf8($octets [, CHECK]);
427 equivalent to C<$string = decode("utf8", $octets [, CHECK])>.
428 The sequence of octets represented by
429 $octets is decoded from UTF-8 into a sequence of logical
430 characters. Not all sequences of octets form valid UTF-8 encodings, so
431 it is possible for this call to fail. For CHECK, see
432 L</"Handling Malformed Data">.
436 =head2 Listing available encodings
439 @list = Encode->encodings();
441 Returns a list of the canonical names of the available encodings that
442 are loaded. To get a list of all available encodings including the
443 ones that are not loaded yet, say
445 @all_encodings = Encode->encodings(":all");
447 Or you can give the name of a specific module.
449 @with_jp = Encode->encodings("Encode::JP");
451 When "::" is not in the name, "Encode::" is assumed.
453 @ebcdic = Encode->encodings("EBCDIC");
455 To find out in detail which encodings are supported by this package,
456 see L<Encode::Supported>.
458 =head2 Defining Aliases
460 To add a new alias to a given encoding, use:
464 define_alias(newName => ENCODING);
466 After that, newName can be used as an alias for ENCODING.
467 ENCODING may be either the name of an encoding or an
470 But before you do so, make sure the alias is nonexistent with
471 C<resolve_alias()>, which returns the canonical name thereof.
474 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
475 Encode::resolve_alias("iso-8859-12") # false; nonexistent
476 Encode::resolve_alias($name) eq $name # true if $name is canonical
478 resolve_alias() does not need C<use Encode::Alias>; it can be
479 exported via C<use Encode qw(resolve_alias)>.
481 See L<Encode::Alias> for details.
483 =head1 Encoding via PerlIO
485 If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode
486 and encode directly via a filehandle. The following two examples
487 are totally identical in their functionality.
490 open my $in, "<:encoding(shiftjis)", $infile or die;
491 open my $out, ">:encoding(euc-jp)", $outfile or die;
492 while(<$in>){ print $out $_; }
495 open my $in, "<", $infile or die;
496 open my $out, ">", $outfile or die;
498 from_to($_, "shiftjis", "euc-jp", 1);
502 Unfortunately, it may be that encodings are PerlIO-savvy. You can check
503 if your encoding is supported by PerlIO by calling the C<perlio_ok>
506 Encode::perlio_ok("hz"); # False
507 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
509 use Encode qw(perlio_ok); # exported upon request
512 Fortunately, all encodings that come with Encode core are PerlIO-savvy
513 except for hz and ISO-2022-kr. For gory details, see L<Encode::Encoding> and L<Encode::PerlIO>.
515 =head1 Handling Malformed Data
519 The I<CHECK> argument is used as follows. When you omit it,
520 the behaviour is the same as if you had passed a value of 0 for
523 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
525 If I<CHECK> is 0, (en|de)code will put a I<substitution character>
526 in place of a malformed character. For UCM-based encodings,
527 E<lt>subcharE<gt> will be used. For Unicode, the code point C<0xFFFD> is used.
528 If the data is supposed to be UTF-8, an optional lexical warning
529 (category utf8) is given.
531 =item I<CHECK> = Encode::FB_CROAK ( == 1)
533 If I<CHECK> is 1, methods will die on error immediately with an error
534 message. Therefore, when I<CHECK> is set to 1, you should trap the
535 fatal error with eval{} unless you really want to let it die on error.
537 =item I<CHECK> = Encode::FB_QUIET
539 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
540 return the portion of the data that has been processed so far when
541 an error occurs. The data argument will be overwritten with
542 everything after that point (that is, the unprocessed part of data).
543 This is handy when you have to call decode repeatedly in the case
544 where your source data may contain partial multi-byte character
545 sequences, for example because you are reading with a fixed-width
546 buffer. Here is some sample code that does exactly this:
548 my $data = ''; my $utf8 = '';
549 while(defined(read $fh, $buffer, 256)){
550 # buffer may end in a partial character so we append
552 $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET);
553 # $data now contains the unprocessed partial character
556 =item I<CHECK> = Encode::FB_WARN
558 This is the same as above, except that it warns on error. Handy when
559 you are debugging the mode above.
561 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
563 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF)
565 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF)
567 For encodings that are implemented by Encode::XS, CHECK ==
568 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
570 When you decode, C<\xI<HH>> will be inserted for a malformed character,
571 where I<HH> is the hex representation of the octet that could not be
572 decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted,
573 where I<HHHH> is the Unicode ID of the character that cannot be found
574 in the character repertoire of the encoding.
576 HTML/XML character reference modes are about the same, in place of
577 C<\x{I<HHHH>}>, HTML uses C<&#I<NNNN>>; where I<NNNN> is a decimal digit and
578 XML uses C<&#xI<HHHH>>; where I<HHHH> is the hexadecimal digit.
582 These modes are actually set via a bitmask. Here is how the FB_XX
583 constants are laid out. You can import the FB_XX constants via
584 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
585 constants via C<use Encode qw(:fallback_all)>.
587 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
590 RETURN_ON_ERR 0x0004 X X
596 =head2 Unimplemented fallback schemes
598 In the future, you will be able to use a code reference to a callback
599 function for the value of I<CHECK> but its API is still undecided.
601 =head1 Defining Encodings
603 To define a new encoding, use:
605 use Encode qw(define_encoding);
606 define_encoding($object, 'canonicalName' [, alias...]);
608 I<canonicalName> will be associated with I<$object>. The object
609 should provide the interface described in L<Encode::Encoding>.
610 If more than two arguments are provided then additional
611 arguments are taken as aliases for I<$object>.
613 See L<Encode::Encoding> for more details.
615 =head1 The UTF-8 flag
617 Before the introduction of utf8 support in perl, The C<eq> operator
618 just compared the strings represented by two scalars. Beginning with
619 perl 5.8, C<eq> compares two strings with simultaneous consideration
620 of I<the utf8 flag>. To explain why we made it so, I will quote page
621 402 of C<Programming Perl, 3rd ed.>
627 Old byte-oriented programs should not spontaneously break on the old
628 byte-oriented data they used to work on.
632 Old byte-oriented programs should magically start working on the new
633 character-oriented data when appropriate.
637 Programs should run just as fast in the new character-oriented mode
638 as in the old byte-oriented mode.
642 Perl should remain one language, rather than forking into a
643 byte-oriented Perl and a character-oriented Perl.
647 Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0
648 was born and many features documented in the book remained
649 unimplemented for a long time. Perl 5.8 corrected this and the introduction
650 of the UTF-8 flag is one of them. You can think of this perl notion as of a
651 byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8
654 Here is how Encode takes care of the utf8 flag.
660 When you encode, the resulting utf8 flag is always off.
664 When you decode, the resulting utf8 flag is on unless you can
665 unambiguously represent data. Here is the definition of
668 After C<$utf8 = decode('foo', $octet);>,
670 When $octet is... The utf8 flag in $utf8 is
671 ---------------------------------------------
672 In ASCII only (or EBCDIC only) OFF
674 In any other Encoding ON
675 ---------------------------------------------
677 As you see, there is one exception, In ASCII. That way you can assue
678 Goal #1. And with Encode Goal #2 is assumed but you still have to be
679 careful in such cases mentioned in B<CAVEAT> paragraphs.
681 This utf8 flag is not visible in perl scripts, exactly for the same
682 reason you cannot (or you I<don't have to>) see if a scalar contains a
683 string, integer, or floating point number. But you can still peek
684 and poke these if you will. See the section below.
688 =head2 Messing with Perl's Internals
690 The following API uses parts of Perl's internals in the current
691 implementation. As such, they are efficient but may change.
695 =item is_utf8(STRING [, CHECK])
697 [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING.
698 If CHECK is true, also checks the data in STRING for being well-formed
699 UTF-8. Returns true if successful, false otherwise.
701 =item _utf8_on(STRING)
703 [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is
704 B<not> checked for being well-formed UTF-8. Do not use unless you
705 B<know> that the STRING is well-formed UTF-8. Returns the previous
706 state of the UTF-8 flag (so please don't treat the return value as
707 indicating success or failure), or C<undef> if STRING is not a string.
709 =item _utf8_off(STRING)
711 [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously.
712 Returns the previous state of the UTF-8 flag (so please don't treat the
713 return value as indicating success or failure), or C<undef> if STRING is
721 L<Encode::Supported>,
728 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
732 This project was originated by Nick Ing-Simmons and later maintained
733 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full
734 list of people involved. For any questions, use
735 E<lt>perl-unicode@perl.orgE<gt> so we can all share.