2 # $Id: Encode.pm,v 2.23 2007/05/29 18:15:32 dankogai Exp dankogai $
7 our $VERSION = sprintf "%d.%02d", q$Revision: 2.23 $ =~ /(\d+)/g;
10 XSLoader::load( __PACKAGE__, $VERSION );
13 use base qw/Exporter/;
15 # Public, encouraged API is exported by default
18 decode decode_utf8 encode encode_utf8 str2bytes bytes2str
19 encodings find_encoding clone_encoding
22 DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC
23 PERLQQ HTMLCREF XMLCREF STOP_AT_PARTIAL
26 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN
27 FB_PERLQQ FB_HTMLCREF FB_XMLCREF
31 _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit
32 is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade
34 @FB_FLAGS, @FB_CONSTS,
38 all => [ @EXPORT, @EXPORT_OK ],
39 fallbacks => [@FB_CONSTS],
40 fallback_all => [ @FB_CONSTS, @FB_FLAGS ],
43 # Documentation moved after __END__ for speed - NI-S
45 our $ON_EBCDIC = ( ord("A") == 193 );
49 # Make a %Encoding package variable to allow a certain amount of cheating
52 require Encode::Config;
53 eval { require Encode::ConfigLocal };
58 if ( @_ and $_[0] eq ":all" ) {
59 %enc = ( %Encoding, %ExtModule );
63 for my $mod ( map { m/::/o ? $_ : "Encode::$_" } @_ ) {
65 for my $enc ( keys %ExtModule ) {
66 $ExtModule{$enc} eq $mod and $enc{$enc} = $mod;
70 return 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
83 $Encoding{$name} = $obj;
85 define_alias( $lc => $obj ) unless $lc eq $name;
88 define_alias( $alias, $obj );
94 my ( $class, $name, $skip_external ) = @_;
96 ref($name) && $name->can('renew') and return $name;
97 exists $Encoding{$name} and return $Encoding{$name};
99 exists $Encoding{$lc} and return $Encoding{$lc};
101 my $oc = $class->find_alias($name);
102 defined($oc) and return $oc;
103 $lc ne $name and $oc = $class->find_alias($lc);
104 defined($oc) and return $oc;
106 unless ($skip_external) {
107 if ( my $mod = $ExtModule{$name} || $ExtModule{$lc} ) {
110 eval { require $mod; };
111 exists $Encoding{$name} and return $Encoding{$name};
117 sub find_encoding($;$) {
118 my ( $name, $skip_external ) = @_;
119 return __PACKAGE__->getEncoding( $name, $skip_external );
122 sub resolve_alias($) {
123 my $obj = find_encoding(shift);
124 defined $obj and return $obj->name;
128 sub clone_encoding($) {
129 my $obj = find_encoding(shift);
131 eval { require Storable };
133 return Storable::dclone($obj);
137 my ( $name, $string, $check ) = @_;
138 return undef unless defined $string;
139 $string .= '' if ref $string; # stringify;
141 my $enc = find_encoding($name);
142 unless ( defined $enc ) {
144 Carp::croak("Unknown encoding '$name'");
146 my $octets = $enc->encode( $string, $check );
147 $_[1] = $string if $check and !ref $check and !( $check & LEAVE_SRC() );
150 *str2bytes = \&encode;
153 my ( $name, $octets, $check ) = @_;
154 return undef unless defined $octets;
155 $octets .= '' if ref $octets;
157 my $enc = find_encoding($name);
158 unless ( defined $enc ) {
160 Carp::croak("Unknown encoding '$name'");
162 my $string = $enc->decode( $octets, $check );
163 $_[1] = $octets if $check and !ref $check and !( $check & LEAVE_SRC() );
166 *bytes2str = \&decode;
169 my ( $string, $from, $to, $check ) = @_;
170 return undef unless defined $string;
172 my $f = find_encoding($from);
173 unless ( defined $f ) {
175 Carp::croak("Unknown encoding '$from'");
177 my $t = find_encoding($to);
178 unless ( defined $t ) {
180 Carp::croak("Unknown encoding '$to'");
182 my $uni = $f->decode($string);
183 $_[0] = $string = $t->encode( $uni, $check );
184 return undef if ( $check && length($uni) );
185 return defined( $_[0] ) ? length($string) : undef;
194 sub decode_utf8($;$) {
195 my ( $str, $check ) = @_;
196 return $str if is_utf8($str);
198 return decode( "utf8", $str, $check );
201 return decode( "utf8", $str );
206 predefine_encodings(1);
209 # This is to restore %Encoding if really needed;
212 sub predefine_encodings {
213 require Encode::Encoding;
214 no warnings 'redefine';
218 # was in Encode::UTF_EBCDIC
219 package Encode::UTF_EBCDIC;
220 push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding';
222 my ( $obj, $str, $chk ) = @_;
224 for ( my $i = 0 ; $i < length($str) ; $i++ ) {
227 utf8::unicode_to_native( ord( substr( $str, $i, 1 ) ) )
234 my ( $obj, $str, $chk ) = @_;
236 for ( my $i = 0 ; $i < length($str) ; $i++ ) {
239 utf8::native_to_unicode( ord( substr( $str, $i, 1 ) ) )
245 $Encode::Encoding{Unicode} =
246 bless { Name => "UTF_EBCDIC" } => "Encode::UTF_EBCDIC";
250 package Encode::Internal;
251 push @Encode::Internal::ISA, 'Encode::Encoding';
253 my ( $obj, $str, $chk ) = @_;
259 $Encode::Encoding{Unicode} =
260 bless { Name => "Internal" } => "Encode::Internal";
265 # was in Encode::utf8
266 package Encode::utf8;
267 push @Encode::utf8::ISA, 'Encode::Encoding';
271 Encode::DEBUG and warn __PACKAGE__, " XS on";
272 *decode = \&decode_xs;
273 *encode = \&encode_xs;
276 Encode::DEBUG and warn __PACKAGE__, " XS off";
278 my ( $obj, $octets, $chk ) = @_;
279 my $str = Encode::decode_utf8($octets);
280 if ( defined $str ) {
287 my ( $obj, $string, $chk ) = @_;
288 my $octets = Encode::encode_utf8($string);
293 *cat_decode = sub { # ($obj, $dst, $src, $pos, $trm, $chk)
294 # currently ignores $chk
295 my ( $obj, undef, undef, $pos, $trm ) = @_;
296 my ( $rdst, $rsrc, $rpos ) = \@_[ 1, 2, 3 ];
298 if ( ( my $npos = index( $$rsrc, $trm, $pos ) ) >= 0 ) {
300 substr( $$rsrc, $pos, $npos - $pos + length($trm) );
301 $$rpos = $npos + length($trm);
304 $$rdst .= substr( $$rsrc, $pos );
305 $$rpos = length($$rsrc);
308 $Encode::Encoding{utf8} =
309 bless { Name => "utf8" } => "Encode::utf8";
310 $Encode::Encoding{"utf-8-strict"} =
311 bless { Name => "utf-8-strict", strict_utf8 => 1 } =>
322 Encode - character encodings
328 =head2 Table of Contents
330 Encode consists of a collection of modules whose details are too big
331 to fit in one document. This POD itself explains the top-level APIs
332 and general topics at a glance. For other topics and more details,
336 --------------------------------------------------------
337 Encode::Alias Alias definitions to encodings
338 Encode::Encoding Encode Implementation Base Class
339 Encode::Supported List of Supported Encodings
340 Encode::CN Simplified Chinese Encodings
341 Encode::JP Japanese Encodings
342 Encode::KR Korean Encodings
343 Encode::TW Traditional Chinese Encodings
344 --------------------------------------------------------
348 The C<Encode> module provides the interfaces between Perl's strings
349 and the rest of the system. Perl strings are sequences of
352 The repertoire of characters that Perl can represent is at least that
353 defined by the Unicode Consortium. On most platforms the ordinal
354 values of the characters (as returned by C<ord(ch)>) is the "Unicode
355 codepoint" for the character (the exceptions are those platforms where
356 the legacy encoding is some variant of EBCDIC rather than a super-set
357 of ASCII - see L<perlebcdic>).
359 Traditionally, computer data has been moved around in 8-bit chunks
360 often called "bytes". These chunks are also known as "octets" in
361 networking standards. Perl is widely used to manipulate data of many
362 types - not only strings of characters representing human or computer
363 languages but also "binary" data being the machine's representation of
364 numbers, pixels in an image - or just about anything.
366 When Perl is processing "binary data", the programmer wants Perl to
367 process "sequences of bytes". This is not a problem for Perl - as a
368 byte has 256 possible values, it easily fits in Perl's much larger
377 I<character>: a character in the range 0..(2**32-1) (or more).
378 (What Perl's strings are made of.)
382 I<byte>: a character in the range 0..255
383 (A special case of a Perl character.)
387 I<octet>: 8 bits of data, with ordinal values 0..255
388 (Term for bytes passed to or from a non-Perl context, e.g. a disk file.)
392 =head1 PERL ENCODING API
396 =item $octets = encode(ENCODING, $string [, CHECK])
398 Encodes a string from Perl's internal form into I<ENCODING> and returns
399 a sequence of octets. ENCODING can be either a canonical name or
400 an alias. For encoding names and aliases, see L</"Defining Aliases">.
401 For CHECK, see L</"Handling Malformed Data">.
403 For example, to convert a string from Perl's internal format to
404 iso-8859-1 (also known as Latin1),
406 $octets = encode("iso-8859-1", $string);
408 B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then
409 $octets B<may not be equal to> $string. Though they both contain the
410 same data, the UTF8 flag for $octets is B<always> off. When you
411 encode anything, UTF8 flag of the result is always off, even when it
412 contains completely valid utf8 string. See L</"The UTF8 flag"> below.
414 If the $string is C<undef> then C<undef> is returned.
416 =item $string = decode(ENCODING, $octets [, CHECK])
418 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
419 internal form and returns the resulting string. As in encode(),
420 ENCODING can be either a canonical name or an alias. For encoding names
421 and aliases, see L</"Defining Aliases">. For CHECK, see
422 L</"Handling Malformed Data">.
424 For example, to convert ISO-8859-1 data to a string in Perl's internal format:
426 $string = decode("iso-8859-1", $octets);
428 B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string
429 B<may not be equal to> $octets. Though they both contain the same data,
430 the UTF8 flag for $string is on unless $octets entirely consists of
431 ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF8 flag">
434 If the $string is C<undef> then C<undef> is returned.
436 =item [$obj =] find_encoding(ENCODING)
438 Returns the I<encoding object> corresponding to ENCODING. Returns
439 undef if no matching ENCODING is find.
441 This object is what actually does the actual (en|de)coding.
443 $utf8 = decode($name, $bytes);
448 $obj = find_encoding($name);
449 croak qq(encoding "$name" not found) unless ref $obj;
453 with more error checking.
455 Therefore you can save time by reusing this object as follows;
457 my $enc = find_encoding("iso-8859-1");
459 my $utf8 = $enc->decode($_);
460 # and do someting with $utf8;
463 Besides C<< ->decode >> and C<< ->encode >>, other methods are
464 available as well. For instance, C<< -> name >> returns the canonical
465 name of the encoding object.
467 find_encoding("latin1")->name; # iso-8859-1
469 See L<Encode::Encoding> for details.
471 =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK])
473 Converts B<in-place> data between two encodings. The data in $octets
474 must be encoded as octets and not as characters in Perl's internal
475 format. For example, to convert ISO-8859-1 data to Microsoft's CP1250
478 from_to($octets, "iso-8859-1", "cp1250");
480 and to convert it back:
482 from_to($octets, "cp1250", "iso-8859-1");
484 Note that because the conversion happens in place, the data to be
485 converted cannot be a string constant; it must be a scalar variable.
487 from_to() returns the length of the converted string in octets on
488 success, I<undef> on error.
490 B<CAVEAT>: The following operations look the same but are not quite so;
492 from_to($data, "iso-8859-1", "utf8"); #1
493 $data = decode("iso-8859-1", $data); #2
495 Both #1 and #2 make $data consist of a completely valid UTF-8 string
496 but only #2 turns UTF8 flag on. #1 is equivalent to
498 $data = encode("utf8", decode("iso-8859-1", $data));
500 See L</"The UTF8 flag"> below.
504 from_to($octets, $from, $to, $check);
508 $octets = encode($to, decode($from, $octets), $check);
510 Yes, it does not respect the $check during decoding. It is
511 deliberately done that way. If you need minute control, C<decode>
512 then C<encode> as follows;
514 $octets = encode($to, decode($from, $octets, $check_from), $check_to);
516 =item $octets = encode_utf8($string);
518 Equivalent to C<$octets = encode("utf8", $string);> The characters
519 that comprise $string are encoded in Perl's internal format and the
520 result is returned as a sequence of octets. All possible
521 characters have a UTF-8 representation so this function cannot fail.
524 =item $string = decode_utf8($octets [, CHECK]);
526 equivalent to C<$string = decode("utf8", $octets [, CHECK])>.
527 The sequence of octets represented by
528 $octets is decoded from UTF-8 into a sequence of logical
529 characters. Not all sequences of octets form valid UTF-8 encodings, so
530 it is possible for this call to fail. For CHECK, see
531 L</"Handling Malformed Data">.
535 =head2 Listing available encodings
538 @list = Encode->encodings();
540 Returns a list of the canonical names of the available encodings that
541 are loaded. To get a list of all available encodings including the
542 ones that are not loaded yet, say
544 @all_encodings = Encode->encodings(":all");
546 Or you can give the name of a specific module.
548 @with_jp = Encode->encodings("Encode::JP");
550 When "::" is not in the name, "Encode::" is assumed.
552 @ebcdic = Encode->encodings("EBCDIC");
554 To find out in detail which encodings are supported by this package,
555 see L<Encode::Supported>.
557 =head2 Defining Aliases
559 To add a new alias to a given encoding, use:
563 define_alias(newName => ENCODING);
565 After that, newName can be used as an alias for ENCODING.
566 ENCODING may be either the name of an encoding or an
569 But before you do so, make sure the alias is nonexistent with
570 C<resolve_alias()>, which returns the canonical name thereof.
573 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
574 Encode::resolve_alias("iso-8859-12") # false; nonexistent
575 Encode::resolve_alias($name) eq $name # true if $name is canonical
577 resolve_alias() does not need C<use Encode::Alias>; it can be
578 exported via C<use Encode qw(resolve_alias)>.
580 See L<Encode::Alias> for details.
582 =head2 Finding IANA Character Set Registry names
584 The canonical name of a given encoding does not necessarily agree with
585 IANA IANA Character Set Registry, commonly seen as C<< Content-Type:
586 text/plain; charset=I<whatever> >>. For most cases canonical names
587 work but sometimes it does not (notably 'utf-8-strict').
589 Therefore as of Encode version 2.21, a new method C<mime_name()> is added.
592 my $enc = find_encoding('UTF-8');
593 warn $enc->name; # utf-8-strict
594 warn $enc->mime_name; # UTF-8
596 See also: L<Encode::Encoding>
598 =head1 Encoding via PerlIO
600 If your perl supports I<PerlIO> (which is the default), you can use a
601 PerlIO layer to decode and encode directly via a filehandle. The
602 following two examples are totally identical in their functionality.
605 open my $in, "<:encoding(shiftjis)", $infile or die;
606 open my $out, ">:encoding(euc-jp)", $outfile or die;
607 while(<$in>){ print $out $_; }
610 open my $in, "<", $infile or die;
611 open my $out, ">", $outfile or die;
613 from_to($_, "shiftjis", "euc-jp", 1);
617 Unfortunately, it may be that encodings are PerlIO-savvy. You can check
618 if your encoding is supported by PerlIO by calling the C<perlio_ok>
621 Encode::perlio_ok("hz"); # False
622 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
624 use Encode qw(perlio_ok); # exported upon request
627 Fortunately, all encodings that come with Encode core are PerlIO-savvy
628 except for hz and ISO-2022-kr. For gory details, see
629 L<Encode::Encoding> and L<Encode::PerlIO>.
631 =head1 Handling Malformed Data
633 The optional I<CHECK> argument tells Encode what to do when it
634 encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 )
637 As of version 2.12 Encode supports coderef values for CHECK. See below.
641 =item B<NOTE:> Not all encoding support this feature
643 Some encodings ignore I<CHECK> argument. For example,
644 L<Encode::Unicode> ignores I<CHECK> and it always croaks on error.
648 Now here is the list of I<CHECK> values available
652 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
654 If I<CHECK> is 0, (en|de)code will put a I<substitution character> in
655 place of a malformed character. When you encode, E<lt>subcharE<gt>
656 will be used. When you decode the code point C<0xFFFD> is used. If
657 the data is supposed to be UTF-8, an optional lexical warning
658 (category utf8) is given.
660 =item I<CHECK> = Encode::FB_CROAK ( == 1)
662 If I<CHECK> is 1, methods will die on error immediately with an error
663 message. Therefore, when I<CHECK> is set to 1, you should trap the
664 error with eval{} unless you really want to let it die.
666 =item I<CHECK> = Encode::FB_QUIET
668 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
669 return the portion of the data that has been processed so far when an
670 error occurs. The data argument will be overwritten with everything
671 after that point (that is, the unprocessed part of data). This is
672 handy when you have to call decode repeatedly in the case where your
673 source data may contain partial multi-byte character sequences,
674 (i.e. you are reading with a fixed-width buffer). Here is a sample
675 code that does exactly this:
677 my $buffer = ''; my $string = '';
678 while(read $fh, $buffer, 256, length($buffer)){
679 $string .= decode($encoding, $buffer, Encode::FB_QUIET);
680 # $buffer now contains the unprocessed partial character
683 =item I<CHECK> = Encode::FB_WARN
685 This is the same as above, except that it warns on error. Handy when
686 you are debugging the mode above.
688 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
690 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF)
692 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF)
694 For encodings that are implemented by Encode::XS, CHECK ==
695 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
697 When you decode, C<\xI<HH>> will be inserted for a malformed character,
698 where I<HH> is the hex representation of the octet that could not be
699 decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted,
700 where I<HHHH> is the Unicode ID of the character that cannot be found
701 in the character repertoire of the encoding.
703 HTML/XML character reference modes are about the same, in place of
704 C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and
705 XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number.
707 In Encode 2.10 or later, C<LEAVE_SRC> is also implied.
711 These modes are actually set via a bitmask. Here is how the FB_XX
712 constants are laid out. You can import the FB_XX constants via
713 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
714 constants via C<use Encode qw(:fallback_all)>.
716 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
719 RETURN_ON_ERR 0x0004 X X
729 =item Encode::LEAVE_SRC
731 If the C<Encode::LEAVE_SRC> bit is not set, but I<CHECK> is, then the second
732 argument to C<encode()> or C<decode()> may be assigned to by the functions. If
733 you're not interested in this, then bitwise-or the bitmask with it.
737 =Head2 coderef for CHECK
739 As of Encode 2.12 CHECK can also be a code reference which takes the
740 ord value of unmapped caharacter as an argument and returns a string
741 that represents the fallback character. For instance,
743 $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift });
745 Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of
748 =head1 Defining Encodings
750 To define a new encoding, use:
752 use Encode qw(define_encoding);
753 define_encoding($object, 'canonicalName' [, alias...]);
755 I<canonicalName> will be associated with I<$object>. The object
756 should provide the interface described in L<Encode::Encoding>.
757 If more than two arguments are provided then additional
758 arguments are taken as aliases for I<$object>.
760 See L<Encode::Encoding> for more details.
764 Before the introduction of Unicode support in perl, The C<eq> operator
765 just compared the strings represented by two scalars. Beginning with
766 perl 5.8, C<eq> compares two strings with simultaneous consideration of
767 I<the UTF8 flag>. To explain why we made it so, I will quote page 402 of
768 C<Programming Perl, 3rd ed.>
774 Old byte-oriented programs should not spontaneously break on the old
775 byte-oriented data they used to work on.
779 Old byte-oriented programs should magically start working on the new
780 character-oriented data when appropriate.
784 Programs should run just as fast in the new character-oriented mode
785 as in the old byte-oriented mode.
789 Perl should remain one language, rather than forking into a
790 byte-oriented Perl and a character-oriented Perl.
794 Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0
795 was born and many features documented in the book remained
796 unimplemented for a long time. Perl 5.8 corrected this and the introduction
797 of the UTF8 flag is one of them. You can think of this perl notion as of a
798 byte-oriented mode (UTF8 flag off) and a character-oriented mode (UTF8
801 Here is how Encode takes care of the UTF8 flag.
807 When you encode, the resulting UTF8 flag is always off.
811 When you decode, the resulting UTF8 flag is on unless you can
812 unambiguously represent data. Here is the definition of
815 After C<$utf8 = decode('foo', $octet);>,
817 When $octet is... The UTF8 flag in $utf8 is
818 ---------------------------------------------
819 In ASCII only (or EBCDIC only) OFF
821 In any other Encoding ON
822 ---------------------------------------------
824 As you see, there is one exception, In ASCII. That way you can assume
825 Goal #1. And with Encode Goal #2 is assumed but you still have to be
826 careful in such cases mentioned in B<CAVEAT> paragraphs.
828 This UTF8 flag is not visible in perl scripts, exactly for the same
829 reason you cannot (or you I<don't have to>) see if a scalar contains a
830 string, integer, or floating point number. But you can still peek
831 and poke these if you will. See the section below.
835 =head2 Messing with Perl's Internals
837 The following API uses parts of Perl's internals in the current
838 implementation. As such, they are efficient but may change.
842 =item is_utf8(STRING [, CHECK])
844 [INTERNAL] Tests whether the UTF8 flag is turned on in the STRING.
845 If CHECK is true, also checks the data in STRING for being well-formed
846 UTF-8. Returns true if successful, false otherwise.
848 As of perl 5.8.1, L<utf8> also has utf8::is_utf8().
850 =item _utf8_on(STRING)
852 [INTERNAL] Turns on the UTF8 flag in STRING. The data in STRING is
853 B<not> checked for being well-formed UTF-8. Do not use unless you
854 B<know> that the STRING is well-formed UTF-8. Returns the previous
855 state of the UTF8 flag (so please don't treat the return value as
856 indicating success or failure), or C<undef> if STRING is not a string.
858 =item _utf8_off(STRING)
860 [INTERNAL] Turns off the UTF8 flag in STRING. Do not use frivolously.
861 Returns the previous state of the UTF8 flag (so please don't treat the
862 return value as indicating success or failure), or C<undef> if STRING is
867 =head1 UTF-8 vs. utf8 vs. UTF8
869 ....We now view strings not as sequences of bytes, but as sequences
870 of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit
871 computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed.
873 That has been the perl's notion of UTF-8 but official UTF-8 is more
874 strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are
875 not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al).
877 Now that is overruled by Larry Wall himself.
879 From: Larry Wall <larry@wall.org>
880 Date: December 04, 2004 11:51:58 JST
881 To: perl-unicode@perl.org
882 Subject: Re: Make Encode.pm support the real UTF-8
883 Message-Id: <20041204025158.GA28754@wall.org>
885 On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote:
886 : I've no problem with 'utf8' being perl's unrestricted uft8 encoding,
887 : but "UTF-8" is the name of the standard and should give the
888 : corresponding behaviour.
890 For what it's worth, that's how I've always kept them straight in my
893 Also for what it's worth, Perl 6 will mostly default to strict but
894 make it easy to switch back to lax.
898 Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8
899 while B<utf8> means liberal, lax, version thereof. And Encode version
900 2.10 or later thus groks the difference between C<UTF-8> and C"utf8".
902 encode("utf8", "\x{FFFF_FFFF}", 1); # okay
903 encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks
905 C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>.
906 Yes, the hyphen between "UTF" and "8" is important. Without it Encode
909 find_encoding("UTF-8")->name # is 'utf-8-strict'
910 find_encoding("utf-8")->name # ditto. names are case insensitive
911 find_encoding("utf_8")->name # ditto. "_" are treated as "-"
912 find_encoding("UTF8")->name # is 'utf8'.
914 The UTF8 flag is internally called UTF8, without a hyphen. It indicates
915 whether a string is internally encoded as utf8, also without a hypen.
920 L<Encode::Supported>,
925 L<perlunicode>, L<perluniintro>, L<perlunifaq>, L<perlunitut>
927 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
931 This project was originated by Nick Ing-Simmons and later maintained
932 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full
933 list of people involved. For any questions, use
934 E<lt>perl-unicode@perl.orgE<gt> so we can all share.
936 While Dan Kogai retains the copyright as a maintainer, the credit
937 should go to all those involoved. See AUTHORS for those submitted
942 Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt>
944 This library is free software; you can redistribute it and/or modify
945 it under the same terms as Perl itself.