2 # $Id: Encode.pm,v 2.25 2008/05/07 20:56:05 dankogai Exp dankogai $
7 our $VERSION = sprintf "%d.%02d", q$Revision: 2.25 $ =~ /(\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 default => [ @EXPORT ],
40 fallbacks => [ @FB_CONSTS ],
41 fallback_all => [ @FB_CONSTS, @FB_FLAGS ],
44 # Documentation moved after __END__ for speed - NI-S
46 our $ON_EBCDIC = ( ord("A") == 193 );
50 # Make a %Encoding package variable to allow a certain amount of cheating
53 require Encode::Config;
54 eval { require Encode::ConfigLocal };
59 if ( @_ and $_[0] eq ":all" ) {
60 %enc = ( %Encoding, %ExtModule );
64 for my $mod ( map { m/::/o ? $_ : "Encode::$_" } @_ ) {
66 for my $enc ( keys %ExtModule ) {
67 $ExtModule{$enc} eq $mod and $enc{$enc} = $mod;
71 return sort { lc $a cmp lc $b }
72 grep { !/^(?:Internal|Unicode|Guess)$/o } keys %enc;
76 my $obj = ref( $_[0] ) ? $_[0] : find_encoding( $_[0] );
77 $obj->can("perlio_ok") and return $obj->perlio_ok();
78 return 0; # safety net
84 $Encoding{$name} = $obj;
86 define_alias( $lc => $obj ) unless $lc eq $name;
89 define_alias( $alias, $obj );
95 my ( $class, $name, $skip_external ) = @_;
97 ref($name) && $name->can('renew') and return $name;
98 exists $Encoding{$name} and return $Encoding{$name};
100 exists $Encoding{$lc} and return $Encoding{$lc};
102 my $oc = $class->find_alias($name);
103 defined($oc) and return $oc;
104 $lc ne $name and $oc = $class->find_alias($lc);
105 defined($oc) and return $oc;
107 unless ($skip_external) {
108 if ( my $mod = $ExtModule{$name} || $ExtModule{$lc} ) {
111 eval { require $mod; };
112 exists $Encoding{$name} and return $Encoding{$name};
118 sub find_encoding($;$) {
119 my ( $name, $skip_external ) = @_;
120 return __PACKAGE__->getEncoding( $name, $skip_external );
123 sub resolve_alias($) {
124 my $obj = find_encoding(shift);
125 defined $obj and return $obj->name;
129 sub clone_encoding($) {
130 my $obj = find_encoding(shift);
132 eval { require Storable };
134 return Storable::dclone($obj);
138 my ( $name, $string, $check ) = @_;
139 return undef unless defined $string;
140 $string .= '' if ref $string; # stringify;
142 my $enc = find_encoding($name);
143 unless ( defined $enc ) {
145 Carp::croak("Unknown encoding '$name'");
147 my $octets = $enc->encode( $string, $check );
148 $_[1] = $string if $check and !ref $check and !( $check & LEAVE_SRC() );
151 *str2bytes = \&encode;
154 my ( $name, $octets, $check ) = @_;
155 return undef unless defined $octets;
156 $octets .= '' if ref $octets;
158 my $enc = find_encoding($name);
159 unless ( defined $enc ) {
161 Carp::croak("Unknown encoding '$name'");
163 my $string = $enc->decode( $octets, $check );
164 $_[1] = $octets if $check and !ref $check and !( $check & LEAVE_SRC() );
167 *bytes2str = \&decode;
170 my ( $string, $from, $to, $check ) = @_;
171 return undef unless defined $string;
173 my $f = find_encoding($from);
174 unless ( defined $f ) {
176 Carp::croak("Unknown encoding '$from'");
178 my $t = find_encoding($to);
179 unless ( defined $t ) {
181 Carp::croak("Unknown encoding '$to'");
183 my $uni = $f->decode($string);
184 $_[0] = $string = $t->encode( $uni, $check );
185 return undef if ( $check && length($uni) );
186 return defined( $_[0] ) ? length($string) : undef;
195 sub decode_utf8($;$) {
196 my ( $str, $check ) = @_;
197 return $str if is_utf8($str);
199 return decode( "utf8", $str, $check );
202 return decode( "utf8", $str );
207 predefine_encodings(1);
210 # This is to restore %Encoding if really needed;
213 sub predefine_encodings {
214 require Encode::Encoding;
215 no warnings 'redefine';
219 # was in Encode::UTF_EBCDIC
220 package Encode::UTF_EBCDIC;
221 push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding';
223 my ( $obj, $str, $chk ) = @_;
225 for ( my $i = 0 ; $i < length($str) ; $i++ ) {
228 utf8::unicode_to_native( ord( substr( $str, $i, 1 ) ) )
235 my ( $obj, $str, $chk ) = @_;
237 for ( my $i = 0 ; $i < length($str) ; $i++ ) {
240 utf8::native_to_unicode( ord( substr( $str, $i, 1 ) ) )
246 $Encode::Encoding{Unicode} =
247 bless { Name => "UTF_EBCDIC" } => "Encode::UTF_EBCDIC";
251 package Encode::Internal;
252 push @Encode::Internal::ISA, 'Encode::Encoding';
254 my ( $obj, $str, $chk ) = @_;
260 $Encode::Encoding{Unicode} =
261 bless { Name => "Internal" } => "Encode::Internal";
266 # was in Encode::utf8
267 package Encode::utf8;
268 push @Encode::utf8::ISA, 'Encode::Encoding';
272 Encode::DEBUG and warn __PACKAGE__, " XS on";
273 *decode = \&decode_xs;
274 *encode = \&encode_xs;
277 Encode::DEBUG and warn __PACKAGE__, " XS off";
279 my ( $obj, $octets, $chk ) = @_;
280 my $str = Encode::decode_utf8($octets);
281 if ( defined $str ) {
288 my ( $obj, $string, $chk ) = @_;
289 my $octets = Encode::encode_utf8($string);
294 *cat_decode = sub { # ($obj, $dst, $src, $pos, $trm, $chk)
295 # currently ignores $chk
296 my ( $obj, undef, undef, $pos, $trm ) = @_;
297 my ( $rdst, $rsrc, $rpos ) = \@_[ 1, 2, 3 ];
299 if ( ( my $npos = index( $$rsrc, $trm, $pos ) ) >= 0 ) {
301 substr( $$rsrc, $pos, $npos - $pos + length($trm) );
302 $$rpos = $npos + length($trm);
305 $$rdst .= substr( $$rsrc, $pos );
306 $$rpos = length($$rsrc);
309 $Encode::Encoding{utf8} =
310 bless { Name => "utf8" } => "Encode::utf8";
311 $Encode::Encoding{"utf-8-strict"} =
312 bless { Name => "utf-8-strict", strict_utf8 => 1 } =>
323 Encode - character encodings
329 =head2 Table of Contents
331 Encode consists of a collection of modules whose details are too big
332 to fit in one document. This POD itself explains the top-level APIs
333 and general topics at a glance. For other topics and more details,
337 --------------------------------------------------------
338 Encode::Alias Alias definitions to encodings
339 Encode::Encoding Encode Implementation Base Class
340 Encode::Supported List of Supported Encodings
341 Encode::CN Simplified Chinese Encodings
342 Encode::JP Japanese Encodings
343 Encode::KR Korean Encodings
344 Encode::TW Traditional Chinese Encodings
345 --------------------------------------------------------
349 The C<Encode> module provides the interfaces between Perl's strings
350 and the rest of the system. Perl strings are sequences of
353 The repertoire of characters that Perl can represent is at least that
354 defined by the Unicode Consortium. On most platforms the ordinal
355 values of the characters (as returned by C<ord(ch)>) is the "Unicode
356 codepoint" for the character (the exceptions are those platforms where
357 the legacy encoding is some variant of EBCDIC rather than a super-set
358 of ASCII - see L<perlebcdic>).
360 Traditionally, computer data has been moved around in 8-bit chunks
361 often called "bytes". These chunks are also known as "octets" in
362 networking standards. Perl is widely used to manipulate data of many
363 types - not only strings of characters representing human or computer
364 languages but also "binary" data being the machine's representation of
365 numbers, pixels in an image - or just about anything.
367 When Perl is processing "binary data", the programmer wants Perl to
368 process "sequences of bytes". This is not a problem for Perl - as a
369 byte has 256 possible values, it easily fits in Perl's much larger
378 I<character>: a character in the range 0..(2**32-1) (or more).
379 (What Perl's strings are made of.)
383 I<byte>: a character in the range 0..255
384 (A special case of a Perl character.)
388 I<octet>: 8 bits of data, with ordinal values 0..255
389 (Term for bytes passed to or from a non-Perl context, e.g. a disk file.)
393 =head1 PERL ENCODING API
397 =item $octets = encode(ENCODING, $string [, CHECK])
399 Encodes a string from Perl's internal form into I<ENCODING> and returns
400 a sequence of octets. ENCODING can be either a canonical name or
401 an alias. For encoding names and aliases, see L</"Defining Aliases">.
402 For CHECK, see L</"Handling Malformed Data">.
404 For example, to convert a string from Perl's internal format to
405 iso-8859-1 (also known as Latin1),
407 $octets = encode("iso-8859-1", $string);
409 B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then
410 $octets B<may not be equal to> $string. Though they both contain the
411 same data, the UTF8 flag for $octets is B<always> off. When you
412 encode anything, UTF8 flag of the result is always off, even when it
413 contains completely valid utf8 string. See L</"The UTF8 flag"> below.
415 If the $string is C<undef> then C<undef> is returned.
417 =item $string = decode(ENCODING, $octets [, CHECK])
419 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
420 internal form and returns the resulting string. As in encode(),
421 ENCODING can be either a canonical name or an alias. For encoding names
422 and aliases, see L</"Defining Aliases">. For CHECK, see
423 L</"Handling Malformed Data">.
425 For example, to convert ISO-8859-1 data to a string in Perl's internal format:
427 $string = decode("iso-8859-1", $octets);
429 B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string
430 B<may not be equal to> $octets. Though they both contain the same data,
431 the UTF8 flag for $string is on unless $octets entirely consists of
432 ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF8 flag">
435 If the $string is C<undef> then C<undef> is returned.
437 =item [$obj =] find_encoding(ENCODING)
439 Returns the I<encoding object> corresponding to ENCODING. Returns
440 undef if no matching ENCODING is find.
442 This object is what actually does the actual (en|de)coding.
444 $utf8 = decode($name, $bytes);
449 $obj = find_encoding($name);
450 croak qq(encoding "$name" not found) unless ref $obj;
454 with more error checking.
456 Therefore you can save time by reusing this object as follows;
458 my $enc = find_encoding("iso-8859-1");
460 my $utf8 = $enc->decode($_);
461 # and do someting with $utf8;
464 Besides C<< ->decode >> and C<< ->encode >>, other methods are
465 available as well. For instance, C<< -> name >> returns the canonical
466 name of the encoding object.
468 find_encoding("latin1")->name; # iso-8859-1
470 See L<Encode::Encoding> for details.
472 =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK])
474 Converts B<in-place> data between two encodings. The data in $octets
475 must be encoded as octets and not as characters in Perl's internal
476 format. For example, to convert ISO-8859-1 data to Microsoft's CP1250
479 from_to($octets, "iso-8859-1", "cp1250");
481 and to convert it back:
483 from_to($octets, "cp1250", "iso-8859-1");
485 Note that because the conversion happens in place, the data to be
486 converted cannot be a string constant; it must be a scalar variable.
488 from_to() returns the length of the converted string in octets on
489 success, I<undef> on error.
491 B<CAVEAT>: The following operations look the same but are not quite so;
493 from_to($data, "iso-8859-1", "utf8"); #1
494 $data = decode("iso-8859-1", $data); #2
496 Both #1 and #2 make $data consist of a completely valid UTF-8 string
497 but only #2 turns UTF8 flag on. #1 is equivalent to
499 $data = encode("utf8", decode("iso-8859-1", $data));
501 See L</"The UTF8 flag"> below.
505 from_to($octets, $from, $to, $check);
509 $octets = encode($to, decode($from, $octets), $check);
511 Yes, it does not respect the $check during decoding. It is
512 deliberately done that way. If you need minute control, C<decode>
513 then C<encode> as follows;
515 $octets = encode($to, decode($from, $octets, $check_from), $check_to);
517 =item $octets = encode_utf8($string);
519 Equivalent to C<$octets = encode("utf8", $string);> The characters
520 that comprise $string are encoded in Perl's internal format and the
521 result is returned as a sequence of octets. All possible
522 characters have a UTF-8 representation so this function cannot fail.
525 =item $string = decode_utf8($octets [, CHECK]);
527 equivalent to C<$string = decode("utf8", $octets [, CHECK])>.
528 The sequence of octets represented by
529 $octets is decoded from UTF-8 into a sequence of logical
530 characters. Not all sequences of octets form valid UTF-8 encodings, so
531 it is possible for this call to fail. For CHECK, see
532 L</"Handling Malformed Data">.
536 =head2 Listing available encodings
539 @list = Encode->encodings();
541 Returns a list of the canonical names of the available encodings that
542 are loaded. To get a list of all available encodings including the
543 ones that are not loaded yet, say
545 @all_encodings = Encode->encodings(":all");
547 Or you can give the name of a specific module.
549 @with_jp = Encode->encodings("Encode::JP");
551 When "::" is not in the name, "Encode::" is assumed.
553 @ebcdic = Encode->encodings("EBCDIC");
555 To find out in detail which encodings are supported by this package,
556 see L<Encode::Supported>.
558 =head2 Defining Aliases
560 To add a new alias to a given encoding, use:
564 define_alias(newName => ENCODING);
566 After that, newName can be used as an alias for ENCODING.
567 ENCODING may be either the name of an encoding or an
570 But before you do so, make sure the alias is nonexistent with
571 C<resolve_alias()>, which returns the canonical name thereof.
574 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
575 Encode::resolve_alias("iso-8859-12") # false; nonexistent
576 Encode::resolve_alias($name) eq $name # true if $name is canonical
578 resolve_alias() does not need C<use Encode::Alias>; it can be
579 exported via C<use Encode qw(resolve_alias)>.
581 See L<Encode::Alias> for details.
583 =head2 Finding IANA Character Set Registry names
585 The canonical name of a given encoding does not necessarily agree with
586 IANA IANA Character Set Registry, commonly seen as C<< Content-Type:
587 text/plain; charset=I<whatever> >>. For most cases canonical names
588 work but sometimes it does not (notably 'utf-8-strict').
590 Therefore as of Encode version 2.21, a new method C<mime_name()> is added.
593 my $enc = find_encoding('UTF-8');
594 warn $enc->name; # utf-8-strict
595 warn $enc->mime_name; # UTF-8
597 See also: L<Encode::Encoding>
599 =head1 Encoding via PerlIO
601 If your perl supports I<PerlIO> (which is the default), you can use a
602 PerlIO layer to decode and encode directly via a filehandle. The
603 following two examples are totally identical in their functionality.
606 open my $in, "<:encoding(shiftjis)", $infile or die;
607 open my $out, ">:encoding(euc-jp)", $outfile or die;
608 while(<$in>){ print $out $_; }
611 open my $in, "<", $infile or die;
612 open my $out, ">", $outfile or die;
614 from_to($_, "shiftjis", "euc-jp", 1);
618 Unfortunately, it may be that encodings are PerlIO-savvy. You can check
619 if your encoding is supported by PerlIO by calling the C<perlio_ok>
622 Encode::perlio_ok("hz"); # False
623 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
625 use Encode qw(perlio_ok); # exported upon request
628 Fortunately, all encodings that come with Encode core are PerlIO-savvy
629 except for hz and ISO-2022-kr. For gory details, see
630 L<Encode::Encoding> and L<Encode::PerlIO>.
632 =head1 Handling Malformed Data
634 The optional I<CHECK> argument tells Encode what to do when it
635 encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 )
638 As of version 2.12 Encode supports coderef values for CHECK. See below.
642 =item B<NOTE:> Not all encoding support this feature
644 Some encodings ignore I<CHECK> argument. For example,
645 L<Encode::Unicode> ignores I<CHECK> and it always croaks on error.
649 Now here is the list of I<CHECK> values available
653 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
655 If I<CHECK> is 0, (en|de)code will put a I<substitution character> in
656 place of a malformed character. When you encode, E<lt>subcharE<gt>
657 will be used. When you decode the code point C<0xFFFD> is used. If
658 the data is supposed to be UTF-8, an optional lexical warning
659 (category utf8) is given.
661 =item I<CHECK> = Encode::FB_CROAK ( == 1)
663 If I<CHECK> is 1, methods will die on error immediately with an error
664 message. Therefore, when I<CHECK> is set to 1, you should trap the
665 error with eval{} unless you really want to let it die.
667 =item I<CHECK> = Encode::FB_QUIET
669 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
670 return the portion of the data that has been processed so far when an
671 error occurs. The data argument will be overwritten with everything
672 after that point (that is, the unprocessed part of data). This is
673 handy when you have to call decode repeatedly in the case where your
674 source data may contain partial multi-byte character sequences,
675 (i.e. you are reading with a fixed-width buffer). Here is a sample
676 code that does exactly this:
678 my $buffer = ''; my $string = '';
679 while(read $fh, $buffer, 256, length($buffer)){
680 $string .= decode($encoding, $buffer, Encode::FB_QUIET);
681 # $buffer now contains the unprocessed partial character
684 =item I<CHECK> = Encode::FB_WARN
686 This is the same as above, except that it warns on error. Handy when
687 you are debugging the mode above.
689 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
691 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF)
693 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF)
695 For encodings that are implemented by Encode::XS, CHECK ==
696 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
698 When you decode, C<\xI<HH>> will be inserted for a malformed character,
699 where I<HH> is the hex representation of the octet that could not be
700 decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted,
701 where I<HHHH> is the Unicode ID of the character that cannot be found
702 in the character repertoire of the encoding.
704 HTML/XML character reference modes are about the same, in place of
705 C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and
706 XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number.
708 In Encode 2.10 or later, C<LEAVE_SRC> is also implied.
712 These modes are actually set via a bitmask. Here is how the FB_XX
713 constants are laid out. You can import the FB_XX constants via
714 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
715 constants via C<use Encode qw(:fallback_all)>.
717 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
720 RETURN_ON_ERR 0x0004 X X
730 =item Encode::LEAVE_SRC
732 If the C<Encode::LEAVE_SRC> bit is not set, but I<CHECK> is, then the second
733 argument to C<encode()> or C<decode()> may be assigned to by the functions. If
734 you're not interested in this, then bitwise-or the bitmask with it.
738 =head2 coderef for CHECK
740 As of Encode 2.12 CHECK can also be a code reference which takes the
741 ord value of unmapped caharacter as an argument and returns a string
742 that represents the fallback character. For instance,
744 $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift });
746 Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of
749 =head1 Defining Encodings
751 To define a new encoding, use:
753 use Encode qw(define_encoding);
754 define_encoding($object, 'canonicalName' [, alias...]);
756 I<canonicalName> will be associated with I<$object>. The object
757 should provide the interface described in L<Encode::Encoding>.
758 If more than two arguments are provided then additional
759 arguments are taken as aliases for I<$object>.
761 See L<Encode::Encoding> for more details.
765 Before the introduction of Unicode support in perl, The C<eq> operator
766 just compared the strings represented by two scalars. Beginning with
767 perl 5.8, C<eq> compares two strings with simultaneous consideration of
768 I<the UTF8 flag>. To explain why we made it so, I will quote page 402 of
769 C<Programming Perl, 3rd ed.>
775 Old byte-oriented programs should not spontaneously break on the old
776 byte-oriented data they used to work on.
780 Old byte-oriented programs should magically start working on the new
781 character-oriented data when appropriate.
785 Programs should run just as fast in the new character-oriented mode
786 as in the old byte-oriented mode.
790 Perl should remain one language, rather than forking into a
791 byte-oriented Perl and a character-oriented Perl.
795 Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0
796 was born and many features documented in the book remained
797 unimplemented for a long time. Perl 5.8 corrected this and the introduction
798 of the UTF8 flag is one of them. You can think of this perl notion as of a
799 byte-oriented mode (UTF8 flag off) and a character-oriented mode (UTF8
802 Here is how Encode takes care of the UTF8 flag.
808 When you encode, the resulting UTF8 flag is always off.
812 When you decode, the resulting UTF8 flag is on unless you can
813 unambiguously represent data. Here is the definition of
816 After C<$utf8 = decode('foo', $octet);>,
818 When $octet is... The UTF8 flag in $utf8 is
819 ---------------------------------------------
820 In ASCII only (or EBCDIC only) OFF
822 In any other Encoding ON
823 ---------------------------------------------
825 As you see, there is one exception, In ASCII. That way you can assume
826 Goal #1. And with Encode Goal #2 is assumed but you still have to be
827 careful in such cases mentioned in B<CAVEAT> paragraphs.
829 This UTF8 flag is not visible in perl scripts, exactly for the same
830 reason you cannot (or you I<don't have to>) see if a scalar contains a
831 string, integer, or floating point number. But you can still peek
832 and poke these if you will. See the section below.
836 =head2 Messing with Perl's Internals
838 The following API uses parts of Perl's internals in the current
839 implementation. As such, they are efficient but may change.
843 =item is_utf8(STRING [, CHECK])
845 [INTERNAL] Tests whether the UTF8 flag is turned on in the STRING.
846 If CHECK is true, also checks the data in STRING for being well-formed
847 UTF-8. Returns true if successful, false otherwise.
849 As of perl 5.8.1, L<utf8> also has utf8::is_utf8().
851 =item _utf8_on(STRING)
853 [INTERNAL] Turns on the UTF8 flag in STRING. The data in STRING is
854 B<not> checked for being well-formed UTF-8. Do not use unless you
855 B<know> that the STRING is well-formed UTF-8. Returns the previous
856 state of the UTF8 flag (so please don't treat the return value as
857 indicating success or failure), or C<undef> if STRING is not a string.
859 =item _utf8_off(STRING)
861 [INTERNAL] Turns off the UTF8 flag in STRING. Do not use frivolously.
862 Returns the previous state of the UTF8 flag (so please don't treat the
863 return value as indicating success or failure), or C<undef> if STRING is
868 =head1 UTF-8 vs. utf8 vs. UTF8
870 ....We now view strings not as sequences of bytes, but as sequences
871 of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit
872 computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed.
874 That has been the perl's notion of UTF-8 but official UTF-8 is more
875 strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are
876 not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al).
878 Now that is overruled by Larry Wall himself.
880 From: Larry Wall <larry@wall.org>
881 Date: December 04, 2004 11:51:58 JST
882 To: perl-unicode@perl.org
883 Subject: Re: Make Encode.pm support the real UTF-8
884 Message-Id: <20041204025158.GA28754@wall.org>
886 On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote:
887 : I've no problem with 'utf8' being perl's unrestricted uft8 encoding,
888 : but "UTF-8" is the name of the standard and should give the
889 : corresponding behaviour.
891 For what it's worth, that's how I've always kept them straight in my
894 Also for what it's worth, Perl 6 will mostly default to strict but
895 make it easy to switch back to lax.
899 Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8
900 while B<utf8> means liberal, lax, version thereof. And Encode version
901 2.10 or later thus groks the difference between C<UTF-8> and C"utf8".
903 encode("utf8", "\x{FFFF_FFFF}", 1); # okay
904 encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks
906 C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>.
907 Yes, the hyphen between "UTF" and "8" is important. Without it Encode
910 find_encoding("UTF-8")->name # is 'utf-8-strict'
911 find_encoding("utf-8")->name # ditto. names are case insensitive
912 find_encoding("utf_8")->name # ditto. "_" are treated as "-"
913 find_encoding("UTF8")->name # is 'utf8'.
915 The UTF8 flag is internally called UTF8, without a hyphen. It indicates
916 whether a string is internally encoded as utf8, also without a hypen.
921 L<Encode::Supported>,
926 L<perlunicode>, L<perluniintro>, L<perlunifaq>, L<perlunitut>
928 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
932 This project was originated by Nick Ing-Simmons and later maintained
933 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full
934 list of people involved. For any questions, use
935 E<lt>perl-unicode@perl.orgE<gt> so we can all share.
937 While Dan Kogai retains the copyright as a maintainer, the credit
938 should go to all those involoved. See AUTHORS for those submitted
943 Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt>
945 This library is free software; you can redistribute it and/or modify
946 it under the same terms as Perl itself.