2 # $Id: Encode.pm,v 2.42 2010/12/31 22:48:10 dankogai Exp $
7 our $VERSION = sprintf "%d.%02d", q$Revision: 2.42 $ =~ /(\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;
55 # https://bugzilla.redhat.com/show_bug.cgi?id=435505#c2
56 # to find why sig handers inside eval{} are disabled.
60 require Encode::ConfigLocal;
66 if ( @_ and $_[0] eq ":all" ) {
67 %enc = ( %Encoding, %ExtModule );
71 for my $mod ( map { m/::/o ? $_ : "Encode::$_" } @_ ) {
73 for my $enc ( keys %ExtModule ) {
74 $ExtModule{$enc} eq $mod and $enc{$enc} = $mod;
78 return sort { lc $a cmp lc $b }
79 grep { !/^(?:Internal|Unicode|Guess)$/o } keys %enc;
83 my $obj = ref( $_[0] ) ? $_[0] : find_encoding( $_[0] );
84 $obj->can("perlio_ok") and return $obj->perlio_ok();
85 return 0; # safety net
91 $Encoding{$name} = $obj;
93 define_alias( $lc => $obj ) unless $lc eq $name;
96 define_alias( $alias, $obj );
102 my ( $class, $name, $skip_external ) = @_;
104 ref($name) && $name->can('renew') and return $name;
105 exists $Encoding{$name} and return $Encoding{$name};
107 exists $Encoding{$lc} and return $Encoding{$lc};
109 my $oc = $class->find_alias($name);
110 defined($oc) and return $oc;
111 $lc ne $name and $oc = $class->find_alias($lc);
112 defined($oc) and return $oc;
114 unless ($skip_external) {
115 if ( my $mod = $ExtModule{$name} || $ExtModule{$lc} ) {
118 eval { require $mod; };
119 exists $Encoding{$name} and return $Encoding{$name};
125 sub find_encoding($;$) {
126 my ( $name, $skip_external ) = @_;
127 return __PACKAGE__->getEncoding( $name, $skip_external );
130 sub resolve_alias($) {
131 my $obj = find_encoding(shift);
132 defined $obj and return $obj->name;
136 sub clone_encoding($) {
137 my $obj = find_encoding(shift);
139 eval { require Storable };
141 return Storable::dclone($obj);
145 my ( $name, $string, $check ) = @_;
146 return undef unless defined $string;
147 $string .= '' if ref $string; # stringify;
149 unless ( defined $name ) {
151 Carp::croak("Encoding name should not be undef");
153 my $enc = find_encoding($name);
154 unless ( defined $enc ) {
156 Carp::croak("Unknown encoding '$name'");
158 my $octets = $enc->encode( $string, $check );
159 $_[1] = $string if $check and !ref $check and !( $check & LEAVE_SRC() );
162 *str2bytes = \&encode;
165 my ( $name, $octets, $check ) = @_;
166 return undef unless defined $octets;
167 $octets .= '' if ref $octets;
169 my $enc = find_encoding($name);
170 unless ( defined $enc ) {
172 Carp::croak("Unknown encoding '$name'");
174 my $string = $enc->decode( $octets, $check );
175 $_[1] = $octets if $check and !ref $check and !( $check & LEAVE_SRC() );
178 *bytes2str = \&decode;
181 my ( $string, $from, $to, $check ) = @_;
182 return undef unless defined $string;
184 my $f = find_encoding($from);
185 unless ( defined $f ) {
187 Carp::croak("Unknown encoding '$from'");
189 my $t = find_encoding($to);
190 unless ( defined $t ) {
192 Carp::croak("Unknown encoding '$to'");
194 my $uni = $f->decode($string);
195 $_[0] = $string = $t->encode( $uni, $check );
196 return undef if ( $check && length($uni) );
197 return defined( $_[0] ) ? length($string) : undef;
208 sub decode_utf8($;$) {
209 my ( $octets, $check ) = @_;
210 return $octets if is_utf8($octets);
211 return undef unless defined $octets;
212 $octets .= '' if ref $octets;
214 $utf8enc ||= find_encoding('utf8');
215 my $string = $utf8enc->decode( $octets, $check );
216 $_[0] = $octets if $check and !ref $check and !( $check & LEAVE_SRC() );
220 # sub decode_utf8($;$) {
221 # my ( $str, $check ) = @_;
222 # return $str if is_utf8($str);
224 # return decode( "utf8", $str, $check );
227 # return decode( "utf8", $str );
232 predefine_encodings(1);
235 # This is to restore %Encoding if really needed;
238 sub predefine_encodings {
239 require Encode::Encoding;
240 no warnings 'redefine';
244 # was in Encode::UTF_EBCDIC
245 package Encode::UTF_EBCDIC;
246 push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding';
248 my ( $obj, $str, $chk ) = @_;
250 for ( my $i = 0 ; $i < length($str) ; $i++ ) {
253 utf8::unicode_to_native( ord( substr( $str, $i, 1 ) ) )
260 my ( $obj, $str, $chk ) = @_;
262 for ( my $i = 0 ; $i < length($str) ; $i++ ) {
265 utf8::native_to_unicode( ord( substr( $str, $i, 1 ) ) )
271 $Encode::Encoding{Unicode} =
272 bless { Name => "UTF_EBCDIC" } => "Encode::UTF_EBCDIC";
276 package Encode::Internal;
277 push @Encode::Internal::ISA, 'Encode::Encoding';
279 my ( $obj, $str, $chk ) = @_;
285 $Encode::Encoding{Unicode} =
286 bless { Name => "Internal" } => "Encode::Internal";
291 # was in Encode::utf8
292 package Encode::utf8;
293 push @Encode::utf8::ISA, 'Encode::Encoding';
297 Encode::DEBUG and warn __PACKAGE__, " XS on";
298 *decode = \&decode_xs;
299 *encode = \&encode_xs;
302 Encode::DEBUG and warn __PACKAGE__, " XS off";
304 my ( $obj, $octets, $chk ) = @_;
305 my $str = Encode::decode_utf8($octets);
306 if ( defined $str ) {
313 my ( $obj, $string, $chk ) = @_;
314 my $octets = Encode::encode_utf8($string);
319 *cat_decode = sub { # ($obj, $dst, $src, $pos, $trm, $chk)
320 # currently ignores $chk
321 my ( $obj, undef, undef, $pos, $trm ) = @_;
322 my ( $rdst, $rsrc, $rpos ) = \@_[ 1, 2, 3 ];
324 if ( ( my $npos = index( $$rsrc, $trm, $pos ) ) >= 0 ) {
326 substr( $$rsrc, $pos, $npos - $pos + length($trm) );
327 $$rpos = $npos + length($trm);
330 $$rdst .= substr( $$rsrc, $pos );
331 $$rpos = length($$rsrc);
334 $Encode::Encoding{utf8} =
335 bless { Name => "utf8" } => "Encode::utf8";
336 $Encode::Encoding{"utf-8-strict"} =
337 bless { Name => "utf-8-strict", strict_utf8 => 1 } =>
348 Encode - character encodings
354 =head2 Table of Contents
356 Encode consists of a collection of modules whose details are too big
357 to fit in one document. This POD itself explains the top-level APIs
358 and general topics at a glance. For other topics and more details,
362 --------------------------------------------------------
363 Encode::Alias Alias definitions to encodings
364 Encode::Encoding Encode Implementation Base Class
365 Encode::Supported List of Supported Encodings
366 Encode::CN Simplified Chinese Encodings
367 Encode::JP Japanese Encodings
368 Encode::KR Korean Encodings
369 Encode::TW Traditional Chinese Encodings
370 --------------------------------------------------------
374 The C<Encode> module provides the interfaces between Perl's strings
375 and the rest of the system. Perl strings are sequences of
378 The repertoire of characters that Perl can represent is at least that
379 defined by the Unicode Consortium. On most platforms the ordinal
380 values of the characters (as returned by C<ord(ch)>) is the "Unicode
381 codepoint" for the character (the exceptions are those platforms where
382 the legacy encoding is some variant of EBCDIC rather than a super-set
383 of ASCII - see L<perlebcdic>).
385 Traditionally, computer data has been moved around in 8-bit chunks
386 often called "bytes". These chunks are also known as "octets" in
387 networking standards. Perl is widely used to manipulate data of many
388 types - not only strings of characters representing human or computer
389 languages but also "binary" data being the machine's representation of
390 numbers, pixels in an image - or just about anything.
392 When Perl is processing "binary data", the programmer wants Perl to
393 process "sequences of bytes". This is not a problem for Perl - as a
394 byte has 256 possible values, it easily fits in Perl's much larger
403 I<character>: a character in the range 0..(2**32-1) (or more).
404 (What Perl's strings are made of.)
408 I<byte>: a character in the range 0..255
409 (A special case of a Perl character.)
413 I<octet>: 8 bits of data, with ordinal values 0..255
414 (Term for bytes passed to or from a non-Perl context, e.g. a disk file.)
418 =head1 PERL ENCODING API
422 =item $octets = encode(ENCODING, $string [, CHECK])
424 Encodes a string from Perl's internal form into I<ENCODING> and returns
425 a sequence of octets. ENCODING can be either a canonical name or
426 an alias. For encoding names and aliases, see L</"Defining Aliases">.
427 For CHECK, see L</"Handling Malformed Data">.
429 For example, to convert a string from Perl's internal format to
430 iso-8859-1 (also known as Latin1),
432 $octets = encode("iso-8859-1", $string);
434 B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then
435 $octets B<may not be equal to> $string. Though they both contain the
436 same data, the UTF8 flag for $octets is B<always> off. When you
437 encode anything, UTF8 flag of the result is always off, even when it
438 contains completely valid utf8 string. See L</"The UTF8 flag"> below.
440 If the $string is C<undef> then C<undef> is returned.
442 =item $string = decode(ENCODING, $octets [, CHECK])
444 Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's
445 internal form and returns the resulting string. As in encode(),
446 ENCODING can be either a canonical name or an alias. For encoding names
447 and aliases, see L</"Defining Aliases">. For CHECK, see
448 L</"Handling Malformed Data">.
450 For example, to convert ISO-8859-1 data to a string in Perl's internal format:
452 $string = decode("iso-8859-1", $octets);
454 B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string
455 B<may not be equal to> $octets. Though they both contain the same data,
456 the UTF8 flag for $string is on unless $octets entirely consists of
457 ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF8 flag">
460 If the $string is C<undef> then C<undef> is returned.
462 =item [$obj =] find_encoding(ENCODING)
464 Returns the I<encoding object> corresponding to ENCODING. Returns
465 undef if no matching ENCODING is find.
467 This object is what actually does the actual (en|de)coding.
469 $utf8 = decode($name, $bytes);
474 $obj = find_encoding($name);
475 croak qq(encoding "$name" not found) unless ref $obj;
479 with more error checking.
481 Therefore you can save time by reusing this object as follows;
483 my $enc = find_encoding("iso-8859-1");
485 my $utf8 = $enc->decode($_);
486 # and do someting with $utf8;
489 Besides C<< ->decode >> and C<< ->encode >>, other methods are
490 available as well. For instance, C<< -> name >> returns the canonical
491 name of the encoding object.
493 find_encoding("latin1")->name; # iso-8859-1
495 See L<Encode::Encoding> for details.
497 =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK])
499 Converts B<in-place> data between two encodings. The data in $octets
500 must be encoded as octets and not as characters in Perl's internal
501 format. For example, to convert ISO-8859-1 data to Microsoft's CP1250
504 from_to($octets, "iso-8859-1", "cp1250");
506 and to convert it back:
508 from_to($octets, "cp1250", "iso-8859-1");
510 Note that because the conversion happens in place, the data to be
511 converted cannot be a string constant; it must be a scalar variable.
513 from_to() returns the length of the converted string in octets on
514 success, I<undef> on error.
516 B<CAVEAT>: The following operations look the same but are not quite so;
518 from_to($data, "iso-8859-1", "utf8"); #1
519 $data = decode("iso-8859-1", $data); #2
521 Both #1 and #2 make $data consist of a completely valid UTF-8 string
522 but only #2 turns UTF8 flag on. #1 is equivalent to
524 $data = encode("utf8", decode("iso-8859-1", $data));
526 See L</"The UTF8 flag"> below.
530 from_to($octets, $from, $to, $check);
534 $octets = encode($to, decode($from, $octets), $check);
536 Yes, it does not respect the $check during decoding. It is
537 deliberately done that way. If you need minute control, C<decode>
538 then C<encode> as follows;
540 $octets = encode($to, decode($from, $octets, $check_from), $check_to);
542 =item $octets = encode_utf8($string);
544 Equivalent to C<$octets = encode("utf8", $string);> The characters
545 that comprise $string are encoded in Perl's internal format and the
546 result is returned as a sequence of octets. All possible
547 characters have a UTF-8 representation so this function cannot fail.
550 =item $string = decode_utf8($octets [, CHECK]);
552 equivalent to C<$string = decode("utf8", $octets [, CHECK])>.
553 The sequence of octets represented by
554 $octets is decoded from UTF-8 into a sequence of logical
555 characters. Not all sequences of octets form valid UTF-8 encodings, so
556 it is possible for this call to fail. For CHECK, see
557 L</"Handling Malformed Data">.
561 =head2 Listing available encodings
564 @list = Encode->encodings();
566 Returns a list of the canonical names of the available encodings that
567 are loaded. To get a list of all available encodings including the
568 ones that are not loaded yet, say
570 @all_encodings = Encode->encodings(":all");
572 Or you can give the name of a specific module.
574 @with_jp = Encode->encodings("Encode::JP");
576 When "::" is not in the name, "Encode::" is assumed.
578 @ebcdic = Encode->encodings("EBCDIC");
580 To find out in detail which encodings are supported by this package,
581 see L<Encode::Supported>.
583 =head2 Defining Aliases
585 To add a new alias to a given encoding, use:
589 define_alias(newName => ENCODING);
591 After that, newName can be used as an alias for ENCODING.
592 ENCODING may be either the name of an encoding or an
595 But before you do so, make sure the alias is nonexistent with
596 C<resolve_alias()>, which returns the canonical name thereof.
599 Encode::resolve_alias("latin1") eq "iso-8859-1" # true
600 Encode::resolve_alias("iso-8859-12") # false; nonexistent
601 Encode::resolve_alias($name) eq $name # true if $name is canonical
603 resolve_alias() does not need C<use Encode::Alias>; it can be
604 exported via C<use Encode qw(resolve_alias)>.
606 See L<Encode::Alias> for details.
608 =head2 Finding IANA Character Set Registry names
610 The canonical name of a given encoding does not necessarily agree with
611 IANA IANA Character Set Registry, commonly seen as C<< Content-Type:
612 text/plain; charset=I<whatever> >>. For most cases canonical names
613 work but sometimes it does not (notably 'utf-8-strict').
615 Therefore as of Encode version 2.21, a new method C<mime_name()> is added.
618 my $enc = find_encoding('UTF-8');
619 warn $enc->name; # utf-8-strict
620 warn $enc->mime_name; # UTF-8
622 See also: L<Encode::Encoding>
624 =head1 Encoding via PerlIO
626 If your perl supports I<PerlIO> (which is the default), you can use a
627 PerlIO layer to decode and encode directly via a filehandle. The
628 following two examples are totally identical in their functionality.
631 open my $in, "<:encoding(shiftjis)", $infile or die;
632 open my $out, ">:encoding(euc-jp)", $outfile or die;
633 while(<$in>){ print $out $_; }
636 open my $in, "<", $infile or die;
637 open my $out, ">", $outfile or die;
639 from_to($_, "shiftjis", "euc-jp", 1);
643 Unfortunately, it may be that encodings are PerlIO-savvy. You can check
644 if your encoding is supported by PerlIO by calling the C<perlio_ok>
647 Encode::perlio_ok("hz"); # False
648 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available
650 use Encode qw(perlio_ok); # exported upon request
653 Fortunately, all encodings that come with Encode core are PerlIO-savvy
654 except for hz and ISO-2022-kr. For gory details, see
655 L<Encode::Encoding> and L<Encode::PerlIO>.
657 =head1 Handling Malformed Data
659 The optional I<CHECK> argument tells Encode what to do when it
660 encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 )
663 As of version 2.12 Encode supports coderef values for CHECK. See below.
667 =item B<NOTE:> Not all encoding support this feature
669 Some encodings ignore I<CHECK> argument. For example,
670 L<Encode::Unicode> ignores I<CHECK> and it always croaks on error.
674 Now here is the list of I<CHECK> values available
678 =item I<CHECK> = Encode::FB_DEFAULT ( == 0)
680 If I<CHECK> is 0, (en|de)code will put a I<substitution character> in
681 place of a malformed character. When you encode, E<lt>subcharE<gt>
682 will be used. When you decode the code point C<0xFFFD> is used. If
683 the data is supposed to be UTF-8, an optional lexical warning
684 (category utf8) is given.
686 =item I<CHECK> = Encode::FB_CROAK ( == 1)
688 If I<CHECK> is 1, methods will die on error immediately with an error
689 message. Therefore, when I<CHECK> is set to 1, you should trap the
690 error with eval{} unless you really want to let it die.
692 =item I<CHECK> = Encode::FB_QUIET
694 If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately
695 return the portion of the data that has been processed so far when an
696 error occurs. The data argument will be overwritten with everything
697 after that point (that is, the unprocessed part of data). This is
698 handy when you have to call decode repeatedly in the case where your
699 source data may contain partial multi-byte character sequences,
700 (i.e. you are reading with a fixed-width buffer). Here is a sample
701 code that does exactly this:
703 my $buffer = ''; my $string = '';
704 while(read $fh, $buffer, 256, length($buffer)){
705 $string .= decode($encoding, $buffer, Encode::FB_QUIET);
706 # $buffer now contains the unprocessed partial character
709 =item I<CHECK> = Encode::FB_WARN
711 This is the same as above, except that it warns on error. Handy when
712 you are debugging the mode above.
714 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ)
716 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF)
718 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF)
720 For encodings that are implemented by Encode::XS, CHECK ==
721 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode.
723 When you decode, C<\xI<HH>> will be inserted for a malformed character,
724 where I<HH> is the hex representation of the octet that could not be
725 decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted,
726 where I<HHHH> is the Unicode ID of the character that cannot be found
727 in the character repertoire of the encoding.
729 HTML/XML character reference modes are about the same, in place of
730 C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and
731 XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number.
733 In Encode 2.10 or later, C<LEAVE_SRC> is also implied.
737 These modes are actually set via a bitmask. Here is how the FB_XX
738 constants are laid out. You can import the FB_XX constants via
739 C<use Encode qw(:fallbacks)>; you can import the generic bitmask
740 constants via C<use Encode qw(:fallback_all)>.
742 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ
745 RETURN_ON_ERR 0x0004 X X
755 =item Encode::LEAVE_SRC
757 If the C<Encode::LEAVE_SRC> bit is not set, but I<CHECK> is, then the second
758 argument to C<encode()> or C<decode()> may be assigned to by the functions. If
759 you're not interested in this, then bitwise-or the bitmask with it.
763 =head2 coderef for CHECK
765 As of Encode 2.12 CHECK can also be a code reference which takes the
766 ord value of unmapped caharacter as an argument and returns a string
767 that represents the fallback character. For instance,
769 $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift });
771 Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of
774 =head1 Defining Encodings
776 To define a new encoding, use:
778 use Encode qw(define_encoding);
779 define_encoding($object, 'canonicalName' [, alias...]);
781 I<canonicalName> will be associated with I<$object>. The object
782 should provide the interface described in L<Encode::Encoding>.
783 If more than two arguments are provided then additional
784 arguments are taken as aliases for I<$object>.
786 See L<Encode::Encoding> for more details.
790 Before the introduction of Unicode support in perl, The C<eq> operator
791 just compared the strings represented by two scalars. Beginning with
792 perl 5.8, C<eq> compares two strings with simultaneous consideration of
793 I<the UTF8 flag>. To explain why we made it so, I will quote page 402 of
794 C<Programming Perl, 3rd ed.>
800 Old byte-oriented programs should not spontaneously break on the old
801 byte-oriented data they used to work on.
805 Old byte-oriented programs should magically start working on the new
806 character-oriented data when appropriate.
810 Programs should run just as fast in the new character-oriented mode
811 as in the old byte-oriented mode.
815 Perl should remain one language, rather than forking into a
816 byte-oriented Perl and a character-oriented Perl.
820 Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0
821 was born and many features documented in the book remained
822 unimplemented for a long time. Perl 5.8 corrected this and the introduction
823 of the UTF8 flag is one of them. You can think of this perl notion as of a
824 byte-oriented mode (UTF8 flag off) and a character-oriented mode (UTF8
827 Here is how Encode takes care of the UTF8 flag.
833 When you encode, the resulting UTF8 flag is always off.
837 When you decode, the resulting UTF8 flag is on unless you can
838 unambiguously represent data. Here is the definition of
841 After C<$utf8 = decode('foo', $octet);>,
843 When $octet is... The UTF8 flag in $utf8 is
844 ---------------------------------------------
845 In ASCII only (or EBCDIC only) OFF
847 In any other Encoding ON
848 ---------------------------------------------
850 As you see, there is one exception, In ASCII. That way you can assume
851 Goal #1. And with Encode Goal #2 is assumed but you still have to be
852 careful in such cases mentioned in B<CAVEAT> paragraphs.
854 This UTF8 flag is not visible in perl scripts, exactly for the same
855 reason you cannot (or you I<don't have to>) see if a scalar contains a
856 string, integer, or floating point number. But you can still peek
857 and poke these if you will. See the section below.
861 =head2 Messing with Perl's Internals
863 The following API uses parts of Perl's internals in the current
864 implementation. As such, they are efficient but may change.
868 =item is_utf8(STRING [, CHECK])
870 [INTERNAL] Tests whether the UTF8 flag is turned on in the STRING.
871 If CHECK is true, also checks the data in STRING for being well-formed
872 UTF-8. Returns true if successful, false otherwise.
874 As of perl 5.8.1, L<utf8> also has utf8::is_utf8().
876 =item _utf8_on(STRING)
878 [INTERNAL] Turns on the UTF8 flag in STRING. The data in STRING is
879 B<not> checked for being well-formed UTF-8. Do not use unless you
880 B<know> that the STRING is well-formed UTF-8. Returns the previous
881 state of the UTF8 flag (so please don't treat the return value as
882 indicating success or failure), or C<undef> if STRING is not a string.
884 This function does not work on tainted values.
886 =item _utf8_off(STRING)
888 [INTERNAL] Turns off the UTF8 flag in STRING. Do not use frivolously.
889 Returns the previous state of the UTF8 flag (so please don't treat the
890 return value as indicating success or failure), or C<undef> if STRING is
893 This function does not work on tainted values.
897 =head1 UTF-8 vs. utf8 vs. UTF8
899 ....We now view strings not as sequences of bytes, but as sequences
900 of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit
901 computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed.
903 That has been the perl's notion of UTF-8 but official UTF-8 is more
904 strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are
905 not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al).
907 Now that is overruled by Larry Wall himself.
909 From: Larry Wall <larry@wall.org>
910 Date: December 04, 2004 11:51:58 JST
911 To: perl-unicode@perl.org
912 Subject: Re: Make Encode.pm support the real UTF-8
913 Message-Id: <20041204025158.GA28754@wall.org>
915 On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote:
916 : I've no problem with 'utf8' being perl's unrestricted uft8 encoding,
917 : but "UTF-8" is the name of the standard and should give the
918 : corresponding behaviour.
920 For what it's worth, that's how I've always kept them straight in my
923 Also for what it's worth, Perl 6 will mostly default to strict but
924 make it easy to switch back to lax.
928 Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8
929 while B<utf8> means liberal, lax, version thereof. And Encode version
930 2.10 or later thus groks the difference between C<UTF-8> and C"utf8".
932 encode("utf8", "\x{FFFF_FFFF}", 1); # okay
933 encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks
935 C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>.
936 Yes, the hyphen between "UTF" and "8" is important. Without it Encode
939 find_encoding("UTF-8")->name # is 'utf-8-strict'
940 find_encoding("utf-8")->name # ditto. names are case insensitive
941 find_encoding("utf_8")->name # ditto. "_" are treated as "-"
942 find_encoding("UTF8")->name # is 'utf8'.
944 The UTF8 flag is internally called UTF8, without a hyphen. It indicates
945 whether a string is internally encoded as utf8, also without a hypen.
950 L<Encode::Supported>,
955 L<perlunicode>, L<perluniintro>, L<perlunifaq>, L<perlunitut>
957 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt>
961 This project was originated by Nick Ing-Simmons and later maintained
962 by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full
963 list of people involved. For any questions, use
964 E<lt>perl-unicode@perl.orgE<gt> so we can all share.
966 While Dan Kogai retains the copyright as a maintainer, the credit
967 should go to all those involoved. See AUTHORS for those submitted
972 Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt>
974 This library is free software; you can redistribute it and/or modify
975 it under the same terms as Perl itself.