| [ Index ] |
PHP Cross Reference of Unnamed Project |
[Summary view] [Print] [Text view]
1 # 2 # $Id: Encode.pm,v 2.23 2007/05/29 18:15:32 dankogai Exp dankogai $ 3 # 4 package Encode; 5 use strict; 6 use warnings; 7 our $VERSION = sprintf "%d.%02d", q$Revision: 2.23 $ =~ /(\d+)/g; 8 sub DEBUG () { 0 } 9 use XSLoader (); 10 XSLoader::load( __PACKAGE__, $VERSION ); 11 12 require Exporter; 13 use base qw/Exporter/; 14 15 # Public, encouraged API is exported by default 16 17 our @EXPORT = qw( 18 decode decode_utf8 encode encode_utf8 str2bytes bytes2str 19 encodings find_encoding clone_encoding 20 ); 21 our @FB_FLAGS = qw( 22 DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC 23 PERLQQ HTMLCREF XMLCREF STOP_AT_PARTIAL 24 ); 25 our @FB_CONSTS = qw( 26 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN 27 FB_PERLQQ FB_HTMLCREF FB_XMLCREF 28 ); 29 our @EXPORT_OK = ( 30 qw( 31 _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit 32 is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade 33 ), 34 @FB_FLAGS, @FB_CONSTS, 35 ); 36 37 our %EXPORT_TAGS = ( 38 all => [ @EXPORT, @EXPORT_OK ], 39 fallbacks => [@FB_CONSTS], 40 fallback_all => [ @FB_CONSTS, @FB_FLAGS ], 41 ); 42 43 # Documentation moved after __END__ for speed - NI-S 44 45 our $ON_EBCDIC = ( ord("A") == 193 ); 46 47 use Encode::Alias; 48 49 # Make a %Encoding package variable to allow a certain amount of cheating 50 our %Encoding; 51 our %ExtModule; 52 require Encode::Config; 53 eval { require Encode::ConfigLocal }; 54 55 sub encodings { 56 my $class = shift; 57 my %enc; 58 if ( @_ and $_[0] eq ":all" ) { 59 %enc = ( %Encoding, %ExtModule ); 60 } 61 else { 62 %enc = %Encoding; 63 for my $mod ( map { m/::/o ? $_ : "Encode::$_" } @_ ) { 64 DEBUG and warn $mod; 65 for my $enc ( keys %ExtModule ) { 66 $ExtModule{$enc} eq $mod and $enc{$enc} = $mod; 67 } 68 } 69 } 70 return sort { lc $a cmp lc $b } 71 grep { !/^(?:Internal|Unicode|Guess)$/o } keys %enc; 72 } 73 74 sub perlio_ok { 75 my $obj = ref( $_[0] ) ? $_[0] : find_encoding( $_[0] ); 76 $obj->can("perlio_ok") and return $obj->perlio_ok(); 77 return 0; # safety net 78 } 79 80 sub define_encoding { 81 my $obj = shift; 82 my $name = shift; 83 $Encoding{$name} = $obj; 84 my $lc = lc($name); 85 define_alias( $lc => $obj ) unless $lc eq $name; 86 while (@_) { 87 my $alias = shift; 88 define_alias( $alias, $obj ); 89 } 90 return $obj; 91 } 92 93 sub getEncoding { 94 my ( $class, $name, $skip_external ) = @_; 95 96 ref($name) && $name->can('renew') and return $name; 97 exists $Encoding{$name} and return $Encoding{$name}; 98 my $lc = lc $name; 99 exists $Encoding{$lc} and return $Encoding{$lc}; 100 101 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 $oc; 105 106 unless ($skip_external) { 107 if ( my $mod = $ExtModule{$name} || $ExtModule{$lc} ) { 108 $mod =~ s,::,/,g; 109 $mod .= '.pm'; 110 eval { require $mod; }; 111 exists $Encoding{$name} and return $Encoding{$name}; 112 } 113 } 114 return; 115 } 116 117 sub find_encoding($;$) { 118 my ( $name, $skip_external ) = @_; 119 return __PACKAGE__->getEncoding( $name, $skip_external ); 120 } 121 122 sub resolve_alias($) { 123 my $obj = find_encoding(shift); 124 defined $obj and return $obj->name; 125 return; 126 } 127 128 sub clone_encoding($) { 129 my $obj = find_encoding(shift); 130 ref $obj or return; 131 eval { require Storable }; 132 $@ and return; 133 return Storable::dclone($obj); 134 } 135 136 sub encode($$;$) { 137 my ( $name, $string, $check ) = @_; 138 return undef unless defined $string; 139 $string .= '' if ref $string; # stringify; 140 $check ||= 0; 141 my $enc = find_encoding($name); 142 unless ( defined $enc ) { 143 require Carp; 144 Carp::croak("Unknown encoding '$name'"); 145 } 146 my $octets = $enc->encode( $string, $check ); 147 $_[1] = $string if $check and !ref $check and !( $check & LEAVE_SRC() ); 148 return $octets; 149 } 150 *str2bytes = \&encode; 151 152 sub decode($$;$) { 153 my ( $name, $octets, $check ) = @_; 154 return undef unless defined $octets; 155 $octets .= '' if ref $octets; 156 $check ||= 0; 157 my $enc = find_encoding($name); 158 unless ( defined $enc ) { 159 require Carp; 160 Carp::croak("Unknown encoding '$name'"); 161 } 162 my $string = $enc->decode( $octets, $check ); 163 $_[1] = $octets if $check and !ref $check and !( $check & LEAVE_SRC() ); 164 return $string; 165 } 166 *bytes2str = \&decode; 167 168 sub from_to($$$;$) { 169 my ( $string, $from, $to, $check ) = @_; 170 return undef unless defined $string; 171 $check ||= 0; 172 my $f = find_encoding($from); 173 unless ( defined $f ) { 174 require Carp; 175 Carp::croak("Unknown encoding '$from'"); 176 } 177 my $t = find_encoding($to); 178 unless ( defined $t ) { 179 require Carp; 180 Carp::croak("Unknown encoding '$to'"); 181 } 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; 186 } 187 188 sub encode_utf8($) { 189 my ($str) = @_; 190 utf8::encode($str); 191 return $str; 192 } 193 194 sub decode_utf8($;$) { 195 my ( $str, $check ) = @_; 196 return $str if is_utf8($str); 197 if ($check) { 198 return decode( "utf8", $str, $check ); 199 } 200 else { 201 return decode( "utf8", $str ); 202 return $str; 203 } 204 } 205 206 predefine_encodings(1); 207 208 # 209 # This is to restore %Encoding if really needed; 210 # 211 212 sub predefine_encodings { 213 require Encode::Encoding; 214 no warnings 'redefine'; 215 my $use_xs = shift; 216 if ($ON_EBCDIC) { 217 218 # was in Encode::UTF_EBCDIC 219 package Encode::UTF_EBCDIC; 220 push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; 221 *decode = sub { 222 my ( $obj, $str, $chk ) = @_; 223 my $res = ''; 224 for ( my $i = 0 ; $i < length($str) ; $i++ ) { 225 $res .= 226 chr( 227 utf8::unicode_to_native( ord( substr( $str, $i, 1 ) ) ) 228 ); 229 } 230 $_[1] = '' if $chk; 231 return $res; 232 }; 233 *encode = sub { 234 my ( $obj, $str, $chk ) = @_; 235 my $res = ''; 236 for ( my $i = 0 ; $i < length($str) ; $i++ ) { 237 $res .= 238 chr( 239 utf8::native_to_unicode( ord( substr( $str, $i, 1 ) ) ) 240 ); 241 } 242 $_[1] = '' if $chk; 243 return $res; 244 }; 245 $Encode::Encoding{Unicode} = 246 bless { Name => "UTF_EBCDIC" } => "Encode::UTF_EBCDIC"; 247 } 248 else { 249 250 package Encode::Internal; 251 push @Encode::Internal::ISA, 'Encode::Encoding'; 252 *decode = sub { 253 my ( $obj, $str, $chk ) = @_; 254 utf8::upgrade($str); 255 $_[1] = '' if $chk; 256 return $str; 257 }; 258 *encode = \&decode; 259 $Encode::Encoding{Unicode} = 260 bless { Name => "Internal" } => "Encode::Internal"; 261 } 262 263 { 264 265 # was in Encode::utf8 266 package Encode::utf8; 267 push @Encode::utf8::ISA, 'Encode::Encoding'; 268 269 # 270 if ($use_xs) { 271 Encode::DEBUG and warn __PACKAGE__, " XS on"; 272 *decode = \&decode_xs; 273 *encode = \&encode_xs; 274 } 275 else { 276 Encode::DEBUG and warn __PACKAGE__, " XS off"; 277 *decode = sub { 278 my ( $obj, $octets, $chk ) = @_; 279 my $str = Encode::decode_utf8($octets); 280 if ( defined $str ) { 281 $_[1] = '' if $chk; 282 return $str; 283 } 284 return undef; 285 }; 286 *encode = sub { 287 my ( $obj, $string, $chk ) = @_; 288 my $octets = Encode::encode_utf8($string); 289 $_[1] = '' if $chk; 290 return $octets; 291 }; 292 } 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 ]; 297 use bytes; 298 if ( ( my $npos = index( $$rsrc, $trm, $pos ) ) >= 0 ) { 299 $$rdst .= 300 substr( $$rsrc, $pos, $npos - $pos + length($trm) ); 301 $$rpos = $npos + length($trm); 302 return 1; 303 } 304 $$rdst .= substr( $$rsrc, $pos ); 305 $$rpos = length($$rsrc); 306 return ''; 307 }; 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 } => 312 "Encode::utf8"; 313 } 314 } 315 316 1; 317 318 __END__ 319 320 =head1 NAME 321 322 Encode - character encodings 323 324 =head1 SYNOPSIS 325 326 use Encode; 327 328 =head2 Table of Contents 329 330 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, 333 see the PODs below: 334 335 Name Description 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 -------------------------------------------------------- 345 346 =head1 DESCRIPTION 347 348 The C<Encode> module provides the interfaces between Perl's strings 349 and the rest of the system. Perl strings are sequences of 350 B<characters>. 351 352 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>). 358 359 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. 365 366 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 369 "logical character". 370 371 =head2 TERMINOLOGY 372 373 =over 2 374 375 =item * 376 377 I<character>: a character in the range 0..(2**32-1) (or more). 378 (What Perl's strings are made of.) 379 380 =item * 381 382 I<byte>: a character in the range 0..255 383 (A special case of a Perl character.) 384 385 =item * 386 387 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.) 389 390 =back 391 392 =head1 PERL ENCODING API 393 394 =over 2 395 396 =item $octets = encode(ENCODING, $string [, CHECK]) 397 398 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">. 402 403 For example, to convert a string from Perl's internal format to 404 iso-8859-1 (also known as Latin1), 405 406 $octets = encode("iso-8859-1", $string); 407 408 B<CAVEAT>: When you run C<$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. 413 414 If the $string is C<undef> then C<undef> is returned. 415 416 =item $string = decode(ENCODING, $octets [, CHECK]) 417 418 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">. 423 424 For example, to convert ISO-8859-1 data to a string in Perl's internal format: 425 426 $string = decode("iso-8859-1", $octets); 427 428 B<CAVEAT>: When you run C<$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"> 432 below. 433 434 If the $string is C<undef> then C<undef> is returned. 435 436 =item [$obj =] find_encoding(ENCODING) 437 438 Returns the I<encoding object> corresponding to ENCODING. Returns 439 undef if no matching ENCODING is find. 440 441 This object is what actually does the actual (en|de)coding. 442 443 $utf8 = decode($name, $bytes); 444 445 is in fact 446 447 $utf8 = do{ 448 $obj = find_encoding($name); 449 croak qq(encoding "$name" not found) unless ref $obj; 450 $obj->decode($bytes) 451 }; 452 453 with more error checking. 454 455 Therefore you can save time by reusing this object as follows; 456 457 my $enc = find_encoding("iso-8859-1"); 458 while(<>){ 459 my $utf8 = $enc->decode($_); 460 # and do someting with $utf8; 461 } 462 463 Besides C<< ->decode >> and C<< ->encode >>, other methods are 464 available as well. For instance, C<< -> name >> returns the canonical 465 name of the encoding object. 466 467 find_encoding("latin1")->name; # iso-8859-1 468 469 See L<Encode::Encoding> for details. 470 471 =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) 472 473 Converts B<in-place> data between two encodings. The data in $octets 474 must be encoded as octets and not as characters in Perl's internal 475 format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 476 encoding: 477 478 from_to($octets, "iso-8859-1", "cp1250"); 479 480 and to convert it back: 481 482 from_to($octets, "cp1250", "iso-8859-1"); 483 484 Note that because the conversion happens in place, the data to be 485 converted cannot be a string constant; it must be a scalar variable. 486 487 from_to() returns the length of the converted string in octets on 488 success, I<undef> on error. 489 490 B<CAVEAT>: The following operations look the same but are not quite so; 491 492 from_to($data, "iso-8859-1", "utf8"); #1 493 $data = decode("iso-8859-1", $data); #2 494 495 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 497 498 $data = encode("utf8", decode("iso-8859-1", $data)); 499 500 See L</"The UTF8 flag"> below. 501 502 Also note that 503 504 from_to($octets, $from, $to, $check); 505 506 is equivalent to 507 508 $octets = encode($to, decode($from, $octets), $check); 509 510 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; 513 514 $octets = encode($to, decode($from, $octets, $check_from), $check_to); 515 516 =item $octets = encode_utf8($string); 517 518 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. 522 523 524 =item $string = decode_utf8($octets [, CHECK]); 525 526 equivalent to C<$string = decode("utf8", $octets [, CHECK])>. 527 The sequence of octets represented by 528 $octets is decoded from UTF-8 into a sequence of logical 529 characters. Not all sequences of octets form valid UTF-8 encodings, so 530 it is possible for this call to fail. For CHECK, see 531 L</"Handling Malformed Data">. 532 533 =back 534 535 =head2 Listing available encodings 536 537 use Encode; 538 @list = Encode->encodings(); 539 540 Returns a list of the canonical names of the available encodings that 541 are loaded. To get a list of all available encodings including the 542 ones that are not loaded yet, say 543 544 @all_encodings = Encode->encodings(":all"); 545 546 Or you can give the name of a specific module. 547 548 @with_jp = Encode->encodings("Encode::JP"); 549 550 When "::" is not in the name, "Encode::" is assumed. 551 552 @ebcdic = Encode->encodings("EBCDIC"); 553 554 To find out in detail which encodings are supported by this package, 555 see L<Encode::Supported>. 556 557 =head2 Defining Aliases 558 559 To add a new alias to a given encoding, use: 560 561 use Encode; 562 use Encode::Alias; 563 define_alias(newName => ENCODING); 564 565 After that, newName can be used as an alias for ENCODING. 566 ENCODING may be either the name of an encoding or an 567 I<encoding object> 568 569 But before you do so, make sure the alias is nonexistent with 570 C<resolve_alias()>, which returns the canonical name thereof. 571 i.e. 572 573 Encode::resolve_alias("latin1") eq "iso-8859-1" # true 574 Encode::resolve_alias("iso-8859-12") # false; nonexistent 575 Encode::resolve_alias($name) eq $name # true if $name is canonical 576 577 resolve_alias() does not need C<use Encode::Alias>; it can be 578 exported via C<use Encode qw(resolve_alias)>. 579 580 See L<Encode::Alias> for details. 581 582 =head2 Finding IANA Character Set Registry names 583 584 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'). 588 589 Therefore as of Encode version 2.21, a new method C<mime_name()> is added. 590 591 use Encode; 592 my $enc = find_encoding('UTF-8'); 593 warn $enc->name; # utf-8-strict 594 warn $enc->mime_name; # UTF-8 595 596 See also: L<Encode::Encoding> 597 598 =head1 Encoding via PerlIO 599 600 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. 603 604 # via PerlIO 605 open my $in, "<:encoding(shiftjis)", $infile or die; 606 open my $out, ">:encoding(euc-jp)", $outfile or die; 607 while(<$in>){ print $out $_; } 608 609 # via from_to 610 open my $in, "<", $infile or die; 611 open my $out, ">", $outfile or die; 612 while(<$in>){ 613 from_to($_, "shiftjis", "euc-jp", 1); 614 print $out $_; 615 } 616 617 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> 619 method. 620 621 Encode::perlio_ok("hz"); # False 622 find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available 623 624 use Encode qw(perlio_ok); # exported upon request 625 perlio_ok("euc-jp") 626 627 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>. 630 631 =head1 Handling Malformed Data 632 633 The optional I<CHECK> argument tells Encode what to do when it 634 encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 ) 635 is assumed. 636 637 As of version 2.12 Encode supports coderef values for CHECK. See below. 638 639 =over 2 640 641 =item B<NOTE:> Not all encoding support this feature 642 643 Some encodings ignore I<CHECK> argument. For example, 644 L<Encode::Unicode> ignores I<CHECK> and it always croaks on error. 645 646 =back 647 648 Now here is the list of I<CHECK> values available 649 650 =over 2 651 652 =item I<CHECK> = Encode::FB_DEFAULT ( == 0) 653 654 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. 659 660 =item I<CHECK> = Encode::FB_CROAK ( == 1) 661 662 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. 665 666 =item I<CHECK> = Encode::FB_QUIET 667 668 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: 676 677 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 681 } 682 683 =item I<CHECK> = Encode::FB_WARN 684 685 This is the same as above, except that it warns on error. Handy when 686 you are debugging the mode above. 687 688 =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) 689 690 =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) 691 692 =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) 693 694 For encodings that are implemented by Encode::XS, CHECK == 695 Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. 696 697 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. 702 703 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. 706 707 In Encode 2.10 or later, C<LEAVE_SRC> is also implied. 708 709 =item The bitmask 710 711 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)>. 715 716 FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ 717 DIE_ON_ERR 0x0001 X 718 WARN_ON_ERR 0x0002 X 719 RETURN_ON_ERR 0x0004 X X 720 LEAVE_SRC 0x0008 X 721 PERLQQ 0x0100 X 722 HTMLCREF 0x0200 723 XMLCREF 0x0400 724 725 =back 726 727 =over 2 728 729 =item Encode::LEAVE_SRC 730 731 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. 734 735 =back 736 737 =Head2 coderef for CHECK 738 739 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, 742 743 $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift }); 744 745 Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of 746 \x{I<XXXX>}. 747 748 =head1 Defining Encodings 749 750 To define a new encoding, use: 751 752 use Encode qw(define_encoding); 753 define_encoding($object, 'canonicalName' [, alias...]); 754 755 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>. 759 760 See L<Encode::Encoding> for more details. 761 762 =head1 The UTF8 flag 763 764 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.> 769 770 =over 2 771 772 =item Goal #1: 773 774 Old byte-oriented programs should not spontaneously break on the old 775 byte-oriented data they used to work on. 776 777 =item Goal #2: 778 779 Old byte-oriented programs should magically start working on the new 780 character-oriented data when appropriate. 781 782 =item Goal #3: 783 784 Programs should run just as fast in the new character-oriented mode 785 as in the old byte-oriented mode. 786 787 =item Goal #4: 788 789 Perl should remain one language, rather than forking into a 790 byte-oriented Perl and a character-oriented Perl. 791 792 =back 793 794 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 799 flag on). 800 801 Here is how Encode takes care of the UTF8 flag. 802 803 =over 2 804 805 =item * 806 807 When you encode, the resulting UTF8 flag is always off. 808 809 =item * 810 811 When you decode, the resulting UTF8 flag is on unless you can 812 unambiguously represent data. Here is the definition of 813 dis-ambiguity. 814 815 After C<$utf8 = decode('foo', $octet);>, 816 817 When $octet is... The UTF8 flag in $utf8 is 818 --------------------------------------------- 819 In ASCII only (or EBCDIC only) OFF 820 In ISO-8859-1 ON 821 In any other Encoding ON 822 --------------------------------------------- 823 824 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. 827 828 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. 832 833 =back 834 835 =head2 Messing with Perl's Internals 836 837 The following API uses parts of Perl's internals in the current 838 implementation. As such, they are efficient but may change. 839 840 =over 2 841 842 =item is_utf8(STRING [, CHECK]) 843 844 [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. 847 848 As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). 849 850 =item _utf8_on(STRING) 851 852 [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. 857 858 =item _utf8_off(STRING) 859 860 [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 863 not a string. 864 865 =back 866 867 =head1 UTF-8 vs. utf8 vs. UTF8 868 869 ....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. 872 873 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). 876 877 Now that is overruled by Larry Wall himself. 878 879 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> 884 885 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. 889 890 For what it's worth, that's how I've always kept them straight in my 891 head. 892 893 Also for what it's worth, Perl 6 will mostly default to strict but 894 make it easy to switch back to lax. 895 896 Larry 897 898 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". 901 902 encode("utf8", "\x{FFFF_FFFF}", 1); # okay 903 encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks 904 905 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 907 goes "liberal" 908 909 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'. 913 914 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. 916 917 =head1 SEE ALSO 918 919 L<Encode::Encoding>, 920 L<Encode::Supported>, 921 L<Encode::PerlIO>, 922 L<encoding>, 923 L<perlebcdic>, 924 L<perlfunc/open>, 925 L<perlunicode>, L<perluniintro>, L<perlunifaq>, L<perlunitut> 926 L<utf8>, 927 the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> 928 929 =head1 MAINTAINER 930 931 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. 935 936 While Dan Kogai retains the copyright as a maintainer, the credit 937 should go to all those involoved. See AUTHORS for those submitted 938 codes. 939 940 =head1 COPYRIGHT 941 942 Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt> 943 944 This library is free software; you can redistribute it and/or modify 945 it under the same terms as Perl itself. 946 947 =cut
title
Description
Body
title
Description
Body
title
Description
Body
title
Body
| Generated: Tue Mar 17 22:47:18 2015 | Cross-referenced by PHPXref 0.7.1 |