Commit | Line | Data |
---|---|---|
86530b38 AT |
1 | # |
2 | # $Id: Encode.pm,v 1.75 2002/06/01 18:07:42 dankogai Exp $ | |
3 | # | |
4 | package Encode; | |
5 | use strict; | |
6 | our $VERSION = do { my @r = (q$Revision: 1.75 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r }; | |
7 | our $DEBUG = 0; | |
8 | use XSLoader (); | |
9 | XSLoader::load(__PACKAGE__, $VERSION); | |
10 | ||
11 | require Exporter; | |
12 | use base qw/Exporter/; | |
13 | ||
14 | # Public, encouraged API is exported by default | |
15 | ||
16 | our @EXPORT = qw( | |
17 | decode decode_utf8 encode encode_utf8 | |
18 | encodings find_encoding | |
19 | ); | |
20 | ||
21 | our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC | |
22 | PERLQQ HTMLCREF XMLCREF); | |
23 | our @FB_CONSTS = qw(FB_DEFAULT FB_CROAK FB_QUIET FB_WARN | |
24 | FB_PERLQQ FB_HTMLCREF FB_XMLCREF); | |
25 | ||
26 | our @EXPORT_OK = | |
27 | ( | |
28 | qw( | |
29 | _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit | |
30 | is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade | |
31 | ), | |
32 | @FB_FLAGS, @FB_CONSTS, | |
33 | ); | |
34 | ||
35 | our %EXPORT_TAGS = | |
36 | ( | |
37 | all => [ @EXPORT, @EXPORT_OK ], | |
38 | fallbacks => [ @FB_CONSTS ], | |
39 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], | |
40 | ); | |
41 | ||
42 | # Documentation moved after __END__ for speed - NI-S | |
43 | ||
44 | our $ON_EBCDIC = (ord("A") == 193); | |
45 | ||
46 | use Encode::Alias; | |
47 | ||
48 | # Make a %Encoding package variable to allow a certain amount of cheating | |
49 | our %Encoding; | |
50 | our %ExtModule; | |
51 | require Encode::Config; | |
52 | eval { require Encode::ConfigLocal }; | |
53 | ||
54 | sub encodings | |
55 | { | |
56 | my $class = shift; | |
57 | my %enc; | |
58 | if (@_ and $_[0] eq ":all"){ | |
59 | %enc = ( %Encoding, %ExtModule ); | |
60 | }else{ | |
61 | %enc = %Encoding; | |
62 | for my $mod (map {m/::/o ? $_ : "Encode::$_" } @_){ | |
63 | $DEBUG and warn $mod; | |
64 | for my $enc (keys %ExtModule){ | |
65 | $ExtModule{$enc} eq $mod and $enc{$enc} = $mod; | |
66 | } | |
67 | } | |
68 | } | |
69 | return | |
70 | 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 | { | |
82 | my $obj = shift; | |
83 | my $name = shift; | |
84 | $Encoding{$name} = $obj; | |
85 | my $lc = lc($name); | |
86 | define_alias($lc => $obj) unless $lc eq $name; | |
87 | while (@_){ | |
88 | my $alias = shift; | |
89 | define_alias($alias, $obj); | |
90 | } | |
91 | return $obj; | |
92 | } | |
93 | ||
94 | sub getEncoding | |
95 | { | |
96 | my ($class, $name, $skip_external) = @_; | |
97 | ||
98 | ref($name) && $name->can('new_sequence') and return $name; | |
99 | exists $Encoding{$name} and return $Encoding{$name}; | |
100 | my $lc = lc $name; | |
101 | exists $Encoding{$lc} and return $Encoding{$lc}; | |
102 | ||
103 | my $oc = $class->find_alias($name); | |
104 | defined($oc) and return $oc; | |
105 | $lc ne $name and $oc = $class->find_alias($lc); | |
106 | defined($oc) and return $oc; | |
107 | ||
108 | unless ($skip_external) | |
109 | { | |
110 | if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){ | |
111 | $mod =~ s,::,/,g ; $mod .= '.pm'; | |
112 | eval{ require $mod; }; | |
113 | exists $Encoding{$name} and return $Encoding{$name}; | |
114 | } | |
115 | } | |
116 | return; | |
117 | } | |
118 | ||
119 | sub find_encoding | |
120 | { | |
121 | my ($name, $skip_external) = @_; | |
122 | return __PACKAGE__->getEncoding($name,$skip_external); | |
123 | } | |
124 | ||
125 | sub resolve_alias { | |
126 | my $obj = find_encoding(shift); | |
127 | defined $obj and return $obj->name; | |
128 | return; | |
129 | } | |
130 | ||
131 | sub encode($$;$) | |
132 | { | |
133 | my ($name, $string, $check) = @_; | |
134 | $check ||=0; | |
135 | my $enc = find_encoding($name); | |
136 | unless(defined $enc){ | |
137 | require Carp; | |
138 | Carp::croak("Unknown encoding '$name'"); | |
139 | } | |
140 | my $octets = $enc->encode($string,$check); | |
141 | return undef if ($check && length($string)); | |
142 | return $octets; | |
143 | } | |
144 | ||
145 | sub decode($$;$) | |
146 | { | |
147 | my ($name,$octets,$check) = @_; | |
148 | $check ||=0; | |
149 | my $enc = find_encoding($name); | |
150 | unless(defined $enc){ | |
151 | require Carp; | |
152 | Carp::croak("Unknown encoding '$name'"); | |
153 | } | |
154 | my $string = $enc->decode($octets,$check); | |
155 | $_[1] = $octets if $check; | |
156 | return $string; | |
157 | } | |
158 | ||
159 | sub from_to($$$;$) | |
160 | { | |
161 | my ($string,$from,$to,$check) = @_; | |
162 | $check ||=0; | |
163 | my $f = find_encoding($from); | |
164 | unless (defined $f){ | |
165 | require Carp; | |
166 | Carp::croak("Unknown encoding '$from'"); | |
167 | } | |
168 | my $t = find_encoding($to); | |
169 | unless (defined $t){ | |
170 | require Carp; | |
171 | Carp::croak("Unknown encoding '$to'"); | |
172 | } | |
173 | my $uni = $f->decode($string,$check); | |
174 | return undef if ($check && length($string)); | |
175 | $string = $t->encode($uni,$check); | |
176 | return undef if ($check && length($uni)); | |
177 | return defined($_[0] = $string) ? length($string) : undef ; | |
178 | } | |
179 | ||
180 | sub encode_utf8($) | |
181 | { | |
182 | my ($str) = @_; | |
183 | utf8::encode($str); | |
184 | return $str; | |
185 | } | |
186 | ||
187 | sub decode_utf8($) | |
188 | { | |
189 | my ($str) = @_; | |
190 | return undef unless utf8::decode($str); | |
191 | return $str; | |
192 | } | |
193 | ||
194 | predefine_encodings(); | |
195 | ||
196 | # | |
197 | # This is to restore %Encoding if really needed; | |
198 | # | |
199 | ||
200 | sub predefine_encodings{ | |
201 | use Encode::Encoding; | |
202 | if ($ON_EBCDIC) { | |
203 | # was in Encode::UTF_EBCDIC | |
204 | package Encode::UTF_EBCDIC; | |
205 | push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; | |
206 | *decode = sub{ | |
207 | my ($obj,$str,$chk) = @_; | |
208 | my $res = ''; | |
209 | for (my $i = 0; $i < length($str); $i++) { | |
210 | $res .= | |
211 | chr(utf8::unicode_to_native(ord(substr($str,$i,1)))); | |
212 | } | |
213 | $_[1] = '' if $chk; | |
214 | return $res; | |
215 | }; | |
216 | *encode = sub{ | |
217 | my ($obj,$str,$chk) = @_; | |
218 | my $res = ''; | |
219 | for (my $i = 0; $i < length($str); $i++) { | |
220 | $res .= | |
221 | chr(utf8::native_to_unicode(ord(substr($str,$i,1)))); | |
222 | } | |
223 | $_[1] = '' if $chk; | |
224 | return $res; | |
225 | }; | |
226 | $Encode::Encoding{Unicode} = | |
227 | bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC"; | |
228 | } else { | |
229 | package Encode::Internal; | |
230 | push @Encode::Internal::ISA, 'Encode::Encoding'; | |
231 | *decode = sub{ | |
232 | my ($obj,$str,$chk) = @_; | |
233 | utf8::upgrade($str); | |
234 | $_[1] = '' if $chk; | |
235 | return $str; | |
236 | }; | |
237 | *encode = \&decode; | |
238 | $Encode::Encoding{Unicode} = | |
239 | bless {Name => "Internal"} => "Encode::Internal"; | |
240 | } | |
241 | ||
242 | { | |
243 | # was in Encode::utf8 | |
244 | package Encode::utf8; | |
245 | push @Encode::utf8::ISA, 'Encode::Encoding'; | |
246 | *decode = sub{ | |
247 | my ($obj,$octets,$chk) = @_; | |
248 | my $str = Encode::decode_utf8($octets); | |
249 | if (defined $str) { | |
250 | $_[1] = '' if $chk; | |
251 | return $str; | |
252 | } | |
253 | return undef; | |
254 | }; | |
255 | *encode = sub { | |
256 | my ($obj,$string,$chk) = @_; | |
257 | my $octets = Encode::encode_utf8($string); | |
258 | $_[1] = '' if $chk; | |
259 | return $octets; | |
260 | }; | |
261 | $Encode::Encoding{utf8} = | |
262 | bless {Name => "utf8"} => "Encode::utf8"; | |
263 | } | |
264 | } | |
265 | ||
266 | 1; | |
267 | ||
268 | __END__ | |
269 | ||
270 | =head1 NAME | |
271 | ||
272 | Encode - character encodings | |
273 | ||
274 | =head1 SYNOPSIS | |
275 | ||
276 | use Encode; | |
277 | ||
278 | =head2 Table of Contents | |
279 | ||
280 | Encode consists of a collection of modules whose details are too big | |
281 | to fit in one document. This POD itself explains the top-level APIs | |
282 | and general topics at a glance. For other topics and more details, | |
283 | see the PODs below: | |
284 | ||
285 | Name Description | |
286 | -------------------------------------------------------- | |
287 | Encode::Alias Alias definitions to encodings | |
288 | Encode::Encoding Encode Implementation Base Class | |
289 | Encode::Supported List of Supported Encodings | |
290 | Encode::CN Simplified Chinese Encodings | |
291 | Encode::JP Japanese Encodings | |
292 | Encode::KR Korean Encodings | |
293 | Encode::TW Traditional Chinese Encodings | |
294 | -------------------------------------------------------- | |
295 | ||
296 | =head1 DESCRIPTION | |
297 | ||
298 | The C<Encode> module provides the interfaces between Perl's strings | |
299 | and the rest of the system. Perl strings are sequences of | |
300 | B<characters>. | |
301 | ||
302 | The repertoire of characters that Perl can represent is at least that | |
303 | defined by the Unicode Consortium. On most platforms the ordinal | |
304 | values of the characters (as returned by C<ord(ch)>) is the "Unicode | |
305 | codepoint" for the character (the exceptions are those platforms where | |
306 | the legacy encoding is some variant of EBCDIC rather than a super-set | |
307 | of ASCII - see L<perlebcdic>). | |
308 | ||
309 | Traditionally, computer data has been moved around in 8-bit chunks | |
310 | often called "bytes". These chunks are also known as "octets" in | |
311 | networking standards. Perl is widely used to manipulate data of many | |
312 | types - not only strings of characters representing human or computer | |
313 | languages but also "binary" data being the machine's representation of | |
314 | numbers, pixels in an image - or just about anything. | |
315 | ||
316 | When Perl is processing "binary data", the programmer wants Perl to | |
317 | process "sequences of bytes". This is not a problem for Perl - as a | |
318 | byte has 256 possible values, it easily fits in Perl's much larger | |
319 | "logical character". | |
320 | ||
321 | =head2 TERMINOLOGY | |
322 | ||
323 | =over 2 | |
324 | ||
325 | =item * | |
326 | ||
327 | I<character>: a character in the range 0..(2**32-1) (or more). | |
328 | (What Perl's strings are made of.) | |
329 | ||
330 | =item * | |
331 | ||
332 | I<byte>: a character in the range 0..255 | |
333 | (A special case of a Perl character.) | |
334 | ||
335 | =item * | |
336 | ||
337 | I<octet>: 8 bits of data, with ordinal values 0..255 | |
338 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) | |
339 | ||
340 | =back | |
341 | ||
342 | =head1 PERL ENCODING API | |
343 | ||
344 | =over 2 | |
345 | ||
346 | =item $octets = encode(ENCODING, $string [, CHECK]) | |
347 | ||
348 | Encodes a string from Perl's internal form into I<ENCODING> and returns | |
349 | a sequence of octets. ENCODING can be either a canonical name or | |
350 | an alias. For encoding names and aliases, see L</"Defining Aliases">. | |
351 | For CHECK, see L</"Handling Malformed Data">. | |
352 | ||
353 | For example, to convert a string from Perl's internal format to | |
354 | iso-8859-1 (also known as Latin1), | |
355 | ||
356 | $octets = encode("iso-8859-1", $string); | |
357 | ||
358 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets | |
359 | B<may not be equal to> $string. Though they both contain the same data, the utf8 flag | |
360 | for $octets is B<always> off. When you encode anything, utf8 flag of | |
361 | the result is always off, even when it contains completely valid utf8 | |
362 | string. See L</"The UTF-8 flag"> below. | |
363 | ||
364 | encode($valid_encoding, undef) is harmless but warns you for | |
365 | C<Use of uninitialized value in subroutine entry>. | |
366 | encode($valid_encoding, '') is harmless and warnless. | |
367 | ||
368 | =item $string = decode(ENCODING, $octets [, CHECK]) | |
369 | ||
370 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's | |
371 | internal form and returns the resulting string. As in encode(), | |
372 | ENCODING can be either a canonical name or an alias. For encoding names | |
373 | and aliases, see L</"Defining Aliases">. For CHECK, see | |
374 | L</"Handling Malformed Data">. | |
375 | ||
376 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: | |
377 | ||
378 | $string = decode("iso-8859-1", $octets); | |
379 | ||
380 | B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string | |
381 | B<may not be equal to> $octets. Though they both contain the same data, | |
382 | the utf8 flag for $string is on unless $octets entirely consists of | |
383 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag"> | |
384 | below. | |
385 | ||
386 | decode($valid_encoding, undef) is harmless but warns you for | |
387 | C<Use of uninitialized value in subroutine entry>. | |
388 | decode($valid_encoding, '') is harmless and warnless. | |
389 | ||
390 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) | |
391 | ||
392 | Converts B<in-place> data between two encodings. The data in $octets | |
393 | must be encoded as octets and not as characters in Perl's internal | |
394 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 encoding: | |
395 | ||
396 | from_to($octets, "iso-8859-1", "cp1250"); | |
397 | ||
398 | and to convert it back: | |
399 | ||
400 | from_to($octets, "cp1250", "iso-8859-1"); | |
401 | ||
402 | Note that because the conversion happens in place, the data to be | |
403 | converted cannot be a string constant; it must be a scalar variable. | |
404 | ||
405 | from_to() returns the length of the converted string in octets on success, undef | |
406 | otherwise. | |
407 | ||
408 | B<CAVEAT>: The following operations look the same but are not quite so; | |
409 | ||
410 | from_to($data, "iso-8859-1", "utf8"); #1 | |
411 | $data = decode("iso-8859-1", $data); #2 | |
412 | ||
413 | Both #1 and #2 make $data consist of a completely valid UTF-8 string | |
414 | but only #2 turns utf8 flag on. #1 is equivalent to | |
415 | ||
416 | $data = encode("utf8", decode("iso-8859-1", $data)); | |
417 | ||
418 | See L</"The UTF-8 flag"> below. | |
419 | ||
420 | =item $octets = encode_utf8($string); | |
421 | ||
422 | Equivalent to C<$octets = encode("utf8", $string);> The characters | |
423 | that comprise $string are encoded in Perl's internal format and the | |
424 | result is returned as a sequence of octets. All possible | |
425 | characters have a UTF-8 representation so this function cannot fail. | |
426 | ||
427 | ||
428 | =item $string = decode_utf8($octets [, CHECK]); | |
429 | ||
430 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. | |
431 | The sequence of octets represented by | |
432 | $octets is decoded from UTF-8 into a sequence of logical | |
433 | characters. Not all sequences of octets form valid UTF-8 encodings, so | |
434 | it is possible for this call to fail. For CHECK, see | |
435 | L</"Handling Malformed Data">. | |
436 | ||
437 | =back | |
438 | ||
439 | =head2 Listing available encodings | |
440 | ||
441 | use Encode; | |
442 | @list = Encode->encodings(); | |
443 | ||
444 | Returns a list of the canonical names of the available encodings that | |
445 | are loaded. To get a list of all available encodings including the | |
446 | ones that are not loaded yet, say | |
447 | ||
448 | @all_encodings = Encode->encodings(":all"); | |
449 | ||
450 | Or you can give the name of a specific module. | |
451 | ||
452 | @with_jp = Encode->encodings("Encode::JP"); | |
453 | ||
454 | When "::" is not in the name, "Encode::" is assumed. | |
455 | ||
456 | @ebcdic = Encode->encodings("EBCDIC"); | |
457 | ||
458 | To find out in detail which encodings are supported by this package, | |
459 | see L<Encode::Supported>. | |
460 | ||
461 | =head2 Defining Aliases | |
462 | ||
463 | To add a new alias to a given encoding, use: | |
464 | ||
465 | use Encode; | |
466 | use Encode::Alias; | |
467 | define_alias(newName => ENCODING); | |
468 | ||
469 | After that, newName can be used as an alias for ENCODING. | |
470 | ENCODING may be either the name of an encoding or an | |
471 | I<encoding object> | |
472 | ||
473 | But before you do so, make sure the alias is nonexistent with | |
474 | C<resolve_alias()>, which returns the canonical name thereof. | |
475 | i.e. | |
476 | ||
477 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true | |
478 | Encode::resolve_alias("iso-8859-12") # false; nonexistent | |
479 | Encode::resolve_alias($name) eq $name # true if $name is canonical | |
480 | ||
481 | resolve_alias() does not need C<use Encode::Alias>; it can be | |
482 | exported via C<use Encode qw(resolve_alias)>. | |
483 | ||
484 | See L<Encode::Alias> for details. | |
485 | ||
486 | =head1 Encoding via PerlIO | |
487 | ||
488 | If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode | |
489 | and encode directly via a filehandle. The following two examples | |
490 | are totally identical in their functionality. | |
491 | ||
492 | # via PerlIO | |
493 | open my $in, "<:encoding(shiftjis)", $infile or die; | |
494 | open my $out, ">:encoding(euc-jp)", $outfile or die; | |
495 | while(<$in>){ print $out $_; } | |
496 | ||
497 | # via from_to | |
498 | open my $in, "<", $infile or die; | |
499 | open my $out, ">", $outfile or die; | |
500 | while(<$in>){ | |
501 | from_to($_, "shiftjis", "euc-jp", 1); | |
502 | print $out $_; | |
503 | } | |
504 | ||
505 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check | |
506 | if your encoding is supported by PerlIO by calling the C<perlio_ok> | |
507 | method. | |
508 | ||
509 | Encode::perlio_ok("hz"); # False | |
510 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available | |
511 | ||
512 | use Encode qw(perlio_ok); # exported upon request | |
513 | perlio_ok("euc-jp") | |
514 | ||
515 | Fortunately, all encodings that come with Encode core are PerlIO-savvy | |
516 | except for hz and ISO-2022-kr. For gory details, see L<Encode::Encoding> and L<Encode::PerlIO>. | |
517 | ||
518 | =head1 Handling Malformed Data | |
519 | ||
520 | =over 2 | |
521 | ||
522 | The I<CHECK> argument is used as follows. When you omit it, | |
523 | the behaviour is the same as if you had passed a value of 0 for | |
524 | I<CHECK>. | |
525 | ||
526 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) | |
527 | ||
528 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> | |
529 | in place of a malformed character. For UCM-based encodings, | |
530 | E<lt>subcharE<gt> will be used. For Unicode, the code point C<0xFFFD> is used. | |
531 | If the data is supposed to be UTF-8, an optional lexical warning | |
532 | (category utf8) is given. | |
533 | ||
534 | =item I<CHECK> = Encode::FB_CROAK ( == 1) | |
535 | ||
536 | If I<CHECK> is 1, methods will die on error immediately with an error | |
537 | message. Therefore, when I<CHECK> is set to 1, you should trap the | |
538 | fatal error with eval{} unless you really want to let it die on error. | |
539 | ||
540 | =item I<CHECK> = Encode::FB_QUIET | |
541 | ||
542 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately | |
543 | return the portion of the data that has been processed so far when | |
544 | an error occurs. The data argument will be overwritten with | |
545 | everything after that point (that is, the unprocessed part of data). | |
546 | This is handy when you have to call decode repeatedly in the case | |
547 | where your source data may contain partial multi-byte character | |
548 | sequences, for example because you are reading with a fixed-width | |
549 | buffer. Here is some sample code that does exactly this: | |
550 | ||
551 | my $data = ''; my $utf8 = ''; | |
552 | while(defined(read $fh, $buffer, 256)){ | |
553 | # buffer may end in a partial character so we append | |
554 | $data .= $buffer; | |
555 | $utf8 .= decode($encoding, $data, Encode::FB_QUIET); | |
556 | # $data now contains the unprocessed partial character | |
557 | } | |
558 | ||
559 | =item I<CHECK> = Encode::FB_WARN | |
560 | ||
561 | This is the same as above, except that it warns on error. Handy when | |
562 | you are debugging the mode above. | |
563 | ||
564 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) | |
565 | ||
566 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) | |
567 | ||
568 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) | |
569 | ||
570 | For encodings that are implemented by Encode::XS, CHECK == | |
571 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. | |
572 | ||
573 | When you decode, C<\xI<HH>> will be inserted for a malformed character, | |
574 | where I<HH> is the hex representation of the octet that could not be | |
575 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, | |
576 | where I<HHHH> is the Unicode ID of the character that cannot be found | |
577 | in the character repertoire of the encoding. | |
578 | ||
579 | HTML/XML character reference modes are about the same, in place of | |
580 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNNN>>; where I<NNNN> is a decimal digit and | |
581 | XML uses C<&#xI<HHHH>>; where I<HHHH> is the hexadecimal digit. | |
582 | ||
583 | =item The bitmask | |
584 | ||
585 | These modes are actually set via a bitmask. Here is how the FB_XX | |
586 | constants are laid out. You can import the FB_XX constants via | |
587 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask | |
588 | constants via C<use Encode qw(:fallback_all)>. | |
589 | ||
590 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ | |
591 | DIE_ON_ERR 0x0001 X | |
592 | WARN_ON_ERR 0x0002 X | |
593 | RETURN_ON_ERR 0x0004 X X | |
594 | LEAVE_SRC 0x0008 | |
595 | PERLQQ 0x0100 X | |
596 | HTMLCREF 0x0200 | |
597 | XMLCREF 0x0400 | |
598 | ||
599 | =head2 Unimplemented fallback schemes | |
600 | ||
601 | In the future, you will be able to use a code reference to a callback | |
602 | function for the value of I<CHECK> but its API is still undecided. | |
603 | ||
604 | The fallback scheme does not work on EBCDIC platforms. | |
605 | ||
606 | =head1 Defining Encodings | |
607 | ||
608 | To define a new encoding, use: | |
609 | ||
610 | use Encode qw(define_encoding); | |
611 | define_encoding($object, 'canonicalName' [, alias...]); | |
612 | ||
613 | I<canonicalName> will be associated with I<$object>. The object | |
614 | should provide the interface described in L<Encode::Encoding>. | |
615 | If more than two arguments are provided then additional | |
616 | arguments are taken as aliases for I<$object>. | |
617 | ||
618 | See L<Encode::Encoding> for more details. | |
619 | ||
620 | =head1 The UTF-8 flag | |
621 | ||
622 | Before the introduction of utf8 support in perl, The C<eq> operator | |
623 | just compared the strings represented by two scalars. Beginning with | |
624 | perl 5.8, C<eq> compares two strings with simultaneous consideration | |
625 | of I<the utf8 flag>. To explain why we made it so, I will quote page | |
626 | 402 of C<Programming Perl, 3rd ed.> | |
627 | ||
628 | =over 2 | |
629 | ||
630 | =item Goal #1: | |
631 | ||
632 | Old byte-oriented programs should not spontaneously break on the old | |
633 | byte-oriented data they used to work on. | |
634 | ||
635 | =item Goal #2: | |
636 | ||
637 | Old byte-oriented programs should magically start working on the new | |
638 | character-oriented data when appropriate. | |
639 | ||
640 | =item Goal #3: | |
641 | ||
642 | Programs should run just as fast in the new character-oriented mode | |
643 | as in the old byte-oriented mode. | |
644 | ||
645 | =item Goal #4: | |
646 | ||
647 | Perl should remain one language, rather than forking into a | |
648 | byte-oriented Perl and a character-oriented Perl. | |
649 | ||
650 | =back | |
651 | ||
652 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 | |
653 | was born and many features documented in the book remained | |
654 | unimplemented for a long time. Perl 5.8 corrected this and the introduction | |
655 | of the UTF-8 flag is one of them. You can think of this perl notion as of a | |
656 | byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8 | |
657 | flag on). | |
658 | ||
659 | Here is how Encode takes care of the utf8 flag. | |
660 | ||
661 | =over 2 | |
662 | ||
663 | =item * | |
664 | ||
665 | When you encode, the resulting utf8 flag is always off. | |
666 | ||
667 | =item | |
668 | ||
669 | When you decode, the resulting utf8 flag is on unless you can | |
670 | unambiguously represent data. Here is the definition of | |
671 | dis-ambiguity. | |
672 | ||
673 | After C<$utf8 = decode('foo', $octet);>, | |
674 | ||
675 | When $octet is... The utf8 flag in $utf8 is | |
676 | --------------------------------------------- | |
677 | In ASCII only (or EBCDIC only) OFF | |
678 | In ISO-8859-1 ON | |
679 | In any other Encoding ON | |
680 | --------------------------------------------- | |
681 | ||
682 | As you see, there is one exception, In ASCII. That way you can assue | |
683 | Goal #1. And with Encode Goal #2 is assumed but you still have to be | |
684 | careful in such cases mentioned in B<CAVEAT> paragraphs. | |
685 | ||
686 | This utf8 flag is not visible in perl scripts, exactly for the same | |
687 | reason you cannot (or you I<don't have to>) see if a scalar contains a | |
688 | string, integer, or floating point number. But you can still peek | |
689 | and poke these if you will. See the section below. | |
690 | ||
691 | =back | |
692 | ||
693 | =head2 Messing with Perl's Internals | |
694 | ||
695 | The following API uses parts of Perl's internals in the current | |
696 | implementation. As such, they are efficient but may change. | |
697 | ||
698 | =over 2 | |
699 | ||
700 | =item is_utf8(STRING [, CHECK]) | |
701 | ||
702 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. | |
703 | If CHECK is true, also checks the data in STRING for being well-formed | |
704 | UTF-8. Returns true if successful, false otherwise. | |
705 | ||
706 | =item _utf8_on(STRING) | |
707 | ||
708 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is | |
709 | B<not> checked for being well-formed UTF-8. Do not use unless you | |
710 | B<know> that the STRING is well-formed UTF-8. Returns the previous | |
711 | state of the UTF-8 flag (so please don't treat the return value as | |
712 | indicating success or failure), or C<undef> if STRING is not a string. | |
713 | ||
714 | =item _utf8_off(STRING) | |
715 | ||
716 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. | |
717 | Returns the previous state of the UTF-8 flag (so please don't treat the | |
718 | return value as indicating success or failure), or C<undef> if STRING is | |
719 | not a string. | |
720 | ||
721 | =back | |
722 | ||
723 | =head1 SEE ALSO | |
724 | ||
725 | L<Encode::Encoding>, | |
726 | L<Encode::Supported>, | |
727 | L<Encode::PerlIO>, | |
728 | L<encoding>, | |
729 | L<perlebcdic>, | |
730 | L<perlfunc/open>, | |
731 | L<perlunicode>, | |
732 | L<utf8>, | |
733 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> | |
734 | ||
735 | =head1 MAINTAINER | |
736 | ||
737 | This project was originated by Nick Ing-Simmons and later maintained | |
738 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full | |
739 | list of people involved. For any questions, use | |
740 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. | |
741 | ||
742 | =cut |