Commit | Line | Data |
2c674647 |
1 | package Encode; |
51ef4e11 |
2 | use strict; |
0ab8f81e |
3 | our $VERSION = do { my @r = (q$Revision: 1.56 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r }; |
5129552c |
4 | our $DEBUG = 0; |
6d1c0808 |
5 | use XSLoader (); |
6 | XSLoader::load 'Encode'; |
2c674647 |
7 | |
2c674647 |
8 | require Exporter; |
6d1c0808 |
9 | our @ISA = qw(Exporter); |
2c674647 |
10 | |
4411f3b6 |
11 | # Public, encouraged API is exported by default |
85982a32 |
12 | |
13 | our @EXPORT = qw( |
14 | decode decode_utf8 encode encode_utf8 |
15 | encodings find_encoding |
4411f3b6 |
16 | ); |
17 | |
85982a32 |
18 | our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC PERLQQ); |
19 | our @FB_CONSTS = qw(FB_DEFAULT FB_QUIET FB_WARN FB_PERLQQ FB_CROAK); |
20 | |
51ef4e11 |
21 | our @EXPORT_OK = |
6d1c0808 |
22 | ( |
85982a32 |
23 | qw( |
24 | _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit |
25 | is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade |
26 | ), |
27 | @FB_FLAGS, @FB_CONSTS, |
28 | ); |
29 | |
6d1c0808 |
30 | our %EXPORT_TAGS = |
85982a32 |
31 | ( |
32 | all => [ @EXPORT, @EXPORT_OK ], |
33 | fallbacks => [ @FB_CONSTS ], |
34 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], |
35 | ); |
36 | |
4411f3b6 |
37 | # Documentation moved after __END__ for speed - NI-S |
2c674647 |
38 | |
bf230f3d |
39 | use Carp; |
40 | |
a63c962f |
41 | our $ON_EBCDIC = (ord("A") == 193); |
f2a2953c |
42 | |
5d030b67 |
43 | use Encode::Alias; |
44 | |
5129552c |
45 | # Make a %Encoding package variable to allow a certain amount of cheating |
46 | our %Encoding; |
aae85ceb |
47 | our %ExtModule; |
48 | require Encode::Config; |
49 | eval { require Encode::ConfigLocal }; |
5129552c |
50 | |
656753f8 |
51 | sub encodings |
52 | { |
5129552c |
53 | my $class = shift; |
071db25d |
54 | my @modules = (@_ and $_[0] eq ":all") ? values %ExtModule : @_; |
c731e18e |
55 | for my $mod (@modules){ |
56 | $mod =~ s,::,/,g or $mod = "Encode/$mod"; |
6d1c0808 |
57 | $mod .= '.pm'; |
c731e18e |
58 | $DEBUG and warn "about to require $mod;"; |
59 | eval { require $mod; }; |
5129552c |
60 | } |
c731e18e |
61 | my %modules = map {$_ => 1} @modules; |
5129552c |
62 | return |
ce912cd4 |
63 | sort { lc $a cmp lc $b } |
64 | grep {!/^(?:Internal|Unicode)$/o} keys %Encoding; |
51ef4e11 |
65 | } |
66 | |
85982a32 |
67 | sub perlio_ok{ |
68 | exists $INC{"PerlIO/encoding.pm"} or return 0; |
0ab8f81e |
69 | my $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]); |
70 | $obj->can("perlio_ok") and return $obj->perlio_ok() unless $@; |
71 | return 0; # safety net |
85982a32 |
72 | } |
73 | |
51ef4e11 |
74 | sub define_encoding |
75 | { |
18586f54 |
76 | my $obj = shift; |
77 | my $name = shift; |
5129552c |
78 | $Encoding{$name} = $obj; |
18586f54 |
79 | my $lc = lc($name); |
80 | define_alias($lc => $obj) unless $lc eq $name; |
81 | while (@_) |
82 | { |
83 | my $alias = shift; |
84 | define_alias($alias,$obj); |
85 | } |
86 | return $obj; |
656753f8 |
87 | } |
88 | |
656753f8 |
89 | sub getEncoding |
90 | { |
dd9703c9 |
91 | my ($class,$name,$skip_external) = @_; |
18586f54 |
92 | my $enc; |
93 | if (ref($name) && $name->can('new_sequence')) |
94 | { |
95 | return $name; |
96 | } |
97 | my $lc = lc $name; |
5129552c |
98 | if (exists $Encoding{$name}) |
18586f54 |
99 | { |
5129552c |
100 | return $Encoding{$name}; |
18586f54 |
101 | } |
5129552c |
102 | if (exists $Encoding{$lc}) |
18586f54 |
103 | { |
5129552c |
104 | return $Encoding{$lc}; |
18586f54 |
105 | } |
c50d192e |
106 | |
5129552c |
107 | my $oc = $class->find_alias($name); |
c50d192e |
108 | return $oc if defined $oc; |
109 | |
5129552c |
110 | $oc = $class->find_alias($lc) if $lc ne $name; |
c50d192e |
111 | return $oc if defined $oc; |
112 | |
c731e18e |
113 | unless ($skip_external) |
d1ed7747 |
114 | { |
c731e18e |
115 | if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){ |
116 | $mod =~ s,::,/,g ; $mod .= '.pm'; |
117 | eval{ require $mod; }; |
118 | return $Encoding{$name} if exists $Encoding{$name}; |
119 | } |
d1ed7747 |
120 | } |
18586f54 |
121 | return; |
656753f8 |
122 | } |
123 | |
4411f3b6 |
124 | sub find_encoding |
125 | { |
dd9703c9 |
126 | my ($name,$skip_external) = @_; |
127 | return __PACKAGE__->getEncoding($name,$skip_external); |
4411f3b6 |
128 | } |
129 | |
fcb875d4 |
130 | sub resolve_alias { |
131 | my $obj = find_encoding(shift); |
132 | defined $obj and return $obj->name; |
133 | return; |
134 | } |
135 | |
b2704119 |
136 | sub encode($$;$) |
4411f3b6 |
137 | { |
18586f54 |
138 | my ($name,$string,$check) = @_; |
b2704119 |
139 | $check ||=0; |
18586f54 |
140 | my $enc = find_encoding($name); |
141 | croak("Unknown encoding '$name'") unless defined $enc; |
142 | my $octets = $enc->encode($string,$check); |
143 | return undef if ($check && length($string)); |
144 | return $octets; |
4411f3b6 |
145 | } |
146 | |
b2704119 |
147 | sub decode($$;$) |
4411f3b6 |
148 | { |
18586f54 |
149 | my ($name,$octets,$check) = @_; |
b2704119 |
150 | $check ||=0; |
18586f54 |
151 | my $enc = find_encoding($name); |
152 | croak("Unknown encoding '$name'") unless defined $enc; |
153 | my $string = $enc->decode($octets,$check); |
154 | $_[1] = $octets if $check; |
155 | return $string; |
4411f3b6 |
156 | } |
157 | |
b2704119 |
158 | sub from_to($$$;$) |
4411f3b6 |
159 | { |
18586f54 |
160 | my ($string,$from,$to,$check) = @_; |
b2704119 |
161 | $check ||=0; |
18586f54 |
162 | my $f = find_encoding($from); |
163 | croak("Unknown encoding '$from'") unless defined $f; |
164 | my $t = find_encoding($to); |
165 | croak("Unknown encoding '$to'") unless defined $t; |
166 | my $uni = $f->decode($string,$check); |
167 | return undef if ($check && length($string)); |
a999c27c |
168 | $string = $t->encode($uni,$check); |
18586f54 |
169 | return undef if ($check && length($uni)); |
3ef515df |
170 | return defined($_[0] = $string) ? length($string) : undef ; |
4411f3b6 |
171 | } |
172 | |
b2704119 |
173 | sub encode_utf8($) |
4411f3b6 |
174 | { |
18586f54 |
175 | my ($str) = @_; |
c731e18e |
176 | utf8::encode($str); |
18586f54 |
177 | return $str; |
4411f3b6 |
178 | } |
179 | |
b2704119 |
180 | sub decode_utf8($) |
4411f3b6 |
181 | { |
18586f54 |
182 | my ($str) = @_; |
183 | return undef unless utf8::decode($str); |
184 | return $str; |
5ad8ef52 |
185 | } |
186 | |
f2a2953c |
187 | predefine_encodings(); |
188 | |
189 | # |
190 | # This is to restore %Encoding if really needed; |
191 | # |
192 | sub predefine_encodings{ |
6d1c0808 |
193 | if ($ON_EBCDIC) { |
f2a2953c |
194 | # was in Encode::UTF_EBCDIC |
195 | package Encode::UTF_EBCDIC; |
196 | *name = sub{ shift->{'Name'} }; |
197 | *new_sequence = sub{ return $_[0] }; |
198 | *decode = sub{ |
199 | my ($obj,$str,$chk) = @_; |
200 | my $res = ''; |
201 | for (my $i = 0; $i < length($str); $i++) { |
6d1c0808 |
202 | $res .= |
f2a2953c |
203 | chr(utf8::unicode_to_native(ord(substr($str,$i,1)))); |
204 | } |
205 | $_[1] = '' if $chk; |
206 | return $res; |
207 | }; |
208 | *encode = sub{ |
209 | my ($obj,$str,$chk) = @_; |
210 | my $res = ''; |
211 | for (my $i = 0; $i < length($str); $i++) { |
6d1c0808 |
212 | $res .= |
f2a2953c |
213 | chr(utf8::native_to_unicode(ord(substr($str,$i,1)))); |
214 | } |
215 | $_[1] = '' if $chk; |
216 | return $res; |
217 | }; |
6d1c0808 |
218 | $Encode::Encoding{Unicode} = |
c731e18e |
219 | bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC"; |
6d1c0808 |
220 | } else { |
f2a2953c |
221 | # was in Encode::UTF_EBCDIC |
222 | package Encode::Internal; |
223 | *name = sub{ shift->{'Name'} }; |
224 | *new_sequence = sub{ return $_[0] }; |
225 | *decode = sub{ |
226 | my ($obj,$str,$chk) = @_; |
227 | utf8::upgrade($str); |
228 | $_[1] = '' if $chk; |
229 | return $str; |
230 | }; |
231 | *encode = \&decode; |
6d1c0808 |
232 | $Encode::Encoding{Unicode} = |
c731e18e |
233 | bless {Name => "Internal"} => "Encode::Internal"; |
f2a2953c |
234 | } |
235 | |
236 | { |
237 | # was in Encode::utf8 |
238 | package Encode::utf8; |
239 | *name = sub{ shift->{'Name'} }; |
240 | *new_sequence = sub{ return $_[0] }; |
241 | *decode = sub{ |
242 | my ($obj,$octets,$chk) = @_; |
243 | my $str = Encode::decode_utf8($octets); |
244 | if (defined $str) { |
245 | $_[1] = '' if $chk; |
246 | return $str; |
247 | } |
248 | return undef; |
249 | }; |
250 | *encode = sub { |
251 | my ($obj,$string,$chk) = @_; |
252 | my $octets = Encode::encode_utf8($string); |
253 | $_[1] = '' if $chk; |
254 | return $octets; |
255 | }; |
0ab8f81e |
256 | $Encode::Encoding{utf8} = |
c731e18e |
257 | bless {Name => "utf8"} => "Encode::utf8"; |
f2a2953c |
258 | } |
f2a2953c |
259 | } |
260 | |
656753f8 |
261 | 1; |
262 | |
2a936312 |
263 | __END__ |
264 | |
4411f3b6 |
265 | =head1 NAME |
266 | |
267 | Encode - character encodings |
268 | |
269 | =head1 SYNOPSIS |
270 | |
271 | use Encode; |
272 | |
67d7b5ef |
273 | =head2 Table of Contents |
274 | |
0ab8f81e |
275 | Encode consists of a collection of modules whose details are too big |
67d7b5ef |
276 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 |
277 | and general topics at a glance. For other topics and more details, |
0ab8f81e |
278 | see the PODs below: |
67d7b5ef |
279 | |
280 | Name Description |
281 | -------------------------------------------------------- |
6d1c0808 |
282 | Encode::Alias Alias definitions to encodings |
67d7b5ef |
283 | Encode::Encoding Encode Implementation Base Class |
284 | Encode::Supported List of Supported Encodings |
285 | Encode::CN Simplified Chinese Encodings |
286 | Encode::JP Japanese Encodings |
287 | Encode::KR Korean Encodings |
288 | Encode::TW Traditional Chinese Encodings |
289 | -------------------------------------------------------- |
290 | |
4411f3b6 |
291 | =head1 DESCRIPTION |
292 | |
47bfe92f |
293 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef |
294 | and the rest of the system. Perl strings are sequences of |
295 | B<characters>. |
296 | |
297 | The repertoire of characters that Perl can represent is at least that |
298 | defined by the Unicode Consortium. On most platforms the ordinal |
299 | values of the characters (as returned by C<ord(ch)>) is the "Unicode |
300 | codepoint" for the character (the exceptions are those platforms where |
301 | the legacy encoding is some variant of EBCDIC rather than a super-set |
302 | of ASCII - see L<perlebcdic>). |
303 | |
0ab8f81e |
304 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef |
305 | often called "bytes". These chunks are also known as "octets" in |
306 | networking standards. Perl is widely used to manipulate data of many |
307 | types - not only strings of characters representing human or computer |
0ab8f81e |
308 | languages but also "binary" data being the machine's representation of |
67d7b5ef |
309 | numbers, pixels in an image - or just about anything. |
310 | |
0ab8f81e |
311 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef |
312 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e |
313 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef |
314 | "logical character". |
315 | |
316 | =head2 TERMINOLOGY |
4411f3b6 |
317 | |
67d7b5ef |
318 | =over 4 |
21938dfa |
319 | |
67d7b5ef |
320 | =item * |
321 | |
322 | I<character>: a character in the range 0..(2**32-1) (or more). |
323 | (What Perl's strings are made of.) |
324 | |
325 | =item * |
326 | |
327 | I<byte>: a character in the range 0..255 |
328 | (A special case of a Perl character.) |
329 | |
330 | =item * |
331 | |
332 | I<octet>: 8 bits of data, with ordinal values 0..255 |
0ab8f81e |
333 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef |
334 | |
335 | =back |
4411f3b6 |
336 | |
67d7b5ef |
337 | The marker [INTERNAL] marks Internal Implementation Details, in |
338 | general meant only for those who think they know what they are doing, |
339 | and such details may change in future releases. |
340 | |
341 | =head1 PERL ENCODING API |
4411f3b6 |
342 | |
343 | =over 4 |
344 | |
f2a2953c |
345 | =item $octets = encode(ENCODING, $string[, CHECK]) |
4411f3b6 |
346 | |
0ab8f81e |
347 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef |
348 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e |
349 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
350 | For CHECK, see L</"Handling Malformed Data">. |
4411f3b6 |
351 | |
0ab8f81e |
352 | For example, to convert (internally UTF-8 encoded) Unicode string to |
6d1c0808 |
353 | iso-8859-1 (also known as Latin1), |
681a7c68 |
354 | |
67d7b5ef |
355 | $octets = encode("iso-8859-1", $unicode); |
681a7c68 |
356 | |
f2a2953c |
357 | =item $string = decode(ENCODING, $octets[, CHECK]) |
4411f3b6 |
358 | |
0ab8f81e |
359 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's |
360 | internal form and returns the resulting string. As in encode(), |
361 | ENCODING can be either a canonical name or an alias. For encoding names |
362 | and aliases, see L</"Defining Aliases">. For CHECK, see |
47bfe92f |
363 | L</"Handling Malformed Data">. |
364 | |
0ab8f81e |
365 | For example, to convert ISO-8859-1 data to UTF-8: |
681a7c68 |
366 | |
67d7b5ef |
367 | $utf8 = decode("iso-8859-1", $latin1); |
681a7c68 |
368 | |
f2a2953c |
369 | =item [$length =] from_to($string, FROM_ENCODING, TO_ENCODING [,CHECK]) |
47bfe92f |
370 | |
0ab8f81e |
371 | Converts B<in-place> data between two encodings. |
372 | For example, to convert ISO-8859-1 data to UTF-8: |
2b106fbe |
373 | |
374 | from_to($data, "iso-8859-1", "utf-8"); |
375 | |
376 | and to convert it back: |
377 | |
378 | from_to($data, "utf-8", "iso-8859-1"); |
4411f3b6 |
379 | |
ab97ca19 |
380 | Note that because the conversion happens in place, the data to be |
0ab8f81e |
381 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 |
382 | |
0ab8f81e |
383 | from_to() returns the length of the converted string on success, undef |
3ef515df |
384 | otherwise. |
385 | |
4411f3b6 |
386 | =back |
387 | |
f2a2953c |
388 | =head2 UTF-8 / utf8 |
389 | |
0ab8f81e |
390 | The Unicode Consortium defines the UTF-8 transformation format as a |
391 | way of encoding the entire Unicode repertoire as sequences of octets. |
392 | This encoding is expected to become very widespread. Perl can use this |
393 | form internally to represent strings, so conversions to and from this |
394 | form are particularly efficient (as octets in memory do not have to |
395 | change, just the meta-data that tells Perl how to treat them). |
f2a2953c |
396 | |
397 | =over 4 |
398 | |
399 | =item $octets = encode_utf8($string); |
400 | |
0ab8f81e |
401 | The characters that comprise $string are encoded in Perl's superset of |
402 | UTF-8 and the resulting octets are returned as a sequence of bytes. All |
403 | possible characters have a UTF-8 representation so this function cannot |
404 | fail. |
f2a2953c |
405 | |
406 | =item $string = decode_utf8($octets [, CHECK]); |
407 | |
408 | The sequence of octets represented by $octets is decoded from UTF-8 |
409 | into a sequence of logical characters. Not all sequences of octets |
410 | form valid UTF-8 encodings, so it is possible for this call to fail. |
0ab8f81e |
411 | For CHECK, see L</"Handling Malformed Data">. |
f2a2953c |
412 | |
413 | =back |
414 | |
51ef4e11 |
415 | =head2 Listing available encodings |
416 | |
5129552c |
417 | use Encode; |
418 | @list = Encode->encodings(); |
419 | |
420 | Returns a list of the canonical names of the available encodings that |
421 | are loaded. To get a list of all available encodings including the |
422 | ones that are not loaded yet, say |
423 | |
424 | @all_encodings = Encode->encodings(":all"); |
425 | |
0ab8f81e |
426 | Or you can give the name of a specific module. |
5129552c |
427 | |
c731e18e |
428 | @with_jp = Encode->encodings("Encode::JP"); |
429 | |
430 | When "::" is not in the name, "Encode::" is assumed. |
51ef4e11 |
431 | |
c731e18e |
432 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 |
433 | |
0ab8f81e |
434 | To find out in detail which encodings are supported by this package, |
5d030b67 |
435 | see L<Encode::Supported>. |
51ef4e11 |
436 | |
437 | =head2 Defining Aliases |
438 | |
0ab8f81e |
439 | To add a new alias to a given encoding, use: |
67d7b5ef |
440 | |
5129552c |
441 | use Encode; |
442 | use Encode::Alias; |
a63c962f |
443 | define_alias(newName => ENCODING); |
51ef4e11 |
444 | |
3ef515df |
445 | After that, newName can be used as an alias for ENCODING. |
f2a2953c |
446 | ENCODING may be either the name of an encoding or an |
447 | I<encoding object> |
51ef4e11 |
448 | |
fcb875d4 |
449 | But before you do so, make sure the alias is nonexistent with |
450 | C<resolve_alias()>, which returns the canonical name thereof. |
451 | i.e. |
452 | |
453 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true |
454 | Encode::resolve_alias("iso-8859-12") # false; nonexistent |
455 | Encode::resolve_alias($name) eq $name # true if $name is canonical |
456 | |
0ab8f81e |
457 | resolve_alias() does not need C<use Encode::Alias>; it can be |
458 | exported via C<use Encode qw(resolve_alias)>. |
fcb875d4 |
459 | |
0ab8f81e |
460 | See L<Encode::Alias> for details. |
51ef4e11 |
461 | |
85982a32 |
462 | =head1 Encoding via PerlIO |
4411f3b6 |
463 | |
0ab8f81e |
464 | If your perl supports I<PerlIO>, you can use a PerlIO layer to decode |
465 | and encode directly via a filehandle. The following two examples |
466 | are totally identical in their functionality. |
4411f3b6 |
467 | |
85982a32 |
468 | # via PerlIO |
469 | open my $in, "<:encoding(shiftjis)", $infile or die; |
470 | open my $out, ">:encoding(euc-jp)", $outfile or die; |
471 | while(<>){ print; } |
8e86646e |
472 | |
85982a32 |
473 | # via from_to |
0ab8f81e |
474 | open my $in, "<", $infile or die; |
475 | open my $out, ">", $outfile or die; |
6d1c0808 |
476 | while(<>){ |
0ab8f81e |
477 | from_to($_, "shiftjis", "euc-jp", 1); |
85982a32 |
478 | } |
4411f3b6 |
479 | |
0ab8f81e |
480 | Unfortunately, there may be encodings are PerlIO-savvy. You can check |
481 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
482 | method. |
483 | |
484 | Encode::perlio_ok("hz"); # False |
485 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available |
486 | |
487 | use Encode qw(perlio_ok); # exported upon request |
488 | perlio_ok("euc-jp") |
4411f3b6 |
489 | |
0ab8f81e |
490 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
491 | except for hz and ISO-2022-kr. See L<Encode::Encoding> for details. |
4411f3b6 |
492 | |
0ab8f81e |
493 | For gory details, see L<Encode::PerlIO>. |
4411f3b6 |
494 | |
85982a32 |
495 | =head1 Handling Malformed Data |
4411f3b6 |
496 | |
85982a32 |
497 | =over 4 |
47bfe92f |
498 | |
0ab8f81e |
499 | The I<CHECK> argument is used as follows. When you omit it, |
500 | the behaviour is the same as if you had passed a value of 0 for |
501 | I<CHECK>. |
47bfe92f |
502 | |
85982a32 |
503 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f |
504 | |
0ab8f81e |
505 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> |
506 | in place of a malformed character. For UCM-based encodings, |
507 | E<lt>subcharE<gt> will be used. For Unicode, "\x{FFFD}" is used. |
508 | If the data is supposed to be UTF-8, an optional lexical warning |
509 | (category utf8) is given. |
e9692b5b |
510 | |
85982a32 |
511 | =item I<CHECK> = Encode::DIE_ON_ERROR (== 1) |
e9692b5b |
512 | |
0ab8f81e |
513 | If I<CHECK> is 1, methods will die immediately with an error |
514 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
515 | fatal error with eval{} unless you really want to let it die on error. |
47bfe92f |
516 | |
85982a32 |
517 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f |
518 | |
85982a32 |
519 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
0ab8f81e |
520 | return the portion of the data that has been processed so far when |
521 | an error occurs. The data argument will be overwritten with |
522 | everything after that point (that is, the unprocessed part of data). |
523 | This is handy when you have to call decode repeatedly in the case |
524 | where your source data may contain partial multi-byte character |
525 | sequences, for example because you are reading with a fixed-width |
526 | buffer. Here is some sample code that does exactly this: |
4411f3b6 |
527 | |
85982a32 |
528 | my $data = ''; |
529 | while(defined(read $fh, $buffer, 256)){ |
0ab8f81e |
530 | # buffer may end in a partial character so we append |
85982a32 |
531 | $data .= $buffer; |
532 | $utf8 .= decode($encoding, $data, ENCODE::FB_QUIET); |
0ab8f81e |
533 | # $data now contains the unprocessed partial character |
85982a32 |
534 | } |
1768d7eb |
535 | |
85982a32 |
536 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef |
537 | |
0ab8f81e |
538 | This is the same as above, except that it warns on error. Handy when |
539 | you are debugging the mode above. |
85982a32 |
540 | |
541 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) |
542 | |
543 | For encodings that are implemented by Encode::XS, CHECK == |
544 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. |
545 | |
0ab8f81e |
546 | When you decode, '\xI<XX>' will be inserted for a malformed character, |
547 | where I<XX> is the hex representation of the octet that could not be |
548 | decoded to utf8. And when you encode, '\x{I<xxxx>}' will be inserted, |
549 | where I<xxxx> is the Unicode ID of the character that cannot be found |
550 | in the character repertoire of the encoding. |
85982a32 |
551 | |
552 | =item The bitmask |
553 | |
0ab8f81e |
554 | These modes are actually set via a bitmask. Here is how the FB_XX |
555 | constants are laid out. You can import the FB_XX constants via |
556 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask |
557 | constants via C<use Encode qw(:fallback_all)>. |
85982a32 |
558 | |
b0b300a3 |
559 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
560 | DIE_ON_ERR 0x0001 X |
561 | WARN_ON_ER 0x0002 X |
562 | RETURN_ON_ERR 0x0004 X X |
563 | LEAVE_SRC 0x0008 |
564 | PERLQQ 0x0100 X |
67d7b5ef |
565 | |
0ab8f81e |
566 | =head2 Unimplemented fallback schemes |
67d7b5ef |
567 | |
0ab8f81e |
568 | In the future, you will be able to use a code reference to a callback |
f2a2953c |
569 | function for the value of I<CHECK> but its API is still undecided. |
67d7b5ef |
570 | |
571 | =head1 Defining Encodings |
572 | |
573 | To define a new encoding, use: |
574 | |
575 | use Encode qw(define_alias); |
576 | define_encoding($object, 'canonicalName' [, alias...]); |
577 | |
578 | I<canonicalName> will be associated with I<$object>. The object |
0ab8f81e |
579 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef |
580 | If more than two arguments are provided then additional |
0ab8f81e |
581 | arguments are taken as aliases for I<$object>, as for C<define_alias>. |
67d7b5ef |
582 | |
f2a2953c |
583 | See L<Encode::Encoding> for more details. |
584 | |
4411f3b6 |
585 | =head1 Messing with Perl's Internals |
586 | |
47bfe92f |
587 | The following API uses parts of Perl's internals in the current |
0ab8f81e |
588 | implementation. As such, they are efficient but may change. |
4411f3b6 |
589 | |
590 | =over 4 |
591 | |
a63c962f |
592 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 |
593 | |
0ab8f81e |
594 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. |
47bfe92f |
595 | If CHECK is true, also checks the data in STRING for being well-formed |
596 | UTF-8. Returns true if successful, false otherwise. |
4411f3b6 |
597 | |
a63c962f |
598 | =item _utf8_on(STRING) |
4411f3b6 |
599 | |
0ab8f81e |
600 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is |
4411f3b6 |
601 | B<not> checked for being well-formed UTF-8. Do not use unless you |
602 | B<know> that the STRING is well-formed UTF-8. Returns the previous |
0ab8f81e |
603 | state of the UTF-8 flag (so please don't treat the return value as |
604 | indicating success or failure), or C<undef> if STRING is not a string. |
4411f3b6 |
605 | |
a63c962f |
606 | =item _utf8_off(STRING) |
4411f3b6 |
607 | |
0ab8f81e |
608 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. |
609 | Returns the previous state of the UTF-8 flag (so please don't treat the |
610 | return value as indicating success or failure), or C<undef> if STRING is |
4411f3b6 |
611 | not a string. |
612 | |
613 | =back |
614 | |
615 | =head1 SEE ALSO |
616 | |
5d030b67 |
617 | L<Encode::Encoding>, |
618 | L<Encode::Supported>, |
6d1c0808 |
619 | L<Encode::PerlIO>, |
5d030b67 |
620 | L<encoding>, |
6d1c0808 |
621 | L<perlebcdic>, |
622 | L<perlfunc/open>, |
623 | L<perlunicode>, |
624 | L<utf8>, |
5d030b67 |
625 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 |
626 | |
85982a32 |
627 | =head1 MAINTAINER |
aae85ceb |
628 | |
629 | This project was originated by Nick Ing-Simmons and later maintained |
0ab8f81e |
630 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full list |
aae85ceb |
631 | of people involved. For any questions, use |
632 | E<lt>perl-unicode@perl.orgE<gt> so others can share. |
633 | |
4411f3b6 |
634 | =cut |