Commit | Line | Data |
10c5ecbb |
1 | # |
5a1dbf39 |
2 | # $Id: Encode.pm,v 2.17 2006/05/09 17:10:42 dankogai Exp dankogai $ |
10c5ecbb |
3 | # |
2c674647 |
4 | package Encode; |
51ef4e11 |
5 | use strict; |
5a1dbf39 |
6 | our $VERSION = sprintf "%d.%02d", q$Revision: 2.17 $ =~ /(\d+)/g; |
8f139f4c |
7 | sub DEBUG () { 0 } |
6d1c0808 |
8 | use XSLoader (); |
d1256cb1 |
9 | XSLoader::load( __PACKAGE__, $VERSION ); |
2c674647 |
10 | |
2c674647 |
11 | require Exporter; |
7e19fb92 |
12 | use base qw/Exporter/; |
2c674647 |
13 | |
4411f3b6 |
14 | # Public, encouraged API is exported by default |
85982a32 |
15 | |
16 | our @EXPORT = qw( |
0a8c69ed |
17 | decode decode_utf8 encode encode_utf8 str2bytes bytes2str |
a0d8a30e |
18 | encodings find_encoding clone_encoding |
4411f3b6 |
19 | ); |
d1256cb1 |
20 | our @FB_FLAGS = qw( |
21 | DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC |
22 | PERLQQ HTMLCREF XMLCREF STOP_AT_PARTIAL |
23 | ); |
24 | our @FB_CONSTS = qw( |
25 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN |
26 | FB_PERLQQ FB_HTMLCREF FB_XMLCREF |
27 | ); |
28 | our @EXPORT_OK = ( |
29 | qw( |
30 | _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit |
31 | is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade |
85982a32 |
32 | ), |
d1256cb1 |
33 | @FB_FLAGS, @FB_CONSTS, |
34 | ); |
85982a32 |
35 | |
d1256cb1 |
36 | our %EXPORT_TAGS = ( |
37 | all => [ @EXPORT, @EXPORT_OK ], |
38 | fallbacks => [@FB_CONSTS], |
39 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], |
40 | ); |
85982a32 |
41 | |
4411f3b6 |
42 | # Documentation moved after __END__ for speed - NI-S |
2c674647 |
43 | |
d1256cb1 |
44 | our $ON_EBCDIC = ( ord("A") == 193 ); |
f2a2953c |
45 | |
5d030b67 |
46 | use Encode::Alias; |
47 | |
5129552c |
48 | # Make a %Encoding package variable to allow a certain amount of cheating |
49 | our %Encoding; |
aae85ceb |
50 | our %ExtModule; |
51 | require Encode::Config; |
52 | eval { require Encode::ConfigLocal }; |
5129552c |
53 | |
d1256cb1 |
54 | sub encodings { |
5129552c |
55 | my $class = shift; |
fc17bd48 |
56 | my %enc; |
d1256cb1 |
57 | if ( @_ and $_[0] eq ":all" ) { |
58 | %enc = ( %Encoding, %ExtModule ); |
5129552c |
59 | } |
d1256cb1 |
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 sort { lc $a cmp lc $b } |
70 | grep { !/^(?:Internal|Unicode|Guess)$/o } keys %enc; |
51ef4e11 |
71 | } |
72 | |
d1256cb1 |
73 | sub perlio_ok { |
74 | my $obj = ref( $_[0] ) ? $_[0] : find_encoding( $_[0] ); |
011b2d2f |
75 | $obj->can("perlio_ok") and return $obj->perlio_ok(); |
d1256cb1 |
76 | return 0; # safety net |
85982a32 |
77 | } |
78 | |
d1256cb1 |
79 | sub define_encoding { |
18586f54 |
80 | my $obj = shift; |
81 | my $name = shift; |
5129552c |
82 | $Encoding{$name} = $obj; |
18586f54 |
83 | my $lc = lc($name); |
d1256cb1 |
84 | define_alias( $lc => $obj ) unless $lc eq $name; |
85 | while (@_) { |
86 | my $alias = shift; |
87 | define_alias( $alias, $obj ); |
18586f54 |
88 | } |
89 | return $obj; |
656753f8 |
90 | } |
91 | |
d1256cb1 |
92 | sub getEncoding { |
93 | my ( $class, $name, $skip_external ) = @_; |
10c5ecbb |
94 | |
a0d8a30e |
95 | ref($name) && $name->can('renew') and return $name; |
10c5ecbb |
96 | exists $Encoding{$name} and return $Encoding{$name}; |
18586f54 |
97 | my $lc = lc $name; |
10c5ecbb |
98 | exists $Encoding{$lc} and return $Encoding{$lc}; |
c50d192e |
99 | |
5129552c |
100 | my $oc = $class->find_alias($name); |
10c5ecbb |
101 | defined($oc) and return $oc; |
102 | $lc ne $name and $oc = $class->find_alias($lc); |
103 | defined($oc) and return $oc; |
c50d192e |
104 | |
d1256cb1 |
105 | unless ($skip_external) { |
106 | if ( my $mod = $ExtModule{$name} || $ExtModule{$lc} ) { |
107 | $mod =~ s,::,/,g; |
108 | $mod .= '.pm'; |
109 | eval { require $mod; }; |
110 | exists $Encoding{$name} and return $Encoding{$name}; |
111 | } |
d1ed7747 |
112 | } |
18586f54 |
113 | return; |
656753f8 |
114 | } |
115 | |
d1256cb1 |
116 | sub find_encoding($;$) { |
117 | my ( $name, $skip_external ) = @_; |
118 | return __PACKAGE__->getEncoding( $name, $skip_external ); |
4411f3b6 |
119 | } |
120 | |
d1256cb1 |
121 | sub resolve_alias($) { |
fcb875d4 |
122 | my $obj = find_encoding(shift); |
123 | defined $obj and return $obj->name; |
124 | return; |
125 | } |
126 | |
d1256cb1 |
127 | sub clone_encoding($) { |
a0d8a30e |
128 | my $obj = find_encoding(shift); |
129 | ref $obj or return; |
130 | eval { require Storable }; |
131 | $@ and return; |
132 | return Storable::dclone($obj); |
133 | } |
134 | |
d1256cb1 |
135 | sub encode($$;$) { |
136 | my ( $name, $string, $check ) = @_; |
0f7c507f |
137 | return undef unless defined $string; |
d1256cb1 |
138 | $string .= '' if ref $string; # stringify; |
139 | $check ||= 0; |
18586f54 |
140 | my $enc = find_encoding($name); |
d1256cb1 |
141 | unless ( defined $enc ) { |
142 | require Carp; |
143 | Carp::croak("Unknown encoding '$name'"); |
10c5ecbb |
144 | } |
d1256cb1 |
145 | my $octets = $enc->encode( $string, $check ); |
146 | $_[1] = $string if $check and !( $check & LEAVE_SRC() ); |
18586f54 |
147 | return $octets; |
4411f3b6 |
148 | } |
0a8c69ed |
149 | *str2bytes = \&encode; |
4411f3b6 |
150 | |
d1256cb1 |
151 | sub decode($$;$) { |
152 | my ( $name, $octets, $check ) = @_; |
0f7c507f |
153 | return undef unless defined $octets; |
78589665 |
154 | $octets .= '' if ref $octets; |
d1256cb1 |
155 | $check ||= 0; |
18586f54 |
156 | my $enc = find_encoding($name); |
d1256cb1 |
157 | unless ( defined $enc ) { |
158 | require Carp; |
159 | Carp::croak("Unknown encoding '$name'"); |
10c5ecbb |
160 | } |
d1256cb1 |
161 | my $string = $enc->decode( $octets, $check ); |
162 | $_[1] = $octets if $check and !( $check & LEAVE_SRC() ); |
18586f54 |
163 | return $string; |
4411f3b6 |
164 | } |
0a8c69ed |
165 | *bytes2str = \&decode; |
4411f3b6 |
166 | |
d1256cb1 |
167 | sub from_to($$$;$) { |
168 | my ( $string, $from, $to, $check ) = @_; |
0f7c507f |
169 | return undef unless defined $string; |
d1256cb1 |
170 | $check ||= 0; |
18586f54 |
171 | my $f = find_encoding($from); |
d1256cb1 |
172 | unless ( defined $f ) { |
173 | require Carp; |
174 | Carp::croak("Unknown encoding '$from'"); |
10c5ecbb |
175 | } |
18586f54 |
176 | my $t = find_encoding($to); |
d1256cb1 |
177 | unless ( defined $t ) { |
178 | require Carp; |
179 | Carp::croak("Unknown encoding '$to'"); |
10c5ecbb |
180 | } |
41c240f5 |
181 | my $uni = $f->decode($string); |
d1256cb1 |
182 | $_[0] = $string = $t->encode( $uni, $check ); |
183 | return undef if ( $check && length($uni) ); |
184 | return defined( $_[0] ) ? length($string) : undef; |
4411f3b6 |
185 | } |
186 | |
d1256cb1 |
187 | sub encode_utf8($) { |
18586f54 |
188 | my ($str) = @_; |
c731e18e |
189 | utf8::encode($str); |
18586f54 |
190 | return $str; |
4411f3b6 |
191 | } |
192 | |
d1256cb1 |
193 | sub decode_utf8($;$) { |
194 | my ( $str, $check ) = @_; |
41c240f5 |
195 | return $str if is_utf8($str); |
d1256cb1 |
196 | if ($check) { |
197 | return decode( "utf8", $str, $check ); |
198 | } |
199 | else { |
200 | return decode( "utf8", $str ); |
201 | return $str; |
c2cbba7d |
202 | } |
5ad8ef52 |
203 | } |
204 | |
b536bf57 |
205 | predefine_encodings(1); |
f2a2953c |
206 | |
207 | # |
208 | # This is to restore %Encoding if really needed; |
209 | # |
10c5ecbb |
210 | |
d1256cb1 |
211 | sub predefine_encodings { |
10c5ecbb |
212 | use Encode::Encoding; |
b536bf57 |
213 | no warnings 'redefine'; |
214 | my $use_xs = shift; |
6d1c0808 |
215 | if ($ON_EBCDIC) { |
d1256cb1 |
216 | |
217 | # was in Encode::UTF_EBCDIC |
218 | package Encode::UTF_EBCDIC; |
219 | push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; |
220 | *decode = sub { |
221 | my ( $obj, $str, $chk ) = @_; |
222 | my $res = ''; |
223 | for ( my $i = 0 ; $i < length($str) ; $i++ ) { |
224 | $res .= |
225 | chr( |
226 | utf8::unicode_to_native( ord( substr( $str, $i, 1 ) ) ) |
227 | ); |
228 | } |
229 | $_[1] = '' if $chk; |
230 | return $res; |
231 | }; |
232 | *encode = sub { |
233 | my ( $obj, $str, $chk ) = @_; |
234 | my $res = ''; |
235 | for ( my $i = 0 ; $i < length($str) ; $i++ ) { |
236 | $res .= |
237 | chr( |
238 | utf8::native_to_unicode( ord( substr( $str, $i, 1 ) ) ) |
239 | ); |
240 | } |
241 | $_[1] = '' if $chk; |
242 | return $res; |
243 | }; |
244 | $Encode::Encoding{Unicode} = |
245 | bless { Name => "UTF_EBCDIC" } => "Encode::UTF_EBCDIC"; |
246 | } |
247 | else { |
248 | |
249 | package Encode::Internal; |
250 | push @Encode::Internal::ISA, 'Encode::Encoding'; |
251 | *decode = sub { |
252 | my ( $obj, $str, $chk ) = @_; |
253 | utf8::upgrade($str); |
254 | $_[1] = '' if $chk; |
255 | return $str; |
256 | }; |
257 | *encode = \&decode; |
258 | $Encode::Encoding{Unicode} = |
259 | bless { Name => "Internal" } => "Encode::Internal"; |
f2a2953c |
260 | } |
261 | |
262 | { |
d1256cb1 |
263 | |
264 | # was in Encode::utf8 |
265 | package Encode::utf8; |
266 | push @Encode::utf8::ISA, 'Encode::Encoding'; |
267 | |
268 | # |
269 | if ($use_xs) { |
270 | Encode::DEBUG and warn __PACKAGE__, " XS on"; |
271 | *decode = \&decode_xs; |
272 | *encode = \&encode_xs; |
273 | } |
274 | else { |
275 | Encode::DEBUG and warn __PACKAGE__, " XS off"; |
276 | *decode = sub { |
277 | my ( $obj, $octets, $chk ) = @_; |
278 | my $str = Encode::decode_utf8($octets); |
279 | if ( defined $str ) { |
280 | $_[1] = '' if $chk; |
281 | return $str; |
282 | } |
283 | return undef; |
284 | }; |
285 | *encode = sub { |
286 | my ( $obj, $string, $chk ) = @_; |
287 | my $octets = Encode::encode_utf8($string); |
288 | $_[1] = '' if $chk; |
289 | return $octets; |
290 | }; |
291 | } |
292 | *cat_decode = sub { # ($obj, $dst, $src, $pos, $trm, $chk) |
293 | # currently ignores $chk |
294 | my ( $obj, undef, undef, $pos, $trm ) = @_; |
295 | my ( $rdst, $rsrc, $rpos ) = \@_[ 1, 2, 3 ]; |
296 | use bytes; |
297 | if ( ( my $npos = index( $$rsrc, $trm, $pos ) ) >= 0 ) { |
298 | $$rdst .= |
299 | substr( $$rsrc, $pos, $npos - $pos + length($trm) ); |
300 | $$rpos = $npos + length($trm); |
301 | return 1; |
302 | } |
303 | $$rdst .= substr( $$rsrc, $pos ); |
304 | $$rpos = length($$rsrc); |
305 | return ''; |
306 | }; |
307 | $Encode::Encoding{utf8} = |
308 | bless { Name => "utf8" } => "Encode::utf8"; |
309 | $Encode::Encoding{"utf-8-strict"} = |
310 | bless { Name => "utf-8-strict", strict_utf8 => 1 } => |
311 | "Encode::utf8"; |
f2a2953c |
312 | } |
f2a2953c |
313 | } |
314 | |
656753f8 |
315 | 1; |
316 | |
2a936312 |
317 | __END__ |
318 | |
4411f3b6 |
319 | =head1 NAME |
320 | |
321 | Encode - character encodings |
322 | |
323 | =head1 SYNOPSIS |
324 | |
325 | use Encode; |
326 | |
67d7b5ef |
327 | =head2 Table of Contents |
328 | |
0ab8f81e |
329 | Encode consists of a collection of modules whose details are too big |
67d7b5ef |
330 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 |
331 | and general topics at a glance. For other topics and more details, |
0ab8f81e |
332 | see the PODs below: |
67d7b5ef |
333 | |
334 | Name Description |
335 | -------------------------------------------------------- |
6d1c0808 |
336 | Encode::Alias Alias definitions to encodings |
67d7b5ef |
337 | Encode::Encoding Encode Implementation Base Class |
338 | Encode::Supported List of Supported Encodings |
339 | Encode::CN Simplified Chinese Encodings |
340 | Encode::JP Japanese Encodings |
341 | Encode::KR Korean Encodings |
342 | Encode::TW Traditional Chinese Encodings |
343 | -------------------------------------------------------- |
344 | |
4411f3b6 |
345 | =head1 DESCRIPTION |
346 | |
47bfe92f |
347 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef |
348 | and the rest of the system. Perl strings are sequences of |
349 | B<characters>. |
350 | |
351 | The repertoire of characters that Perl can represent is at least that |
352 | defined by the Unicode Consortium. On most platforms the ordinal |
353 | values of the characters (as returned by C<ord(ch)>) is the "Unicode |
354 | codepoint" for the character (the exceptions are those platforms where |
355 | the legacy encoding is some variant of EBCDIC rather than a super-set |
356 | of ASCII - see L<perlebcdic>). |
357 | |
0ab8f81e |
358 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef |
359 | often called "bytes". These chunks are also known as "octets" in |
360 | networking standards. Perl is widely used to manipulate data of many |
361 | types - not only strings of characters representing human or computer |
0ab8f81e |
362 | languages but also "binary" data being the machine's representation of |
67d7b5ef |
363 | numbers, pixels in an image - or just about anything. |
364 | |
0ab8f81e |
365 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef |
366 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e |
367 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef |
368 | "logical character". |
369 | |
370 | =head2 TERMINOLOGY |
4411f3b6 |
371 | |
7e19fb92 |
372 | =over 2 |
21938dfa |
373 | |
67d7b5ef |
374 | =item * |
375 | |
376 | I<character>: a character in the range 0..(2**32-1) (or more). |
377 | (What Perl's strings are made of.) |
378 | |
379 | =item * |
380 | |
381 | I<byte>: a character in the range 0..255 |
382 | (A special case of a Perl character.) |
383 | |
384 | =item * |
385 | |
386 | I<octet>: 8 bits of data, with ordinal values 0..255 |
0ab8f81e |
387 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef |
388 | |
389 | =back |
4411f3b6 |
390 | |
67d7b5ef |
391 | =head1 PERL ENCODING API |
4411f3b6 |
392 | |
7e19fb92 |
393 | =over 2 |
4411f3b6 |
394 | |
b7a5c9de |
395 | =item $octets = encode(ENCODING, $string [, CHECK]) |
4411f3b6 |
396 | |
0ab8f81e |
397 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef |
398 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e |
399 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
400 | For CHECK, see L</"Handling Malformed Data">. |
4411f3b6 |
401 | |
b7a5c9de |
402 | For example, to convert a string from Perl's internal format to |
6d1c0808 |
403 | iso-8859-1 (also known as Latin1), |
681a7c68 |
404 | |
b7a5c9de |
405 | $octets = encode("iso-8859-1", $string); |
7e19fb92 |
406 | |
b7a5c9de |
407 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets |
408 | B<may not be equal to> $string. Though they both contain the same data, the utf8 flag |
7e19fb92 |
409 | for $octets is B<always> off. When you encode anything, utf8 flag of |
410 | the result is always off, even when it contains completely valid utf8 |
411 | string. See L</"The UTF-8 flag"> below. |
681a7c68 |
412 | |
7f0d54d7 |
413 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
414 | |
b7a5c9de |
415 | =item $string = decode(ENCODING, $octets [, CHECK]) |
4411f3b6 |
416 | |
0ab8f81e |
417 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's |
418 | internal form and returns the resulting string. As in encode(), |
419 | ENCODING can be either a canonical name or an alias. For encoding names |
420 | and aliases, see L</"Defining Aliases">. For CHECK, see |
47bfe92f |
421 | L</"Handling Malformed Data">. |
422 | |
b7a5c9de |
423 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: |
681a7c68 |
424 | |
b7a5c9de |
425 | $string = decode("iso-8859-1", $octets); |
681a7c68 |
426 | |
b7a5c9de |
427 | B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string |
428 | B<may not be equal to> $octets. Though they both contain the same data, |
429 | the utf8 flag for $string is on unless $octets entirely consists of |
7e19fb92 |
430 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag"> |
431 | below. |
47bfe92f |
432 | |
7f0d54d7 |
433 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
434 | |
b7a5c9de |
435 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) |
7e19fb92 |
436 | |
b7a5c9de |
437 | Converts B<in-place> data between two encodings. The data in $octets |
438 | must be encoded as octets and not as characters in Perl's internal |
f9d05ba3 |
439 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 |
440 | encoding: |
2b106fbe |
441 | |
b7a5c9de |
442 | from_to($octets, "iso-8859-1", "cp1250"); |
2b106fbe |
443 | |
444 | and to convert it back: |
445 | |
b7a5c9de |
446 | from_to($octets, "cp1250", "iso-8859-1"); |
4411f3b6 |
447 | |
ab97ca19 |
448 | Note that because the conversion happens in place, the data to be |
0ab8f81e |
449 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 |
450 | |
f9d05ba3 |
451 | from_to() returns the length of the converted string in octets on |
452 | success, I<undef> on error. |
3ef515df |
453 | |
b7a5c9de |
454 | B<CAVEAT>: The following operations look the same but are not quite so; |
7e19fb92 |
455 | |
b7a5c9de |
456 | from_to($data, "iso-8859-1", "utf8"); #1 |
7e19fb92 |
457 | $data = decode("iso-8859-1", $data); #2 |
4411f3b6 |
458 | |
b7a5c9de |
459 | Both #1 and #2 make $data consist of a completely valid UTF-8 string |
7e19fb92 |
460 | but only #2 turns utf8 flag on. #1 is equivalent to |
f2a2953c |
461 | |
7e19fb92 |
462 | $data = encode("utf8", decode("iso-8859-1", $data)); |
f2a2953c |
463 | |
7e19fb92 |
464 | See L</"The UTF-8 flag"> below. |
f2a2953c |
465 | |
466 | =item $octets = encode_utf8($string); |
467 | |
7e19fb92 |
468 | Equivalent to C<$octets = encode("utf8", $string);> The characters |
b7a5c9de |
469 | that comprise $string are encoded in Perl's internal format and the |
470 | result is returned as a sequence of octets. All possible |
7e19fb92 |
471 | characters have a UTF-8 representation so this function cannot fail. |
472 | |
f2a2953c |
473 | |
474 | =item $string = decode_utf8($octets [, CHECK]); |
475 | |
7e19fb92 |
476 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. |
b7a5c9de |
477 | The sequence of octets represented by |
7e19fb92 |
478 | $octets is decoded from UTF-8 into a sequence of logical |
479 | characters. Not all sequences of octets form valid UTF-8 encodings, so |
480 | it is possible for this call to fail. For CHECK, see |
481 | L</"Handling Malformed Data">. |
f2a2953c |
482 | |
483 | =back |
484 | |
51ef4e11 |
485 | =head2 Listing available encodings |
486 | |
5129552c |
487 | use Encode; |
488 | @list = Encode->encodings(); |
489 | |
490 | Returns a list of the canonical names of the available encodings that |
491 | are loaded. To get a list of all available encodings including the |
492 | ones that are not loaded yet, say |
493 | |
494 | @all_encodings = Encode->encodings(":all"); |
495 | |
0ab8f81e |
496 | Or you can give the name of a specific module. |
5129552c |
497 | |
c731e18e |
498 | @with_jp = Encode->encodings("Encode::JP"); |
499 | |
500 | When "::" is not in the name, "Encode::" is assumed. |
51ef4e11 |
501 | |
c731e18e |
502 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 |
503 | |
0ab8f81e |
504 | To find out in detail which encodings are supported by this package, |
5d030b67 |
505 | see L<Encode::Supported>. |
51ef4e11 |
506 | |
507 | =head2 Defining Aliases |
508 | |
0ab8f81e |
509 | To add a new alias to a given encoding, use: |
67d7b5ef |
510 | |
5129552c |
511 | use Encode; |
512 | use Encode::Alias; |
a63c962f |
513 | define_alias(newName => ENCODING); |
51ef4e11 |
514 | |
3ef515df |
515 | After that, newName can be used as an alias for ENCODING. |
f2a2953c |
516 | ENCODING may be either the name of an encoding or an |
517 | I<encoding object> |
51ef4e11 |
518 | |
fcb875d4 |
519 | But before you do so, make sure the alias is nonexistent with |
520 | C<resolve_alias()>, which returns the canonical name thereof. |
521 | i.e. |
522 | |
523 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true |
524 | Encode::resolve_alias("iso-8859-12") # false; nonexistent |
525 | Encode::resolve_alias($name) eq $name # true if $name is canonical |
526 | |
0ab8f81e |
527 | resolve_alias() does not need C<use Encode::Alias>; it can be |
528 | exported via C<use Encode qw(resolve_alias)>. |
fcb875d4 |
529 | |
0ab8f81e |
530 | See L<Encode::Alias> for details. |
51ef4e11 |
531 | |
85982a32 |
532 | =head1 Encoding via PerlIO |
4411f3b6 |
533 | |
b7a5c9de |
534 | If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode |
0ab8f81e |
535 | and encode directly via a filehandle. The following two examples |
536 | are totally identical in their functionality. |
4411f3b6 |
537 | |
85982a32 |
538 | # via PerlIO |
539 | open my $in, "<:encoding(shiftjis)", $infile or die; |
540 | open my $out, ">:encoding(euc-jp)", $outfile or die; |
b7a5c9de |
541 | while(<$in>){ print $out $_; } |
8e86646e |
542 | |
85982a32 |
543 | # via from_to |
0ab8f81e |
544 | open my $in, "<", $infile or die; |
545 | open my $out, ">", $outfile or die; |
b7a5c9de |
546 | while(<$in>){ |
0ab8f81e |
547 | from_to($_, "shiftjis", "euc-jp", 1); |
b7a5c9de |
548 | print $out $_; |
85982a32 |
549 | } |
4411f3b6 |
550 | |
b7a5c9de |
551 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check |
0ab8f81e |
552 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
553 | method. |
554 | |
555 | Encode::perlio_ok("hz"); # False |
556 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available |
557 | |
558 | use Encode qw(perlio_ok); # exported upon request |
559 | perlio_ok("euc-jp") |
4411f3b6 |
560 | |
0ab8f81e |
561 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
f9d05ba3 |
562 | except for hz and ISO-2022-kr. For gory details, see |
563 | L<Encode::Encoding> and L<Encode::PerlIO>. |
4411f3b6 |
564 | |
85982a32 |
565 | =head1 Handling Malformed Data |
4411f3b6 |
566 | |
8e180e82 |
567 | The optional I<CHECK> argument tells Encode what to do when it |
568 | encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 ) |
569 | is assumed. |
570 | |
571 | As of version 2.12 Encode supports coderef values for CHECK. See below. |
f9d05ba3 |
572 | |
573 | =over 2 |
574 | |
3c4b39be |
575 | =item B<NOTE:> Not all encoding support this feature |
f9d05ba3 |
576 | |
577 | Some encodings ignore I<CHECK> argument. For example, |
578 | L<Encode::Unicode> ignores I<CHECK> and it always croaks on error. |
579 | |
580 | =back |
581 | |
582 | Now here is the list of I<CHECK> values available |
47bfe92f |
583 | |
151b5d36 |
584 | =over 2 |
585 | |
85982a32 |
586 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f |
587 | |
f9d05ba3 |
588 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> in |
78589665 |
589 | place of a malformed character. When you encode, E<lt>subcharE<gt> |
590 | will be used. When you decode the code point C<0xFFFD> is used. If |
591 | the data is supposed to be UTF-8, an optional lexical warning |
592 | (category utf8) is given. |
e9692b5b |
593 | |
7e19fb92 |
594 | =item I<CHECK> = Encode::FB_CROAK ( == 1) |
e9692b5b |
595 | |
b7a5c9de |
596 | If I<CHECK> is 1, methods will die on error immediately with an error |
0ab8f81e |
597 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
f9d05ba3 |
598 | error with eval{} unless you really want to let it die. |
47bfe92f |
599 | |
85982a32 |
600 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f |
601 | |
85982a32 |
602 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
f9d05ba3 |
603 | return the portion of the data that has been processed so far when an |
604 | error occurs. The data argument will be overwritten with everything |
605 | after that point (that is, the unprocessed part of data). This is |
606 | handy when you have to call decode repeatedly in the case where your |
607 | source data may contain partial multi-byte character sequences, |
608 | (i.e. you are reading with a fixed-width buffer). Here is a sample |
609 | code that does exactly this: |
4411f3b6 |
610 | |
78589665 |
611 | my $buffer = ''; my $string = ''; |
612 | while(read $fh, $buffer, 256, length($buffer)){ |
613 | $string .= decode($encoding, $buffer, Encode::FB_QUIET); |
614 | # $buffer now contains the unprocessed partial character |
85982a32 |
615 | } |
1768d7eb |
616 | |
85982a32 |
617 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef |
618 | |
0ab8f81e |
619 | This is the same as above, except that it warns on error. Handy when |
620 | you are debugging the mode above. |
85982a32 |
621 | |
622 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) |
623 | |
af1f55d9 |
624 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) |
625 | |
626 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) |
627 | |
85982a32 |
628 | For encodings that are implemented by Encode::XS, CHECK == |
629 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. |
630 | |
b7a5c9de |
631 | When you decode, C<\xI<HH>> will be inserted for a malformed character, |
632 | where I<HH> is the hex representation of the octet that could not be |
633 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, |
634 | where I<HHHH> is the Unicode ID of the character that cannot be found |
0ab8f81e |
635 | in the character repertoire of the encoding. |
85982a32 |
636 | |
af1f55d9 |
637 | HTML/XML character reference modes are about the same, in place of |
78589665 |
638 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and |
639 | XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number. |
af1f55d9 |
640 | |
7f0d54d7 |
641 | In Encode 2.10 or later, C<LEAVE_SRC> is also implied. |
642 | |
85982a32 |
643 | =item The bitmask |
644 | |
0ab8f81e |
645 | These modes are actually set via a bitmask. Here is how the FB_XX |
646 | constants are laid out. You can import the FB_XX constants via |
647 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask |
648 | constants via C<use Encode qw(:fallback_all)>. |
85982a32 |
649 | |
b0b300a3 |
650 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
651 | DIE_ON_ERR 0x0001 X |
4089adc4 |
652 | WARN_ON_ERR 0x0002 X |
b0b300a3 |
653 | RETURN_ON_ERR 0x0004 X X |
7f0d54d7 |
654 | LEAVE_SRC 0x0008 X |
b0b300a3 |
655 | PERLQQ 0x0100 X |
b7a5c9de |
656 | HTMLCREF 0x0200 |
657 | XMLCREF 0x0400 |
67d7b5ef |
658 | |
151b5d36 |
659 | =back |
660 | |
8e180e82 |
661 | =head2 coderef for CHECK |
662 | |
663 | As of Encode 2.12 CHECK can also be a code reference which takes the |
664 | ord value of unmapped caharacter as an argument and returns a string |
665 | that represents the fallback character. For instance, |
67d7b5ef |
666 | |
8e180e82 |
667 | $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift }); |
67d7b5ef |
668 | |
8e180e82 |
669 | Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of |
670 | \x{I<XXXX>}. |
982a4085 |
671 | |
67d7b5ef |
672 | =head1 Defining Encodings |
673 | |
674 | To define a new encoding, use: |
675 | |
b7a5c9de |
676 | use Encode qw(define_encoding); |
67d7b5ef |
677 | define_encoding($object, 'canonicalName' [, alias...]); |
678 | |
679 | I<canonicalName> will be associated with I<$object>. The object |
0ab8f81e |
680 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef |
681 | If more than two arguments are provided then additional |
b7a5c9de |
682 | arguments are taken as aliases for I<$object>. |
67d7b5ef |
683 | |
f2a2953c |
684 | See L<Encode::Encoding> for more details. |
685 | |
7e19fb92 |
686 | =head1 The UTF-8 flag |
687 | |
688 | Before the introduction of utf8 support in perl, The C<eq> operator |
b7a5c9de |
689 | just compared the strings represented by two scalars. Beginning with |
690 | perl 5.8, C<eq> compares two strings with simultaneous consideration |
691 | of I<the utf8 flag>. To explain why we made it so, I will quote page |
692 | 402 of C<Programming Perl, 3rd ed.> |
7e19fb92 |
693 | |
694 | =over 2 |
695 | |
696 | =item Goal #1: |
697 | |
698 | Old byte-oriented programs should not spontaneously break on the old |
699 | byte-oriented data they used to work on. |
700 | |
701 | =item Goal #2: |
702 | |
703 | Old byte-oriented programs should magically start working on the new |
704 | character-oriented data when appropriate. |
705 | |
706 | =item Goal #3: |
707 | |
708 | Programs should run just as fast in the new character-oriented mode |
709 | as in the old byte-oriented mode. |
710 | |
711 | =item Goal #4: |
712 | |
713 | Perl should remain one language, rather than forking into a |
714 | byte-oriented Perl and a character-oriented Perl. |
715 | |
716 | =back |
717 | |
718 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 |
719 | was born and many features documented in the book remained |
b7a5c9de |
720 | unimplemented for a long time. Perl 5.8 corrected this and the introduction |
721 | of the UTF-8 flag is one of them. You can think of this perl notion as of a |
722 | byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8 |
7e19fb92 |
723 | flag on). |
724 | |
725 | Here is how Encode takes care of the utf8 flag. |
726 | |
4bdf5738 |
727 | =over 2 |
7e19fb92 |
728 | |
729 | =item * |
730 | |
731 | When you encode, the resulting utf8 flag is always off. |
732 | |
151b5d36 |
733 | =item * |
7e19fb92 |
734 | |
b7a5c9de |
735 | When you decode, the resulting utf8 flag is on unless you can |
7e19fb92 |
736 | unambiguously represent data. Here is the definition of |
737 | dis-ambiguity. |
738 | |
b7a5c9de |
739 | After C<$utf8 = decode('foo', $octet);>, |
7e19fb92 |
740 | |
741 | When $octet is... The utf8 flag in $utf8 is |
742 | --------------------------------------------- |
743 | In ASCII only (or EBCDIC only) OFF |
744 | In ISO-8859-1 ON |
745 | In any other Encoding ON |
746 | --------------------------------------------- |
747 | |
3c4b39be |
748 | As you see, there is one exception, In ASCII. That way you can assume |
7e19fb92 |
749 | Goal #1. And with Encode Goal #2 is assumed but you still have to be |
750 | careful in such cases mentioned in B<CAVEAT> paragraphs. |
751 | |
752 | This utf8 flag is not visible in perl scripts, exactly for the same |
753 | reason you cannot (or you I<don't have to>) see if a scalar contains a |
754 | string, integer, or floating point number. But you can still peek |
755 | and poke these if you will. See the section below. |
756 | |
757 | =back |
758 | |
759 | =head2 Messing with Perl's Internals |
4411f3b6 |
760 | |
47bfe92f |
761 | The following API uses parts of Perl's internals in the current |
0ab8f81e |
762 | implementation. As such, they are efficient but may change. |
4411f3b6 |
763 | |
7e19fb92 |
764 | =over 2 |
4411f3b6 |
765 | |
a63c962f |
766 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 |
767 | |
0ab8f81e |
768 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. |
47bfe92f |
769 | If CHECK is true, also checks the data in STRING for being well-formed |
770 | UTF-8. Returns true if successful, false otherwise. |
4411f3b6 |
771 | |
2c246b25 |
772 | As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). |
b5ab1f6f |
773 | |
a63c962f |
774 | =item _utf8_on(STRING) |
4411f3b6 |
775 | |
0ab8f81e |
776 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is |
4411f3b6 |
777 | B<not> checked for being well-formed UTF-8. Do not use unless you |
778 | B<know> that the STRING is well-formed UTF-8. Returns the previous |
0ab8f81e |
779 | state of the UTF-8 flag (so please don't treat the return value as |
780 | indicating success or failure), or C<undef> if STRING is not a string. |
4411f3b6 |
781 | |
a63c962f |
782 | =item _utf8_off(STRING) |
4411f3b6 |
783 | |
0ab8f81e |
784 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. |
785 | Returns the previous state of the UTF-8 flag (so please don't treat the |
786 | return value as indicating success or failure), or C<undef> if STRING is |
4411f3b6 |
787 | not a string. |
788 | |
789 | =back |
790 | |
7f0d54d7 |
791 | =head1 UTF-8 vs. utf8 |
792 | |
793 | ....We now view strings not as sequences of bytes, but as sequences |
794 | of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit |
795 | computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed. |
796 | |
797 | That has been the perl's notion of UTF-8 but official UTF-8 is more |
798 | strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are |
799 | not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al). |
800 | |
801 | Now that is overruled by Larry Wall himself. |
802 | |
803 | From: Larry Wall <larry@wall.org> |
804 | Date: December 04, 2004 11:51:58 JST |
805 | To: perl-unicode@perl.org |
806 | Subject: Re: Make Encode.pm support the real UTF-8 |
807 | Message-Id: <20041204025158.GA28754@wall.org> |
808 | |
809 | On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote: |
810 | : I've no problem with 'utf8' being perl's unrestricted uft8 encoding, |
811 | : but "UTF-8" is the name of the standard and should give the |
812 | : corresponding behaviour. |
813 | |
814 | For what it's worth, that's how I've always kept them straight in my |
815 | head. |
8e180e82 |
816 | |
7f0d54d7 |
817 | Also for what it's worth, Perl 6 will mostly default to strict but |
818 | make it easy to switch back to lax. |
819 | |
820 | Larry |
821 | |
822 | Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8 |
823 | while B<utf8> means liberal, lax, version thereof. And Encode version |
824 | 2.10 or later thus groks the difference between C<UTF-8> and C"utf8". |
825 | |
826 | encode("utf8", "\x{FFFF_FFFF}", 1); # okay |
827 | encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks |
828 | |
829 | C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>. |
830 | Yes, the hyphen between "UTF" and "8" is important. Without it Encode |
831 | goes "liberal" |
832 | |
833 | find_encoding("UTF-8")->name # is 'utf-8-strict' |
834 | find_encoding("utf-8")->name # ditto. names are case insensitive |
50c1ac04 |
835 | find_encoding("utf_8")->name # ditto. "_" are treated as "-" |
7f0d54d7 |
836 | find_encoding("UTF8")->name # is 'utf8'. |
837 | |
838 | |
4411f3b6 |
839 | =head1 SEE ALSO |
840 | |
5d030b67 |
841 | L<Encode::Encoding>, |
842 | L<Encode::Supported>, |
6d1c0808 |
843 | L<Encode::PerlIO>, |
5d030b67 |
844 | L<encoding>, |
6d1c0808 |
845 | L<perlebcdic>, |
846 | L<perlfunc/open>, |
847 | L<perlunicode>, |
848 | L<utf8>, |
5d030b67 |
849 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 |
850 | |
85982a32 |
851 | =head1 MAINTAINER |
aae85ceb |
852 | |
853 | This project was originated by Nick Ing-Simmons and later maintained |
7e19fb92 |
854 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full |
855 | list of people involved. For any questions, use |
b7a5c9de |
856 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. |
aae85ceb |
857 | |
d1256cb1 |
858 | While Dan Kogai retains the copyright as a maintainer, the credit |
859 | should go to all those involoved. See AUTHORS for those submitted |
860 | codes. |
861 | |
862 | =head1 COPYRIGHT |
863 | |
864 | Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt> |
865 | |
866 | This library is free software; you can redistribute it and/or modify |
867 | it under the same terms as Perl itself. |
868 | |
4411f3b6 |
869 | =cut |