Commit | Line | Data |
10c5ecbb |
1 | # |
656ebd29 |
2 | # $Id: Encode.pm,v 2.18 2006/06/03 20:28:48 dankogai Exp dankogai $ |
10c5ecbb |
3 | # |
2c674647 |
4 | package Encode; |
51ef4e11 |
5 | use strict; |
656ebd29 |
6 | use warnings; |
4438fa18 |
7 | our $VERSION = "2.18_01"; |
8f139f4c |
8 | sub DEBUG () { 0 } |
6d1c0808 |
9 | use XSLoader (); |
d1256cb1 |
10 | XSLoader::load( __PACKAGE__, $VERSION ); |
2c674647 |
11 | |
2c674647 |
12 | require Exporter; |
7e19fb92 |
13 | use base qw/Exporter/; |
2c674647 |
14 | |
4411f3b6 |
15 | # Public, encouraged API is exported by default |
85982a32 |
16 | |
17 | our @EXPORT = qw( |
0a8c69ed |
18 | decode decode_utf8 encode encode_utf8 str2bytes bytes2str |
a0d8a30e |
19 | encodings find_encoding clone_encoding |
4411f3b6 |
20 | ); |
d1256cb1 |
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 |
85982a32 |
33 | ), |
d1256cb1 |
34 | @FB_FLAGS, @FB_CONSTS, |
35 | ); |
85982a32 |
36 | |
d1256cb1 |
37 | our %EXPORT_TAGS = ( |
38 | all => [ @EXPORT, @EXPORT_OK ], |
39 | fallbacks => [@FB_CONSTS], |
40 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], |
41 | ); |
85982a32 |
42 | |
4411f3b6 |
43 | # Documentation moved after __END__ for speed - NI-S |
2c674647 |
44 | |
d1256cb1 |
45 | our $ON_EBCDIC = ( ord("A") == 193 ); |
f2a2953c |
46 | |
5d030b67 |
47 | use Encode::Alias; |
48 | |
5129552c |
49 | # Make a %Encoding package variable to allow a certain amount of cheating |
50 | our %Encoding; |
aae85ceb |
51 | our %ExtModule; |
52 | require Encode::Config; |
53 | eval { require Encode::ConfigLocal }; |
5129552c |
54 | |
d1256cb1 |
55 | sub encodings { |
5129552c |
56 | my $class = shift; |
fc17bd48 |
57 | my %enc; |
d1256cb1 |
58 | if ( @_ and $_[0] eq ":all" ) { |
59 | %enc = ( %Encoding, %ExtModule ); |
5129552c |
60 | } |
d1256cb1 |
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; |
51ef4e11 |
72 | } |
73 | |
d1256cb1 |
74 | sub perlio_ok { |
75 | my $obj = ref( $_[0] ) ? $_[0] : find_encoding( $_[0] ); |
011b2d2f |
76 | $obj->can("perlio_ok") and return $obj->perlio_ok(); |
d1256cb1 |
77 | return 0; # safety net |
85982a32 |
78 | } |
79 | |
d1256cb1 |
80 | sub define_encoding { |
18586f54 |
81 | my $obj = shift; |
82 | my $name = shift; |
5129552c |
83 | $Encoding{$name} = $obj; |
18586f54 |
84 | my $lc = lc($name); |
d1256cb1 |
85 | define_alias( $lc => $obj ) unless $lc eq $name; |
86 | while (@_) { |
87 | my $alias = shift; |
88 | define_alias( $alias, $obj ); |
18586f54 |
89 | } |
90 | return $obj; |
656753f8 |
91 | } |
92 | |
d1256cb1 |
93 | sub getEncoding { |
94 | my ( $class, $name, $skip_external ) = @_; |
10c5ecbb |
95 | |
a0d8a30e |
96 | ref($name) && $name->can('renew') and return $name; |
10c5ecbb |
97 | exists $Encoding{$name} and return $Encoding{$name}; |
18586f54 |
98 | my $lc = lc $name; |
10c5ecbb |
99 | exists $Encoding{$lc} and return $Encoding{$lc}; |
c50d192e |
100 | |
5129552c |
101 | my $oc = $class->find_alias($name); |
10c5ecbb |
102 | defined($oc) and return $oc; |
103 | $lc ne $name and $oc = $class->find_alias($lc); |
104 | defined($oc) and return $oc; |
c50d192e |
105 | |
d1256cb1 |
106 | unless ($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 | } |
d1ed7747 |
113 | } |
18586f54 |
114 | return; |
656753f8 |
115 | } |
116 | |
d1256cb1 |
117 | sub find_encoding($;$) { |
118 | my ( $name, $skip_external ) = @_; |
119 | return __PACKAGE__->getEncoding( $name, $skip_external ); |
4411f3b6 |
120 | } |
121 | |
d1256cb1 |
122 | sub resolve_alias($) { |
fcb875d4 |
123 | my $obj = find_encoding(shift); |
124 | defined $obj and return $obj->name; |
125 | return; |
126 | } |
127 | |
d1256cb1 |
128 | sub clone_encoding($) { |
a0d8a30e |
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 | |
d1256cb1 |
136 | sub encode($$;$) { |
137 | my ( $name, $string, $check ) = @_; |
0f7c507f |
138 | return undef unless defined $string; |
d1256cb1 |
139 | $string .= '' if ref $string; # stringify; |
140 | $check ||= 0; |
18586f54 |
141 | my $enc = find_encoding($name); |
d1256cb1 |
142 | unless ( defined $enc ) { |
143 | require Carp; |
144 | Carp::croak("Unknown encoding '$name'"); |
10c5ecbb |
145 | } |
d1256cb1 |
146 | my $octets = $enc->encode( $string, $check ); |
147 | $_[1] = $string if $check and !( $check & LEAVE_SRC() ); |
18586f54 |
148 | return $octets; |
4411f3b6 |
149 | } |
0a8c69ed |
150 | *str2bytes = \&encode; |
4411f3b6 |
151 | |
d1256cb1 |
152 | sub decode($$;$) { |
153 | my ( $name, $octets, $check ) = @_; |
0f7c507f |
154 | return undef unless defined $octets; |
78589665 |
155 | $octets .= '' if ref $octets; |
d1256cb1 |
156 | $check ||= 0; |
18586f54 |
157 | my $enc = find_encoding($name); |
d1256cb1 |
158 | unless ( defined $enc ) { |
159 | require Carp; |
160 | Carp::croak("Unknown encoding '$name'"); |
10c5ecbb |
161 | } |
d1256cb1 |
162 | my $string = $enc->decode( $octets, $check ); |
163 | $_[1] = $octets if $check and !( $check & LEAVE_SRC() ); |
18586f54 |
164 | return $string; |
4411f3b6 |
165 | } |
0a8c69ed |
166 | *bytes2str = \&decode; |
4411f3b6 |
167 | |
d1256cb1 |
168 | sub from_to($$$;$) { |
169 | my ( $string, $from, $to, $check ) = @_; |
0f7c507f |
170 | return undef unless defined $string; |
d1256cb1 |
171 | $check ||= 0; |
18586f54 |
172 | my $f = find_encoding($from); |
d1256cb1 |
173 | unless ( defined $f ) { |
174 | require Carp; |
175 | Carp::croak("Unknown encoding '$from'"); |
10c5ecbb |
176 | } |
18586f54 |
177 | my $t = find_encoding($to); |
d1256cb1 |
178 | unless ( defined $t ) { |
179 | require Carp; |
180 | Carp::croak("Unknown encoding '$to'"); |
10c5ecbb |
181 | } |
41c240f5 |
182 | my $uni = $f->decode($string); |
d1256cb1 |
183 | $_[0] = $string = $t->encode( $uni, $check ); |
184 | return undef if ( $check && length($uni) ); |
185 | return defined( $_[0] ) ? length($string) : undef; |
4411f3b6 |
186 | } |
187 | |
d1256cb1 |
188 | sub encode_utf8($) { |
18586f54 |
189 | my ($str) = @_; |
c731e18e |
190 | utf8::encode($str); |
18586f54 |
191 | return $str; |
4411f3b6 |
192 | } |
193 | |
d1256cb1 |
194 | sub decode_utf8($;$) { |
195 | my ( $str, $check ) = @_; |
41c240f5 |
196 | return $str if is_utf8($str); |
d1256cb1 |
197 | if ($check) { |
198 | return decode( "utf8", $str, $check ); |
199 | } |
200 | else { |
201 | return decode( "utf8", $str ); |
202 | return $str; |
c2cbba7d |
203 | } |
5ad8ef52 |
204 | } |
205 | |
b536bf57 |
206 | predefine_encodings(1); |
f2a2953c |
207 | |
208 | # |
209 | # This is to restore %Encoding if really needed; |
210 | # |
10c5ecbb |
211 | |
d1256cb1 |
212 | sub predefine_encodings { |
10c5ecbb |
213 | use Encode::Encoding; |
b536bf57 |
214 | no warnings 'redefine'; |
215 | my $use_xs = shift; |
6d1c0808 |
216 | if ($ON_EBCDIC) { |
d1256cb1 |
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"; |
f2a2953c |
261 | } |
262 | |
263 | { |
d1256cb1 |
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"; |
f2a2953c |
313 | } |
f2a2953c |
314 | } |
315 | |
656753f8 |
316 | 1; |
317 | |
2a936312 |
318 | __END__ |
319 | |
4411f3b6 |
320 | =head1 NAME |
321 | |
322 | Encode - character encodings |
323 | |
324 | =head1 SYNOPSIS |
325 | |
326 | use Encode; |
327 | |
67d7b5ef |
328 | =head2 Table of Contents |
329 | |
0ab8f81e |
330 | Encode consists of a collection of modules whose details are too big |
67d7b5ef |
331 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 |
332 | and general topics at a glance. For other topics and more details, |
0ab8f81e |
333 | see the PODs below: |
67d7b5ef |
334 | |
335 | Name Description |
336 | -------------------------------------------------------- |
6d1c0808 |
337 | Encode::Alias Alias definitions to encodings |
67d7b5ef |
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 | |
4411f3b6 |
346 | =head1 DESCRIPTION |
347 | |
47bfe92f |
348 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef |
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 | |
0ab8f81e |
359 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef |
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 |
0ab8f81e |
363 | languages but also "binary" data being the machine's representation of |
67d7b5ef |
364 | numbers, pixels in an image - or just about anything. |
365 | |
0ab8f81e |
366 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef |
367 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e |
368 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef |
369 | "logical character". |
370 | |
371 | =head2 TERMINOLOGY |
4411f3b6 |
372 | |
7e19fb92 |
373 | =over 2 |
21938dfa |
374 | |
67d7b5ef |
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 |
0ab8f81e |
388 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef |
389 | |
390 | =back |
4411f3b6 |
391 | |
67d7b5ef |
392 | =head1 PERL ENCODING API |
4411f3b6 |
393 | |
7e19fb92 |
394 | =over 2 |
4411f3b6 |
395 | |
b7a5c9de |
396 | =item $octets = encode(ENCODING, $string [, CHECK]) |
4411f3b6 |
397 | |
0ab8f81e |
398 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef |
399 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e |
400 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
401 | For CHECK, see L</"Handling Malformed Data">. |
4411f3b6 |
402 | |
b7a5c9de |
403 | For example, to convert a string from Perl's internal format to |
6d1c0808 |
404 | iso-8859-1 (also known as Latin1), |
681a7c68 |
405 | |
b7a5c9de |
406 | $octets = encode("iso-8859-1", $string); |
7e19fb92 |
407 | |
b7a5c9de |
408 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets |
409 | B<may not be equal to> $string. Though they both contain the same data, the utf8 flag |
7e19fb92 |
410 | for $octets is B<always> off. When you encode anything, utf8 flag of |
411 | the result is always off, even when it contains completely valid utf8 |
412 | string. See L</"The UTF-8 flag"> below. |
681a7c68 |
413 | |
7f0d54d7 |
414 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
415 | |
b7a5c9de |
416 | =item $string = decode(ENCODING, $octets [, CHECK]) |
4411f3b6 |
417 | |
0ab8f81e |
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 |
47bfe92f |
422 | L</"Handling Malformed Data">. |
423 | |
b7a5c9de |
424 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: |
681a7c68 |
425 | |
b7a5c9de |
426 | $string = decode("iso-8859-1", $octets); |
681a7c68 |
427 | |
b7a5c9de |
428 | B<CAVEAT>: When you run C<$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 |
7e19fb92 |
431 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag"> |
432 | below. |
47bfe92f |
433 | |
7f0d54d7 |
434 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
435 | |
b7a5c9de |
436 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) |
7e19fb92 |
437 | |
b7a5c9de |
438 | Converts B<in-place> data between two encodings. The data in $octets |
439 | must be encoded as octets and not as characters in Perl's internal |
f9d05ba3 |
440 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 |
441 | encoding: |
2b106fbe |
442 | |
b7a5c9de |
443 | from_to($octets, "iso-8859-1", "cp1250"); |
2b106fbe |
444 | |
445 | and to convert it back: |
446 | |
b7a5c9de |
447 | from_to($octets, "cp1250", "iso-8859-1"); |
4411f3b6 |
448 | |
ab97ca19 |
449 | Note that because the conversion happens in place, the data to be |
0ab8f81e |
450 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 |
451 | |
f9d05ba3 |
452 | from_to() returns the length of the converted string in octets on |
453 | success, I<undef> on error. |
3ef515df |
454 | |
b7a5c9de |
455 | B<CAVEAT>: The following operations look the same but are not quite so; |
7e19fb92 |
456 | |
b7a5c9de |
457 | from_to($data, "iso-8859-1", "utf8"); #1 |
7e19fb92 |
458 | $data = decode("iso-8859-1", $data); #2 |
4411f3b6 |
459 | |
b7a5c9de |
460 | Both #1 and #2 make $data consist of a completely valid UTF-8 string |
7e19fb92 |
461 | but only #2 turns utf8 flag on. #1 is equivalent to |
f2a2953c |
462 | |
7e19fb92 |
463 | $data = encode("utf8", decode("iso-8859-1", $data)); |
f2a2953c |
464 | |
7e19fb92 |
465 | See L</"The UTF-8 flag"> below. |
f2a2953c |
466 | |
467 | =item $octets = encode_utf8($string); |
468 | |
7e19fb92 |
469 | Equivalent to C<$octets = encode("utf8", $string);> The characters |
b7a5c9de |
470 | that comprise $string are encoded in Perl's internal format and the |
471 | result is returned as a sequence of octets. All possible |
7e19fb92 |
472 | characters have a UTF-8 representation so this function cannot fail. |
473 | |
f2a2953c |
474 | |
475 | =item $string = decode_utf8($octets [, CHECK]); |
476 | |
7e19fb92 |
477 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. |
b7a5c9de |
478 | The sequence of octets represented by |
7e19fb92 |
479 | $octets is decoded from UTF-8 into a sequence of logical |
480 | characters. Not all sequences of octets form valid UTF-8 encodings, so |
481 | it is possible for this call to fail. For CHECK, see |
482 | L</"Handling Malformed Data">. |
f2a2953c |
483 | |
484 | =back |
485 | |
51ef4e11 |
486 | =head2 Listing available encodings |
487 | |
5129552c |
488 | use Encode; |
489 | @list = Encode->encodings(); |
490 | |
491 | Returns a list of the canonical names of the available encodings that |
492 | are loaded. To get a list of all available encodings including the |
493 | ones that are not loaded yet, say |
494 | |
495 | @all_encodings = Encode->encodings(":all"); |
496 | |
0ab8f81e |
497 | Or you can give the name of a specific module. |
5129552c |
498 | |
c731e18e |
499 | @with_jp = Encode->encodings("Encode::JP"); |
500 | |
501 | When "::" is not in the name, "Encode::" is assumed. |
51ef4e11 |
502 | |
c731e18e |
503 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 |
504 | |
0ab8f81e |
505 | To find out in detail which encodings are supported by this package, |
5d030b67 |
506 | see L<Encode::Supported>. |
51ef4e11 |
507 | |
508 | =head2 Defining Aliases |
509 | |
0ab8f81e |
510 | To add a new alias to a given encoding, use: |
67d7b5ef |
511 | |
5129552c |
512 | use Encode; |
513 | use Encode::Alias; |
a63c962f |
514 | define_alias(newName => ENCODING); |
51ef4e11 |
515 | |
3ef515df |
516 | After that, newName can be used as an alias for ENCODING. |
f2a2953c |
517 | ENCODING may be either the name of an encoding or an |
518 | I<encoding object> |
51ef4e11 |
519 | |
fcb875d4 |
520 | But before you do so, make sure the alias is nonexistent with |
521 | C<resolve_alias()>, which returns the canonical name thereof. |
522 | i.e. |
523 | |
524 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true |
525 | Encode::resolve_alias("iso-8859-12") # false; nonexistent |
526 | Encode::resolve_alias($name) eq $name # true if $name is canonical |
527 | |
0ab8f81e |
528 | resolve_alias() does not need C<use Encode::Alias>; it can be |
529 | exported via C<use Encode qw(resolve_alias)>. |
fcb875d4 |
530 | |
0ab8f81e |
531 | See L<Encode::Alias> for details. |
51ef4e11 |
532 | |
85982a32 |
533 | =head1 Encoding via PerlIO |
4411f3b6 |
534 | |
b7a5c9de |
535 | If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode |
0ab8f81e |
536 | and encode directly via a filehandle. The following two examples |
537 | are totally identical in their functionality. |
4411f3b6 |
538 | |
85982a32 |
539 | # via PerlIO |
540 | open my $in, "<:encoding(shiftjis)", $infile or die; |
541 | open my $out, ">:encoding(euc-jp)", $outfile or die; |
b7a5c9de |
542 | while(<$in>){ print $out $_; } |
8e86646e |
543 | |
85982a32 |
544 | # via from_to |
0ab8f81e |
545 | open my $in, "<", $infile or die; |
546 | open my $out, ">", $outfile or die; |
b7a5c9de |
547 | while(<$in>){ |
0ab8f81e |
548 | from_to($_, "shiftjis", "euc-jp", 1); |
b7a5c9de |
549 | print $out $_; |
85982a32 |
550 | } |
4411f3b6 |
551 | |
b7a5c9de |
552 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check |
0ab8f81e |
553 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
554 | method. |
555 | |
556 | Encode::perlio_ok("hz"); # False |
557 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available |
558 | |
559 | use Encode qw(perlio_ok); # exported upon request |
560 | perlio_ok("euc-jp") |
4411f3b6 |
561 | |
0ab8f81e |
562 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
f9d05ba3 |
563 | except for hz and ISO-2022-kr. For gory details, see |
564 | L<Encode::Encoding> and L<Encode::PerlIO>. |
4411f3b6 |
565 | |
85982a32 |
566 | =head1 Handling Malformed Data |
4411f3b6 |
567 | |
8e180e82 |
568 | The optional I<CHECK> argument tells Encode what to do when it |
569 | encounters malformed data. Without CHECK, Encode::FB_DEFAULT ( == 0 ) |
570 | is assumed. |
571 | |
572 | As of version 2.12 Encode supports coderef values for CHECK. See below. |
f9d05ba3 |
573 | |
574 | =over 2 |
575 | |
3c4b39be |
576 | =item B<NOTE:> Not all encoding support this feature |
f9d05ba3 |
577 | |
578 | Some encodings ignore I<CHECK> argument. For example, |
579 | L<Encode::Unicode> ignores I<CHECK> and it always croaks on error. |
580 | |
581 | =back |
582 | |
583 | Now here is the list of I<CHECK> values available |
47bfe92f |
584 | |
151b5d36 |
585 | =over 2 |
586 | |
85982a32 |
587 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f |
588 | |
f9d05ba3 |
589 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> in |
78589665 |
590 | place of a malformed character. When you encode, E<lt>subcharE<gt> |
591 | will be used. When you decode the code point C<0xFFFD> is used. If |
592 | the data is supposed to be UTF-8, an optional lexical warning |
593 | (category utf8) is given. |
e9692b5b |
594 | |
7e19fb92 |
595 | =item I<CHECK> = Encode::FB_CROAK ( == 1) |
e9692b5b |
596 | |
b7a5c9de |
597 | If I<CHECK> is 1, methods will die on error immediately with an error |
0ab8f81e |
598 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
f9d05ba3 |
599 | error with eval{} unless you really want to let it die. |
47bfe92f |
600 | |
85982a32 |
601 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f |
602 | |
85982a32 |
603 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
f9d05ba3 |
604 | return the portion of the data that has been processed so far when an |
605 | error occurs. The data argument will be overwritten with everything |
606 | after that point (that is, the unprocessed part of data). This is |
607 | handy when you have to call decode repeatedly in the case where your |
608 | source data may contain partial multi-byte character sequences, |
609 | (i.e. you are reading with a fixed-width buffer). Here is a sample |
610 | code that does exactly this: |
4411f3b6 |
611 | |
78589665 |
612 | my $buffer = ''; my $string = ''; |
613 | while(read $fh, $buffer, 256, length($buffer)){ |
614 | $string .= decode($encoding, $buffer, Encode::FB_QUIET); |
615 | # $buffer now contains the unprocessed partial character |
85982a32 |
616 | } |
1768d7eb |
617 | |
85982a32 |
618 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef |
619 | |
0ab8f81e |
620 | This is the same as above, except that it warns on error. Handy when |
621 | you are debugging the mode above. |
85982a32 |
622 | |
623 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) |
624 | |
af1f55d9 |
625 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) |
626 | |
627 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) |
628 | |
85982a32 |
629 | For encodings that are implemented by Encode::XS, CHECK == |
630 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. |
631 | |
b7a5c9de |
632 | When you decode, C<\xI<HH>> will be inserted for a malformed character, |
633 | where I<HH> is the hex representation of the octet that could not be |
634 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, |
635 | where I<HHHH> is the Unicode ID of the character that cannot be found |
0ab8f81e |
636 | in the character repertoire of the encoding. |
85982a32 |
637 | |
af1f55d9 |
638 | HTML/XML character reference modes are about the same, in place of |
78589665 |
639 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and |
640 | XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number. |
af1f55d9 |
641 | |
7f0d54d7 |
642 | In Encode 2.10 or later, C<LEAVE_SRC> is also implied. |
643 | |
85982a32 |
644 | =item The bitmask |
645 | |
0ab8f81e |
646 | These modes are actually set via a bitmask. Here is how the FB_XX |
647 | constants are laid out. You can import the FB_XX constants via |
648 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask |
649 | constants via C<use Encode qw(:fallback_all)>. |
85982a32 |
650 | |
b0b300a3 |
651 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
652 | DIE_ON_ERR 0x0001 X |
4089adc4 |
653 | WARN_ON_ERR 0x0002 X |
b0b300a3 |
654 | RETURN_ON_ERR 0x0004 X X |
7f0d54d7 |
655 | LEAVE_SRC 0x0008 X |
b0b300a3 |
656 | PERLQQ 0x0100 X |
b7a5c9de |
657 | HTMLCREF 0x0200 |
658 | XMLCREF 0x0400 |
67d7b5ef |
659 | |
151b5d36 |
660 | =back |
661 | |
8e180e82 |
662 | =head2 coderef for CHECK |
663 | |
664 | As of Encode 2.12 CHECK can also be a code reference which takes the |
665 | ord value of unmapped caharacter as an argument and returns a string |
666 | that represents the fallback character. For instance, |
67d7b5ef |
667 | |
8e180e82 |
668 | $ascii = encode("ascii", $utf8, sub{ sprintf "<U+%04X>", shift }); |
67d7b5ef |
669 | |
8e180e82 |
670 | Acts like FB_PERLQQ but E<lt>U+I<XXXX>E<gt> is used instead of |
671 | \x{I<XXXX>}. |
982a4085 |
672 | |
67d7b5ef |
673 | =head1 Defining Encodings |
674 | |
675 | To define a new encoding, use: |
676 | |
b7a5c9de |
677 | use Encode qw(define_encoding); |
67d7b5ef |
678 | define_encoding($object, 'canonicalName' [, alias...]); |
679 | |
680 | I<canonicalName> will be associated with I<$object>. The object |
0ab8f81e |
681 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef |
682 | If more than two arguments are provided then additional |
b7a5c9de |
683 | arguments are taken as aliases for I<$object>. |
67d7b5ef |
684 | |
f2a2953c |
685 | See L<Encode::Encoding> for more details. |
686 | |
7e19fb92 |
687 | =head1 The UTF-8 flag |
688 | |
689 | Before the introduction of utf8 support in perl, The C<eq> operator |
b7a5c9de |
690 | just compared the strings represented by two scalars. Beginning with |
691 | perl 5.8, C<eq> compares two strings with simultaneous consideration |
692 | of I<the utf8 flag>. To explain why we made it so, I will quote page |
693 | 402 of C<Programming Perl, 3rd ed.> |
7e19fb92 |
694 | |
695 | =over 2 |
696 | |
697 | =item Goal #1: |
698 | |
699 | Old byte-oriented programs should not spontaneously break on the old |
700 | byte-oriented data they used to work on. |
701 | |
702 | =item Goal #2: |
703 | |
704 | Old byte-oriented programs should magically start working on the new |
705 | character-oriented data when appropriate. |
706 | |
707 | =item Goal #3: |
708 | |
709 | Programs should run just as fast in the new character-oriented mode |
710 | as in the old byte-oriented mode. |
711 | |
712 | =item Goal #4: |
713 | |
714 | Perl should remain one language, rather than forking into a |
715 | byte-oriented Perl and a character-oriented Perl. |
716 | |
717 | =back |
718 | |
719 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 |
720 | was born and many features documented in the book remained |
b7a5c9de |
721 | unimplemented for a long time. Perl 5.8 corrected this and the introduction |
722 | of the UTF-8 flag is one of them. You can think of this perl notion as of a |
723 | byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8 |
7e19fb92 |
724 | flag on). |
725 | |
726 | Here is how Encode takes care of the utf8 flag. |
727 | |
4bdf5738 |
728 | =over 2 |
7e19fb92 |
729 | |
730 | =item * |
731 | |
732 | When you encode, the resulting utf8 flag is always off. |
733 | |
151b5d36 |
734 | =item * |
7e19fb92 |
735 | |
b7a5c9de |
736 | When you decode, the resulting utf8 flag is on unless you can |
7e19fb92 |
737 | unambiguously represent data. Here is the definition of |
738 | dis-ambiguity. |
739 | |
b7a5c9de |
740 | After C<$utf8 = decode('foo', $octet);>, |
7e19fb92 |
741 | |
742 | When $octet is... The utf8 flag in $utf8 is |
743 | --------------------------------------------- |
744 | In ASCII only (or EBCDIC only) OFF |
745 | In ISO-8859-1 ON |
746 | In any other Encoding ON |
747 | --------------------------------------------- |
748 | |
3c4b39be |
749 | As you see, there is one exception, In ASCII. That way you can assume |
7e19fb92 |
750 | Goal #1. And with Encode Goal #2 is assumed but you still have to be |
751 | careful in such cases mentioned in B<CAVEAT> paragraphs. |
752 | |
753 | This utf8 flag is not visible in perl scripts, exactly for the same |
754 | reason you cannot (or you I<don't have to>) see if a scalar contains a |
755 | string, integer, or floating point number. But you can still peek |
756 | and poke these if you will. See the section below. |
757 | |
758 | =back |
759 | |
760 | =head2 Messing with Perl's Internals |
4411f3b6 |
761 | |
47bfe92f |
762 | The following API uses parts of Perl's internals in the current |
0ab8f81e |
763 | implementation. As such, they are efficient but may change. |
4411f3b6 |
764 | |
7e19fb92 |
765 | =over 2 |
4411f3b6 |
766 | |
a63c962f |
767 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 |
768 | |
0ab8f81e |
769 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. |
47bfe92f |
770 | If CHECK is true, also checks the data in STRING for being well-formed |
771 | UTF-8. Returns true if successful, false otherwise. |
4411f3b6 |
772 | |
2c246b25 |
773 | As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). |
b5ab1f6f |
774 | |
a63c962f |
775 | =item _utf8_on(STRING) |
4411f3b6 |
776 | |
0ab8f81e |
777 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is |
4411f3b6 |
778 | B<not> checked for being well-formed UTF-8. Do not use unless you |
779 | B<know> that the STRING is well-formed UTF-8. Returns the previous |
0ab8f81e |
780 | state of the UTF-8 flag (so please don't treat the return value as |
781 | indicating success or failure), or C<undef> if STRING is not a string. |
4411f3b6 |
782 | |
a63c962f |
783 | =item _utf8_off(STRING) |
4411f3b6 |
784 | |
0ab8f81e |
785 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. |
786 | Returns the previous state of the UTF-8 flag (so please don't treat the |
787 | return value as indicating success or failure), or C<undef> if STRING is |
4411f3b6 |
788 | not a string. |
789 | |
790 | =back |
791 | |
7f0d54d7 |
792 | =head1 UTF-8 vs. utf8 |
793 | |
794 | ....We now view strings not as sequences of bytes, but as sequences |
795 | of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit |
796 | computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed. |
797 | |
798 | That has been the perl's notion of UTF-8 but official UTF-8 is more |
799 | strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are |
800 | not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al). |
801 | |
802 | Now that is overruled by Larry Wall himself. |
803 | |
804 | From: Larry Wall <larry@wall.org> |
805 | Date: December 04, 2004 11:51:58 JST |
806 | To: perl-unicode@perl.org |
807 | Subject: Re: Make Encode.pm support the real UTF-8 |
808 | Message-Id: <20041204025158.GA28754@wall.org> |
809 | |
810 | On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote: |
811 | : I've no problem with 'utf8' being perl's unrestricted uft8 encoding, |
812 | : but "UTF-8" is the name of the standard and should give the |
813 | : corresponding behaviour. |
814 | |
815 | For what it's worth, that's how I've always kept them straight in my |
816 | head. |
8e180e82 |
817 | |
7f0d54d7 |
818 | Also for what it's worth, Perl 6 will mostly default to strict but |
819 | make it easy to switch back to lax. |
820 | |
821 | Larry |
822 | |
823 | Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8 |
824 | while B<utf8> means liberal, lax, version thereof. And Encode version |
825 | 2.10 or later thus groks the difference between C<UTF-8> and C"utf8". |
826 | |
827 | encode("utf8", "\x{FFFF_FFFF}", 1); # okay |
828 | encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks |
829 | |
830 | C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>. |
831 | Yes, the hyphen between "UTF" and "8" is important. Without it Encode |
832 | goes "liberal" |
833 | |
834 | find_encoding("UTF-8")->name # is 'utf-8-strict' |
835 | find_encoding("utf-8")->name # ditto. names are case insensitive |
50c1ac04 |
836 | find_encoding("utf_8")->name # ditto. "_" are treated as "-" |
7f0d54d7 |
837 | find_encoding("UTF8")->name # is 'utf8'. |
838 | |
839 | |
4411f3b6 |
840 | =head1 SEE ALSO |
841 | |
5d030b67 |
842 | L<Encode::Encoding>, |
843 | L<Encode::Supported>, |
6d1c0808 |
844 | L<Encode::PerlIO>, |
5d030b67 |
845 | L<encoding>, |
6d1c0808 |
846 | L<perlebcdic>, |
847 | L<perlfunc/open>, |
848 | L<perlunicode>, |
849 | L<utf8>, |
5d030b67 |
850 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 |
851 | |
85982a32 |
852 | =head1 MAINTAINER |
aae85ceb |
853 | |
854 | This project was originated by Nick Ing-Simmons and later maintained |
7e19fb92 |
855 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full |
856 | list of people involved. For any questions, use |
b7a5c9de |
857 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. |
aae85ceb |
858 | |
d1256cb1 |
859 | While Dan Kogai retains the copyright as a maintainer, the credit |
860 | should go to all those involoved. See AUTHORS for those submitted |
861 | codes. |
862 | |
863 | =head1 COPYRIGHT |
864 | |
865 | Copyright 2002-2006 Dan Kogai E<lt>dankogai@dan.co.jpE<gt> |
866 | |
867 | This library is free software; you can redistribute it and/or modify |
868 | it under the same terms as Perl itself. |
869 | |
4411f3b6 |
870 | =cut |