Commit | Line | Data |
10c5ecbb |
1 | # |
78589665 |
2 | # $Id: Encode.pm,v 2.9 2004/12/03 19:16:40 dankogai Exp $ |
10c5ecbb |
3 | # |
2c674647 |
4 | package Encode; |
51ef4e11 |
5 | use strict; |
7f0d54d7 |
6 | # our $VERSION = do { my @r = (q$Revision: 2.9 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r }; |
b162af07 |
7 | our $VERSION = '2.0902_01'; |
8f139f4c |
8 | sub DEBUG () { 0 } |
6d1c0808 |
9 | use XSLoader (); |
10c5ecbb |
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( |
18 | decode decode_utf8 encode encode_utf8 |
a0d8a30e |
19 | encodings find_encoding clone_encoding |
4411f3b6 |
20 | ); |
21 | |
b7a5c9de |
22 | our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC |
af1f55d9 |
23 | PERLQQ HTMLCREF XMLCREF); |
b7a5c9de |
24 | our @FB_CONSTS = qw(FB_DEFAULT FB_CROAK FB_QUIET FB_WARN |
af1f55d9 |
25 | FB_PERLQQ FB_HTMLCREF FB_XMLCREF); |
85982a32 |
26 | |
51ef4e11 |
27 | our @EXPORT_OK = |
6d1c0808 |
28 | ( |
85982a32 |
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 |
32 | ), |
33 | @FB_FLAGS, @FB_CONSTS, |
34 | ); |
35 | |
6d1c0808 |
36 | our %EXPORT_TAGS = |
85982a32 |
37 | ( |
38 | all => [ @EXPORT, @EXPORT_OK ], |
39 | fallbacks => [ @FB_CONSTS ], |
40 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], |
41 | ); |
42 | |
4411f3b6 |
43 | # Documentation moved after __END__ for speed - NI-S |
2c674647 |
44 | |
a63c962f |
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 | |
656753f8 |
55 | sub encodings |
56 | { |
5129552c |
57 | my $class = shift; |
fc17bd48 |
58 | my %enc; |
59 | if (@_ and $_[0] eq ":all"){ |
60 | %enc = ( %Encoding, %ExtModule ); |
61 | }else{ |
62 | %enc = %Encoding; |
63 | for my $mod (map {m/::/o ? $_ : "Encode::$_" } @_){ |
8f139f4c |
64 | DEBUG and warn $mod; |
fc17bd48 |
65 | for my $enc (keys %ExtModule){ |
66 | $ExtModule{$enc} eq $mod and $enc{$enc} = $mod; |
67 | } |
68 | } |
5129552c |
69 | } |
70 | return |
ce912cd4 |
71 | sort { lc $a cmp lc $b } |
fc17bd48 |
72 | grep {!/^(?:Internal|Unicode|Guess)$/o} keys %enc; |
51ef4e11 |
73 | } |
74 | |
85982a32 |
75 | sub perlio_ok{ |
0ab8f81e |
76 | my $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]); |
011b2d2f |
77 | $obj->can("perlio_ok") and return $obj->perlio_ok(); |
0ab8f81e |
78 | return 0; # safety net |
85982a32 |
79 | } |
80 | |
51ef4e11 |
81 | sub define_encoding |
82 | { |
18586f54 |
83 | my $obj = shift; |
84 | my $name = shift; |
5129552c |
85 | $Encoding{$name} = $obj; |
18586f54 |
86 | my $lc = lc($name); |
87 | define_alias($lc => $obj) unless $lc eq $name; |
10c5ecbb |
88 | while (@_){ |
18586f54 |
89 | my $alias = shift; |
10c5ecbb |
90 | define_alias($alias, $obj); |
18586f54 |
91 | } |
92 | return $obj; |
656753f8 |
93 | } |
94 | |
656753f8 |
95 | sub getEncoding |
96 | { |
10c5ecbb |
97 | my ($class, $name, $skip_external) = @_; |
98 | |
a0d8a30e |
99 | ref($name) && $name->can('renew') and return $name; |
10c5ecbb |
100 | exists $Encoding{$name} and return $Encoding{$name}; |
18586f54 |
101 | my $lc = lc $name; |
10c5ecbb |
102 | exists $Encoding{$lc} and return $Encoding{$lc}; |
c50d192e |
103 | |
5129552c |
104 | my $oc = $class->find_alias($name); |
10c5ecbb |
105 | defined($oc) and return $oc; |
106 | $lc ne $name and $oc = $class->find_alias($lc); |
107 | defined($oc) and return $oc; |
c50d192e |
108 | |
c731e18e |
109 | unless ($skip_external) |
d1ed7747 |
110 | { |
c731e18e |
111 | if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){ |
112 | $mod =~ s,::,/,g ; $mod .= '.pm'; |
113 | eval{ require $mod; }; |
10c5ecbb |
114 | exists $Encoding{$name} and return $Encoding{$name}; |
c731e18e |
115 | } |
d1ed7747 |
116 | } |
18586f54 |
117 | return; |
656753f8 |
118 | } |
119 | |
a0d8a30e |
120 | sub find_encoding($;$) |
4411f3b6 |
121 | { |
10c5ecbb |
122 | my ($name, $skip_external) = @_; |
dd9703c9 |
123 | return __PACKAGE__->getEncoding($name,$skip_external); |
4411f3b6 |
124 | } |
125 | |
a0d8a30e |
126 | sub resolve_alias($){ |
fcb875d4 |
127 | my $obj = find_encoding(shift); |
128 | defined $obj and return $obj->name; |
129 | return; |
130 | } |
131 | |
a0d8a30e |
132 | sub clone_encoding($){ |
133 | my $obj = find_encoding(shift); |
134 | ref $obj or return; |
135 | eval { require Storable }; |
136 | $@ and return; |
137 | return Storable::dclone($obj); |
138 | } |
139 | |
b2704119 |
140 | sub encode($$;$) |
4411f3b6 |
141 | { |
e8c86ba6 |
142 | my ($name, $string, $check) = @_; |
0f7c507f |
143 | return undef unless defined $string; |
78589665 |
144 | $string .= '' if ref $string; # stringify; |
b2704119 |
145 | $check ||=0; |
18586f54 |
146 | my $enc = find_encoding($name); |
10c5ecbb |
147 | unless(defined $enc){ |
148 | require Carp; |
149 | Carp::croak("Unknown encoding '$name'"); |
150 | } |
18586f54 |
151 | my $octets = $enc->encode($string,$check); |
7f0d54d7 |
152 | $_[1] = $string if $check and !($check & LEAVE_SRC()); |
18586f54 |
153 | return $octets; |
4411f3b6 |
154 | } |
155 | |
b2704119 |
156 | sub decode($$;$) |
4411f3b6 |
157 | { |
18586f54 |
158 | my ($name,$octets,$check) = @_; |
0f7c507f |
159 | return undef unless defined $octets; |
78589665 |
160 | $octets .= '' if ref $octets; |
b2704119 |
161 | $check ||=0; |
18586f54 |
162 | my $enc = find_encoding($name); |
10c5ecbb |
163 | unless(defined $enc){ |
164 | require Carp; |
165 | Carp::croak("Unknown encoding '$name'"); |
166 | } |
18586f54 |
167 | my $string = $enc->decode($octets,$check); |
7f0d54d7 |
168 | $_[1] = $octets if $check and !($check & LEAVE_SRC()); |
18586f54 |
169 | return $string; |
4411f3b6 |
170 | } |
171 | |
b2704119 |
172 | sub from_to($$$;$) |
4411f3b6 |
173 | { |
18586f54 |
174 | my ($string,$from,$to,$check) = @_; |
0f7c507f |
175 | return undef unless defined $string; |
b2704119 |
176 | $check ||=0; |
18586f54 |
177 | my $f = find_encoding($from); |
10c5ecbb |
178 | unless (defined $f){ |
179 | require Carp; |
180 | Carp::croak("Unknown encoding '$from'"); |
181 | } |
18586f54 |
182 | my $t = find_encoding($to); |
10c5ecbb |
183 | unless (defined $t){ |
184 | require Carp; |
185 | Carp::croak("Unknown encoding '$to'"); |
186 | } |
18586f54 |
187 | my $uni = $f->decode($string,$check); |
188 | return undef if ($check && length($string)); |
a999c27c |
189 | $string = $t->encode($uni,$check); |
18586f54 |
190 | return undef if ($check && length($uni)); |
3ef515df |
191 | return defined($_[0] = $string) ? length($string) : undef ; |
4411f3b6 |
192 | } |
193 | |
b2704119 |
194 | sub encode_utf8($) |
4411f3b6 |
195 | { |
18586f54 |
196 | my ($str) = @_; |
c731e18e |
197 | utf8::encode($str); |
18586f54 |
198 | return $str; |
4411f3b6 |
199 | } |
200 | |
c2cbba7d |
201 | sub decode_utf8($;$) |
4411f3b6 |
202 | { |
c2cbba7d |
203 | my ($str, $check) = @_; |
204 | if ($check){ |
205 | return decode("utf8", $str, $check); |
206 | }else{ |
207 | return undef unless utf8::decode($str); |
208 | return $str; |
209 | } |
5ad8ef52 |
210 | } |
211 | |
b536bf57 |
212 | predefine_encodings(1); |
f2a2953c |
213 | |
214 | # |
215 | # This is to restore %Encoding if really needed; |
216 | # |
10c5ecbb |
217 | |
f2a2953c |
218 | sub predefine_encodings{ |
10c5ecbb |
219 | use Encode::Encoding; |
b536bf57 |
220 | no warnings 'redefine'; |
221 | my $use_xs = shift; |
6d1c0808 |
222 | if ($ON_EBCDIC) { |
f2a2953c |
223 | # was in Encode::UTF_EBCDIC |
224 | package Encode::UTF_EBCDIC; |
10c5ecbb |
225 | push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; |
f2a2953c |
226 | *decode = sub{ |
227 | my ($obj,$str,$chk) = @_; |
228 | my $res = ''; |
229 | for (my $i = 0; $i < length($str); $i++) { |
6d1c0808 |
230 | $res .= |
f2a2953c |
231 | chr(utf8::unicode_to_native(ord(substr($str,$i,1)))); |
232 | } |
233 | $_[1] = '' if $chk; |
234 | return $res; |
235 | }; |
236 | *encode = sub{ |
237 | my ($obj,$str,$chk) = @_; |
238 | my $res = ''; |
239 | for (my $i = 0; $i < length($str); $i++) { |
6d1c0808 |
240 | $res .= |
f2a2953c |
241 | chr(utf8::native_to_unicode(ord(substr($str,$i,1)))); |
242 | } |
243 | $_[1] = '' if $chk; |
244 | return $res; |
245 | }; |
6d1c0808 |
246 | $Encode::Encoding{Unicode} = |
c731e18e |
247 | bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC"; |
6d1c0808 |
248 | } else { |
f2a2953c |
249 | package Encode::Internal; |
10c5ecbb |
250 | push @Encode::Internal::ISA, 'Encode::Encoding'; |
f2a2953c |
251 | *decode = sub{ |
252 | my ($obj,$str,$chk) = @_; |
253 | utf8::upgrade($str); |
254 | $_[1] = '' if $chk; |
255 | return $str; |
256 | }; |
257 | *encode = \&decode; |
6d1c0808 |
258 | $Encode::Encoding{Unicode} = |
c731e18e |
259 | bless {Name => "Internal"} => "Encode::Internal"; |
f2a2953c |
260 | } |
261 | |
262 | { |
263 | # was in Encode::utf8 |
264 | package Encode::utf8; |
10c5ecbb |
265 | push @Encode::utf8::ISA, 'Encode::Encoding'; |
b536bf57 |
266 | # |
267 | if ($use_xs){ |
8f139f4c |
268 | Encode::DEBUG and warn __PACKAGE__, " XS on"; |
b536bf57 |
269 | *decode = \&decode_xs; |
270 | *encode = \&encode_xs; |
271 | }else{ |
8f139f4c |
272 | Encode::DEBUG and warn __PACKAGE__, " XS off"; |
b536bf57 |
273 | *decode = sub{ |
274 | my ($obj,$octets,$chk) = @_; |
275 | my $str = Encode::decode_utf8($octets); |
276 | if (defined $str) { |
277 | $_[1] = '' if $chk; |
278 | return $str; |
279 | } |
280 | return undef; |
281 | }; |
282 | *encode = sub { |
283 | my ($obj,$string,$chk) = @_; |
284 | my $octets = Encode::encode_utf8($string); |
285 | $_[1] = '' if $chk; |
286 | return $octets; |
287 | }; |
288 | } |
220e2d4e |
289 | *cat_decode = sub{ # ($obj, $dst, $src, $pos, $trm, $chk) |
290 | my ($obj, undef, undef, $pos, $trm) = @_; # currently ignores $chk |
291 | my ($rdst, $rsrc, $rpos) = \@_[1,2,3]; |
292 | use bytes; |
293 | if ((my $npos = index($$rsrc, $trm, $pos)) >= 0) { |
294 | $$rdst .= substr($$rsrc, $pos, $npos - $pos + length($trm)); |
295 | $$rpos = $npos + length($trm); |
296 | return 1; |
297 | } |
298 | $$rdst .= substr($$rsrc, $pos); |
299 | $$rpos = length($$rsrc); |
300 | return ''; |
301 | }; |
b7a5c9de |
302 | $Encode::Encoding{utf8} = |
c731e18e |
303 | bless {Name => "utf8"} => "Encode::utf8"; |
7f0d54d7 |
304 | $Encode::Encoding{"utf-8-strict"} = |
305 | bless {Name => "utf-8-strict", strict_utf8 => 1 } => "Encode::utf8"; |
f2a2953c |
306 | } |
f2a2953c |
307 | } |
308 | |
656753f8 |
309 | 1; |
310 | |
2a936312 |
311 | __END__ |
312 | |
4411f3b6 |
313 | =head1 NAME |
314 | |
315 | Encode - character encodings |
316 | |
317 | =head1 SYNOPSIS |
318 | |
319 | use Encode; |
320 | |
67d7b5ef |
321 | =head2 Table of Contents |
322 | |
0ab8f81e |
323 | Encode consists of a collection of modules whose details are too big |
67d7b5ef |
324 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 |
325 | and general topics at a glance. For other topics and more details, |
0ab8f81e |
326 | see the PODs below: |
67d7b5ef |
327 | |
328 | Name Description |
329 | -------------------------------------------------------- |
6d1c0808 |
330 | Encode::Alias Alias definitions to encodings |
67d7b5ef |
331 | Encode::Encoding Encode Implementation Base Class |
332 | Encode::Supported List of Supported Encodings |
333 | Encode::CN Simplified Chinese Encodings |
334 | Encode::JP Japanese Encodings |
335 | Encode::KR Korean Encodings |
336 | Encode::TW Traditional Chinese Encodings |
337 | -------------------------------------------------------- |
338 | |
4411f3b6 |
339 | =head1 DESCRIPTION |
340 | |
47bfe92f |
341 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef |
342 | and the rest of the system. Perl strings are sequences of |
343 | B<characters>. |
344 | |
345 | The repertoire of characters that Perl can represent is at least that |
346 | defined by the Unicode Consortium. On most platforms the ordinal |
347 | values of the characters (as returned by C<ord(ch)>) is the "Unicode |
348 | codepoint" for the character (the exceptions are those platforms where |
349 | the legacy encoding is some variant of EBCDIC rather than a super-set |
350 | of ASCII - see L<perlebcdic>). |
351 | |
0ab8f81e |
352 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef |
353 | often called "bytes". These chunks are also known as "octets" in |
354 | networking standards. Perl is widely used to manipulate data of many |
355 | types - not only strings of characters representing human or computer |
0ab8f81e |
356 | languages but also "binary" data being the machine's representation of |
67d7b5ef |
357 | numbers, pixels in an image - or just about anything. |
358 | |
0ab8f81e |
359 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef |
360 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e |
361 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef |
362 | "logical character". |
363 | |
364 | =head2 TERMINOLOGY |
4411f3b6 |
365 | |
7e19fb92 |
366 | =over 2 |
21938dfa |
367 | |
67d7b5ef |
368 | =item * |
369 | |
370 | I<character>: a character in the range 0..(2**32-1) (or more). |
371 | (What Perl's strings are made of.) |
372 | |
373 | =item * |
374 | |
375 | I<byte>: a character in the range 0..255 |
376 | (A special case of a Perl character.) |
377 | |
378 | =item * |
379 | |
380 | I<octet>: 8 bits of data, with ordinal values 0..255 |
0ab8f81e |
381 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef |
382 | |
383 | =back |
4411f3b6 |
384 | |
67d7b5ef |
385 | =head1 PERL ENCODING API |
4411f3b6 |
386 | |
7e19fb92 |
387 | =over 2 |
4411f3b6 |
388 | |
b7a5c9de |
389 | =item $octets = encode(ENCODING, $string [, CHECK]) |
4411f3b6 |
390 | |
0ab8f81e |
391 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef |
392 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e |
393 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
394 | For CHECK, see L</"Handling Malformed Data">. |
4411f3b6 |
395 | |
b7a5c9de |
396 | For example, to convert a string from Perl's internal format to |
6d1c0808 |
397 | iso-8859-1 (also known as Latin1), |
681a7c68 |
398 | |
b7a5c9de |
399 | $octets = encode("iso-8859-1", $string); |
7e19fb92 |
400 | |
b7a5c9de |
401 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets |
402 | B<may not be equal to> $string. Though they both contain the same data, the utf8 flag |
7e19fb92 |
403 | for $octets is B<always> off. When you encode anything, utf8 flag of |
404 | the result is always off, even when it contains completely valid utf8 |
405 | string. See L</"The UTF-8 flag"> below. |
681a7c68 |
406 | |
7f0d54d7 |
407 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
408 | |
b7a5c9de |
409 | =item $string = decode(ENCODING, $octets [, CHECK]) |
4411f3b6 |
410 | |
0ab8f81e |
411 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's |
412 | internal form and returns the resulting string. As in encode(), |
413 | ENCODING can be either a canonical name or an alias. For encoding names |
414 | and aliases, see L</"Defining Aliases">. For CHECK, see |
47bfe92f |
415 | L</"Handling Malformed Data">. |
416 | |
b7a5c9de |
417 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: |
681a7c68 |
418 | |
b7a5c9de |
419 | $string = decode("iso-8859-1", $octets); |
681a7c68 |
420 | |
b7a5c9de |
421 | B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string |
422 | B<may not be equal to> $octets. Though they both contain the same data, |
423 | the utf8 flag for $string is on unless $octets entirely consists of |
7e19fb92 |
424 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag"> |
425 | below. |
47bfe92f |
426 | |
7f0d54d7 |
427 | If the $string is C<undef> then C<undef> is returned. |
4089adc4 |
428 | |
b7a5c9de |
429 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) |
7e19fb92 |
430 | |
b7a5c9de |
431 | Converts B<in-place> data between two encodings. The data in $octets |
432 | must be encoded as octets and not as characters in Perl's internal |
f9d05ba3 |
433 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 |
434 | encoding: |
2b106fbe |
435 | |
b7a5c9de |
436 | from_to($octets, "iso-8859-1", "cp1250"); |
2b106fbe |
437 | |
438 | and to convert it back: |
439 | |
b7a5c9de |
440 | from_to($octets, "cp1250", "iso-8859-1"); |
4411f3b6 |
441 | |
ab97ca19 |
442 | Note that because the conversion happens in place, the data to be |
0ab8f81e |
443 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 |
444 | |
f9d05ba3 |
445 | from_to() returns the length of the converted string in octets on |
446 | success, I<undef> on error. |
3ef515df |
447 | |
b7a5c9de |
448 | B<CAVEAT>: The following operations look the same but are not quite so; |
7e19fb92 |
449 | |
b7a5c9de |
450 | from_to($data, "iso-8859-1", "utf8"); #1 |
7e19fb92 |
451 | $data = decode("iso-8859-1", $data); #2 |
4411f3b6 |
452 | |
b7a5c9de |
453 | Both #1 and #2 make $data consist of a completely valid UTF-8 string |
7e19fb92 |
454 | but only #2 turns utf8 flag on. #1 is equivalent to |
f2a2953c |
455 | |
7e19fb92 |
456 | $data = encode("utf8", decode("iso-8859-1", $data)); |
f2a2953c |
457 | |
7e19fb92 |
458 | See L</"The UTF-8 flag"> below. |
f2a2953c |
459 | |
460 | =item $octets = encode_utf8($string); |
461 | |
7e19fb92 |
462 | Equivalent to C<$octets = encode("utf8", $string);> The characters |
b7a5c9de |
463 | that comprise $string are encoded in Perl's internal format and the |
464 | result is returned as a sequence of octets. All possible |
7e19fb92 |
465 | characters have a UTF-8 representation so this function cannot fail. |
466 | |
f2a2953c |
467 | |
468 | =item $string = decode_utf8($octets [, CHECK]); |
469 | |
7e19fb92 |
470 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. |
b7a5c9de |
471 | The sequence of octets represented by |
7e19fb92 |
472 | $octets is decoded from UTF-8 into a sequence of logical |
473 | characters. Not all sequences of octets form valid UTF-8 encodings, so |
474 | it is possible for this call to fail. For CHECK, see |
475 | L</"Handling Malformed Data">. |
f2a2953c |
476 | |
477 | =back |
478 | |
51ef4e11 |
479 | =head2 Listing available encodings |
480 | |
5129552c |
481 | use Encode; |
482 | @list = Encode->encodings(); |
483 | |
484 | Returns a list of the canonical names of the available encodings that |
485 | are loaded. To get a list of all available encodings including the |
486 | ones that are not loaded yet, say |
487 | |
488 | @all_encodings = Encode->encodings(":all"); |
489 | |
0ab8f81e |
490 | Or you can give the name of a specific module. |
5129552c |
491 | |
c731e18e |
492 | @with_jp = Encode->encodings("Encode::JP"); |
493 | |
494 | When "::" is not in the name, "Encode::" is assumed. |
51ef4e11 |
495 | |
c731e18e |
496 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 |
497 | |
0ab8f81e |
498 | To find out in detail which encodings are supported by this package, |
5d030b67 |
499 | see L<Encode::Supported>. |
51ef4e11 |
500 | |
501 | =head2 Defining Aliases |
502 | |
0ab8f81e |
503 | To add a new alias to a given encoding, use: |
67d7b5ef |
504 | |
5129552c |
505 | use Encode; |
506 | use Encode::Alias; |
a63c962f |
507 | define_alias(newName => ENCODING); |
51ef4e11 |
508 | |
3ef515df |
509 | After that, newName can be used as an alias for ENCODING. |
f2a2953c |
510 | ENCODING may be either the name of an encoding or an |
511 | I<encoding object> |
51ef4e11 |
512 | |
fcb875d4 |
513 | But before you do so, make sure the alias is nonexistent with |
514 | C<resolve_alias()>, which returns the canonical name thereof. |
515 | i.e. |
516 | |
517 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true |
518 | Encode::resolve_alias("iso-8859-12") # false; nonexistent |
519 | Encode::resolve_alias($name) eq $name # true if $name is canonical |
520 | |
0ab8f81e |
521 | resolve_alias() does not need C<use Encode::Alias>; it can be |
522 | exported via C<use Encode qw(resolve_alias)>. |
fcb875d4 |
523 | |
0ab8f81e |
524 | See L<Encode::Alias> for details. |
51ef4e11 |
525 | |
85982a32 |
526 | =head1 Encoding via PerlIO |
4411f3b6 |
527 | |
b7a5c9de |
528 | If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode |
0ab8f81e |
529 | and encode directly via a filehandle. The following two examples |
530 | are totally identical in their functionality. |
4411f3b6 |
531 | |
85982a32 |
532 | # via PerlIO |
533 | open my $in, "<:encoding(shiftjis)", $infile or die; |
534 | open my $out, ">:encoding(euc-jp)", $outfile or die; |
b7a5c9de |
535 | while(<$in>){ print $out $_; } |
8e86646e |
536 | |
85982a32 |
537 | # via from_to |
0ab8f81e |
538 | open my $in, "<", $infile or die; |
539 | open my $out, ">", $outfile or die; |
b7a5c9de |
540 | while(<$in>){ |
0ab8f81e |
541 | from_to($_, "shiftjis", "euc-jp", 1); |
b7a5c9de |
542 | print $out $_; |
85982a32 |
543 | } |
4411f3b6 |
544 | |
b7a5c9de |
545 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check |
0ab8f81e |
546 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
547 | method. |
548 | |
549 | Encode::perlio_ok("hz"); # False |
550 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available |
551 | |
552 | use Encode qw(perlio_ok); # exported upon request |
553 | perlio_ok("euc-jp") |
4411f3b6 |
554 | |
0ab8f81e |
555 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
f9d05ba3 |
556 | except for hz and ISO-2022-kr. For gory details, see |
557 | L<Encode::Encoding> and L<Encode::PerlIO>. |
4411f3b6 |
558 | |
85982a32 |
559 | =head1 Handling Malformed Data |
4411f3b6 |
560 | |
f9d05ba3 |
561 | The optional I<CHECK> argument is used as follows. When you omit it, |
562 | Encode::FB_DEFAULT ( == 0 ) is assumed. |
563 | |
564 | =over 2 |
565 | |
566 | =item B<NOTE:> Not all encoding suppport this feature |
567 | |
568 | Some encodings ignore I<CHECK> argument. For example, |
569 | L<Encode::Unicode> ignores I<CHECK> and it always croaks on error. |
570 | |
571 | =back |
572 | |
573 | Now here is the list of I<CHECK> values available |
47bfe92f |
574 | |
151b5d36 |
575 | =over 2 |
576 | |
85982a32 |
577 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f |
578 | |
f9d05ba3 |
579 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> in |
78589665 |
580 | place of a malformed character. When you encode, E<lt>subcharE<gt> |
581 | will be used. When you decode the code point C<0xFFFD> is used. If |
582 | the data is supposed to be UTF-8, an optional lexical warning |
583 | (category utf8) is given. |
e9692b5b |
584 | |
7e19fb92 |
585 | =item I<CHECK> = Encode::FB_CROAK ( == 1) |
e9692b5b |
586 | |
b7a5c9de |
587 | If I<CHECK> is 1, methods will die on error immediately with an error |
0ab8f81e |
588 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
f9d05ba3 |
589 | error with eval{} unless you really want to let it die. |
47bfe92f |
590 | |
85982a32 |
591 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f |
592 | |
85982a32 |
593 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
f9d05ba3 |
594 | return the portion of the data that has been processed so far when an |
595 | error occurs. The data argument will be overwritten with everything |
596 | after that point (that is, the unprocessed part of data). This is |
597 | handy when you have to call decode repeatedly in the case where your |
598 | source data may contain partial multi-byte character sequences, |
599 | (i.e. you are reading with a fixed-width buffer). Here is a sample |
600 | code that does exactly this: |
4411f3b6 |
601 | |
78589665 |
602 | my $buffer = ''; my $string = ''; |
603 | while(read $fh, $buffer, 256, length($buffer)){ |
604 | $string .= decode($encoding, $buffer, Encode::FB_QUIET); |
605 | # $buffer now contains the unprocessed partial character |
85982a32 |
606 | } |
1768d7eb |
607 | |
85982a32 |
608 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef |
609 | |
0ab8f81e |
610 | This is the same as above, except that it warns on error. Handy when |
611 | you are debugging the mode above. |
85982a32 |
612 | |
613 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) |
614 | |
af1f55d9 |
615 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) |
616 | |
617 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) |
618 | |
85982a32 |
619 | For encodings that are implemented by Encode::XS, CHECK == |
620 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. |
621 | |
b7a5c9de |
622 | When you decode, C<\xI<HH>> will be inserted for a malformed character, |
623 | where I<HH> is the hex representation of the octet that could not be |
624 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, |
625 | where I<HHHH> is the Unicode ID of the character that cannot be found |
0ab8f81e |
626 | in the character repertoire of the encoding. |
85982a32 |
627 | |
af1f55d9 |
628 | HTML/XML character reference modes are about the same, in place of |
78589665 |
629 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNN>;> where I<NNN> is a decimal number and |
630 | XML uses C<&#xI<HHHH>;> where I<HHHH> is the hexadecimal number. |
af1f55d9 |
631 | |
7f0d54d7 |
632 | In Encode 2.10 or later, C<LEAVE_SRC> is also implied. |
633 | |
85982a32 |
634 | =item The bitmask |
635 | |
0ab8f81e |
636 | These modes are actually set via a bitmask. Here is how the FB_XX |
637 | constants are laid out. You can import the FB_XX constants via |
638 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask |
639 | constants via C<use Encode qw(:fallback_all)>. |
85982a32 |
640 | |
b0b300a3 |
641 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
642 | DIE_ON_ERR 0x0001 X |
4089adc4 |
643 | WARN_ON_ERR 0x0002 X |
b0b300a3 |
644 | RETURN_ON_ERR 0x0004 X X |
7f0d54d7 |
645 | LEAVE_SRC 0x0008 X |
b0b300a3 |
646 | PERLQQ 0x0100 X |
b7a5c9de |
647 | HTMLCREF 0x0200 |
648 | XMLCREF 0x0400 |
67d7b5ef |
649 | |
151b5d36 |
650 | =back |
651 | |
0ab8f81e |
652 | =head2 Unimplemented fallback schemes |
67d7b5ef |
653 | |
0ab8f81e |
654 | In the future, you will be able to use a code reference to a callback |
f2a2953c |
655 | function for the value of I<CHECK> but its API is still undecided. |
67d7b5ef |
656 | |
982a4085 |
657 | The fallback scheme does not work on EBCDIC platforms. |
658 | |
67d7b5ef |
659 | =head1 Defining Encodings |
660 | |
661 | To define a new encoding, use: |
662 | |
b7a5c9de |
663 | use Encode qw(define_encoding); |
67d7b5ef |
664 | define_encoding($object, 'canonicalName' [, alias...]); |
665 | |
666 | I<canonicalName> will be associated with I<$object>. The object |
0ab8f81e |
667 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef |
668 | If more than two arguments are provided then additional |
b7a5c9de |
669 | arguments are taken as aliases for I<$object>. |
67d7b5ef |
670 | |
f2a2953c |
671 | See L<Encode::Encoding> for more details. |
672 | |
7e19fb92 |
673 | =head1 The UTF-8 flag |
674 | |
675 | Before the introduction of utf8 support in perl, The C<eq> operator |
b7a5c9de |
676 | just compared the strings represented by two scalars. Beginning with |
677 | perl 5.8, C<eq> compares two strings with simultaneous consideration |
678 | of I<the utf8 flag>. To explain why we made it so, I will quote page |
679 | 402 of C<Programming Perl, 3rd ed.> |
7e19fb92 |
680 | |
681 | =over 2 |
682 | |
683 | =item Goal #1: |
684 | |
685 | Old byte-oriented programs should not spontaneously break on the old |
686 | byte-oriented data they used to work on. |
687 | |
688 | =item Goal #2: |
689 | |
690 | Old byte-oriented programs should magically start working on the new |
691 | character-oriented data when appropriate. |
692 | |
693 | =item Goal #3: |
694 | |
695 | Programs should run just as fast in the new character-oriented mode |
696 | as in the old byte-oriented mode. |
697 | |
698 | =item Goal #4: |
699 | |
700 | Perl should remain one language, rather than forking into a |
701 | byte-oriented Perl and a character-oriented Perl. |
702 | |
703 | =back |
704 | |
705 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 |
706 | was born and many features documented in the book remained |
b7a5c9de |
707 | unimplemented for a long time. Perl 5.8 corrected this and the introduction |
708 | of the UTF-8 flag is one of them. You can think of this perl notion as of a |
709 | byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8 |
7e19fb92 |
710 | flag on). |
711 | |
712 | Here is how Encode takes care of the utf8 flag. |
713 | |
4bdf5738 |
714 | =over 2 |
7e19fb92 |
715 | |
716 | =item * |
717 | |
718 | When you encode, the resulting utf8 flag is always off. |
719 | |
151b5d36 |
720 | =item * |
7e19fb92 |
721 | |
b7a5c9de |
722 | When you decode, the resulting utf8 flag is on unless you can |
7e19fb92 |
723 | unambiguously represent data. Here is the definition of |
724 | dis-ambiguity. |
725 | |
b7a5c9de |
726 | After C<$utf8 = decode('foo', $octet);>, |
7e19fb92 |
727 | |
728 | When $octet is... The utf8 flag in $utf8 is |
729 | --------------------------------------------- |
730 | In ASCII only (or EBCDIC only) OFF |
731 | In ISO-8859-1 ON |
732 | In any other Encoding ON |
733 | --------------------------------------------- |
734 | |
735 | As you see, there is one exception, In ASCII. That way you can assue |
736 | Goal #1. And with Encode Goal #2 is assumed but you still have to be |
737 | careful in such cases mentioned in B<CAVEAT> paragraphs. |
738 | |
739 | This utf8 flag is not visible in perl scripts, exactly for the same |
740 | reason you cannot (or you I<don't have to>) see if a scalar contains a |
741 | string, integer, or floating point number. But you can still peek |
742 | and poke these if you will. See the section below. |
743 | |
744 | =back |
745 | |
746 | =head2 Messing with Perl's Internals |
4411f3b6 |
747 | |
47bfe92f |
748 | The following API uses parts of Perl's internals in the current |
0ab8f81e |
749 | implementation. As such, they are efficient but may change. |
4411f3b6 |
750 | |
7e19fb92 |
751 | =over 2 |
4411f3b6 |
752 | |
a63c962f |
753 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 |
754 | |
0ab8f81e |
755 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. |
47bfe92f |
756 | If CHECK is true, also checks the data in STRING for being well-formed |
757 | UTF-8. Returns true if successful, false otherwise. |
4411f3b6 |
758 | |
2c246b25 |
759 | As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). |
b5ab1f6f |
760 | |
a63c962f |
761 | =item _utf8_on(STRING) |
4411f3b6 |
762 | |
0ab8f81e |
763 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is |
4411f3b6 |
764 | B<not> checked for being well-formed UTF-8. Do not use unless you |
765 | B<know> that the STRING is well-formed UTF-8. Returns the previous |
0ab8f81e |
766 | state of the UTF-8 flag (so please don't treat the return value as |
767 | indicating success or failure), or C<undef> if STRING is not a string. |
4411f3b6 |
768 | |
a63c962f |
769 | =item _utf8_off(STRING) |
4411f3b6 |
770 | |
0ab8f81e |
771 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. |
772 | Returns the previous state of the UTF-8 flag (so please don't treat the |
773 | return value as indicating success or failure), or C<undef> if STRING is |
4411f3b6 |
774 | not a string. |
775 | |
776 | =back |
777 | |
7f0d54d7 |
778 | =head1 UTF-8 vs. utf8 |
779 | |
780 | ....We now view strings not as sequences of bytes, but as sequences |
781 | of numbers in the range 0 .. 2**32-1 (or in the case of 64-bit |
782 | computers, 0 .. 2**64-1) -- Programming Perl, 3rd ed. |
783 | |
784 | That has been the perl's notion of UTF-8 but official UTF-8 is more |
785 | strict; Its ranges is much narrower (0 .. 10FFFF), some sequences are |
786 | not allowed (i.e. Those used in the surrogate pair, 0xFFFE, et al). |
787 | |
788 | Now that is overruled by Larry Wall himself. |
789 | |
790 | From: Larry Wall <larry@wall.org> |
791 | Date: December 04, 2004 11:51:58 JST |
792 | To: perl-unicode@perl.org |
793 | Subject: Re: Make Encode.pm support the real UTF-8 |
794 | Message-Id: <20041204025158.GA28754@wall.org> |
795 | |
796 | On Fri, Dec 03, 2004 at 10:12:12PM +0000, Tim Bunce wrote: |
797 | : I've no problem with 'utf8' being perl's unrestricted uft8 encoding, |
798 | : but "UTF-8" is the name of the standard and should give the |
799 | : corresponding behaviour. |
800 | |
801 | For what it's worth, that's how I've always kept them straight in my |
802 | head. |
803 | |
804 | Also for what it's worth, Perl 6 will mostly default to strict but |
805 | make it easy to switch back to lax. |
806 | |
807 | Larry |
808 | |
809 | Do you copy? As of Perl 5.8.7, B<UTF-8> means strict, official UTF-8 |
810 | while B<utf8> means liberal, lax, version thereof. And Encode version |
811 | 2.10 or later thus groks the difference between C<UTF-8> and C"utf8". |
812 | |
813 | encode("utf8", "\x{FFFF_FFFF}", 1); # okay |
814 | encode("UTF-8", "\x{FFFF_FFFF}", 1); # croaks |
815 | |
816 | C<UTF-8> in Encode is actually a canonical name for C<utf-8-strict>. |
817 | Yes, the hyphen between "UTF" and "8" is important. Without it Encode |
818 | goes "liberal" |
819 | |
820 | find_encoding("UTF-8")->name # is 'utf-8-strict' |
821 | find_encoding("utf-8")->name # ditto. names are case insensitive |
822 | find_encoding("utf8")->name # ditto. "_" are treated as "-" |
823 | find_encoding("UTF8")->name # is 'utf8'. |
824 | |
825 | |
4411f3b6 |
826 | =head1 SEE ALSO |
827 | |
5d030b67 |
828 | L<Encode::Encoding>, |
829 | L<Encode::Supported>, |
6d1c0808 |
830 | L<Encode::PerlIO>, |
5d030b67 |
831 | L<encoding>, |
6d1c0808 |
832 | L<perlebcdic>, |
833 | L<perlfunc/open>, |
834 | L<perlunicode>, |
835 | L<utf8>, |
5d030b67 |
836 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 |
837 | |
85982a32 |
838 | =head1 MAINTAINER |
aae85ceb |
839 | |
840 | This project was originated by Nick Ing-Simmons and later maintained |
7e19fb92 |
841 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full |
842 | list of people involved. For any questions, use |
b7a5c9de |
843 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. |
aae85ceb |
844 | |
4411f3b6 |
845 | =cut |