Commit | Line | Data |
---|---|---|
10c5ecbb | 1 | # |
c2cbba7d | 2 | # $Id: Encode.pm,v 1.99 2003/12/29 02:47:16 dankogai Exp dankogai $ |
10c5ecbb | 3 | # |
2c674647 | 4 | package Encode; |
51ef4e11 | 5 | use strict; |
c2cbba7d | 6 | our $VERSION = do { my @r = (q$Revision: 1.99 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r }; |
8f139f4c | 7 | sub DEBUG () { 0 } |
6d1c0808 | 8 | use XSLoader (); |
10c5ecbb | 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 JH |
15 | |
16 | our @EXPORT = qw( | |
17 | decode decode_utf8 encode encode_utf8 | |
a0d8a30e | 18 | encodings find_encoding clone_encoding |
4411f3b6 NIS |
19 | ); |
20 | ||
b7a5c9de | 21 | our @FB_FLAGS = qw(DIE_ON_ERR WARN_ON_ERR RETURN_ON_ERR LEAVE_SRC |
af1f55d9 | 22 | PERLQQ HTMLCREF XMLCREF); |
b7a5c9de | 23 | our @FB_CONSTS = qw(FB_DEFAULT FB_CROAK FB_QUIET FB_WARN |
af1f55d9 | 24 | FB_PERLQQ FB_HTMLCREF FB_XMLCREF); |
85982a32 | 25 | |
51ef4e11 | 26 | our @EXPORT_OK = |
6d1c0808 | 27 | ( |
85982a32 JH |
28 | qw( |
29 | _utf8_off _utf8_on define_encoding from_to is_16bit is_8bit | |
30 | is_utf8 perlio_ok resolve_alias utf8_downgrade utf8_upgrade | |
31 | ), | |
32 | @FB_FLAGS, @FB_CONSTS, | |
33 | ); | |
34 | ||
6d1c0808 | 35 | our %EXPORT_TAGS = |
85982a32 JH |
36 | ( |
37 | all => [ @EXPORT, @EXPORT_OK ], | |
38 | fallbacks => [ @FB_CONSTS ], | |
39 | fallback_all => [ @FB_CONSTS, @FB_FLAGS ], | |
40 | ); | |
41 | ||
4411f3b6 | 42 | # Documentation moved after __END__ for speed - NI-S |
2c674647 | 43 | |
a63c962f | 44 | our $ON_EBCDIC = (ord("A") == 193); |
f2a2953c | 45 | |
5d030b67 JH |
46 | use Encode::Alias; |
47 | ||
5129552c JH |
48 | # Make a %Encoding package variable to allow a certain amount of cheating |
49 | our %Encoding; | |
aae85ceb DK |
50 | our %ExtModule; |
51 | require Encode::Config; | |
52 | eval { require Encode::ConfigLocal }; | |
5129552c | 53 | |
656753f8 NIS |
54 | sub encodings |
55 | { | |
5129552c | 56 | my $class = shift; |
fc17bd48 JH |
57 | my %enc; |
58 | if (@_ and $_[0] eq ":all"){ | |
59 | %enc = ( %Encoding, %ExtModule ); | |
60 | }else{ | |
61 | %enc = %Encoding; | |
62 | for my $mod (map {m/::/o ? $_ : "Encode::$_" } @_){ | |
8f139f4c | 63 | DEBUG and warn $mod; |
fc17bd48 JH |
64 | for my $enc (keys %ExtModule){ |
65 | $ExtModule{$enc} eq $mod and $enc{$enc} = $mod; | |
66 | } | |
67 | } | |
5129552c JH |
68 | } |
69 | return | |
ce912cd4 | 70 | sort { lc $a cmp lc $b } |
fc17bd48 | 71 | grep {!/^(?:Internal|Unicode|Guess)$/o} keys %enc; |
51ef4e11 NIS |
72 | } |
73 | ||
85982a32 | 74 | sub perlio_ok{ |
0ab8f81e | 75 | my $obj = ref($_[0]) ? $_[0] : find_encoding($_[0]); |
011b2d2f | 76 | $obj->can("perlio_ok") and return $obj->perlio_ok(); |
0ab8f81e | 77 | return 0; # safety net |
85982a32 JH |
78 | } |
79 | ||
51ef4e11 NIS |
80 | sub define_encoding |
81 | { | |
18586f54 NIS |
82 | my $obj = shift; |
83 | my $name = shift; | |
5129552c | 84 | $Encoding{$name} = $obj; |
18586f54 NIS |
85 | my $lc = lc($name); |
86 | define_alias($lc => $obj) unless $lc eq $name; | |
10c5ecbb | 87 | while (@_){ |
18586f54 | 88 | my $alias = shift; |
10c5ecbb | 89 | define_alias($alias, $obj); |
18586f54 NIS |
90 | } |
91 | return $obj; | |
656753f8 NIS |
92 | } |
93 | ||
656753f8 NIS |
94 | sub getEncoding |
95 | { | |
10c5ecbb JH |
96 | my ($class, $name, $skip_external) = @_; |
97 | ||
a0d8a30e | 98 | ref($name) && $name->can('renew') and return $name; |
10c5ecbb | 99 | exists $Encoding{$name} and return $Encoding{$name}; |
18586f54 | 100 | my $lc = lc $name; |
10c5ecbb | 101 | exists $Encoding{$lc} and return $Encoding{$lc}; |
c50d192e | 102 | |
5129552c | 103 | my $oc = $class->find_alias($name); |
10c5ecbb JH |
104 | defined($oc) and return $oc; |
105 | $lc ne $name and $oc = $class->find_alias($lc); | |
106 | defined($oc) and return $oc; | |
c50d192e | 107 | |
c731e18e | 108 | unless ($skip_external) |
d1ed7747 | 109 | { |
c731e18e JH |
110 | if (my $mod = $ExtModule{$name} || $ExtModule{$lc}){ |
111 | $mod =~ s,::,/,g ; $mod .= '.pm'; | |
112 | eval{ require $mod; }; | |
10c5ecbb | 113 | exists $Encoding{$name} and return $Encoding{$name}; |
c731e18e | 114 | } |
d1ed7747 | 115 | } |
18586f54 | 116 | return; |
656753f8 NIS |
117 | } |
118 | ||
a0d8a30e | 119 | sub find_encoding($;$) |
4411f3b6 | 120 | { |
10c5ecbb | 121 | my ($name, $skip_external) = @_; |
dd9703c9 | 122 | return __PACKAGE__->getEncoding($name,$skip_external); |
4411f3b6 NIS |
123 | } |
124 | ||
a0d8a30e | 125 | sub resolve_alias($){ |
fcb875d4 JH |
126 | my $obj = find_encoding(shift); |
127 | defined $obj and return $obj->name; | |
128 | return; | |
129 | } | |
130 | ||
a0d8a30e DK |
131 | sub clone_encoding($){ |
132 | my $obj = find_encoding(shift); | |
133 | ref $obj or return; | |
134 | eval { require Storable }; | |
135 | $@ and return; | |
136 | return Storable::dclone($obj); | |
137 | } | |
138 | ||
b2704119 | 139 | sub encode($$;$) |
4411f3b6 | 140 | { |
e8c86ba6 | 141 | my ($name, $string, $check) = @_; |
0f7c507f | 142 | return undef unless defined $string; |
b2704119 | 143 | $check ||=0; |
18586f54 | 144 | my $enc = find_encoding($name); |
10c5ecbb JH |
145 | unless(defined $enc){ |
146 | require Carp; | |
147 | Carp::croak("Unknown encoding '$name'"); | |
148 | } | |
18586f54 | 149 | my $octets = $enc->encode($string,$check); |
23f3589e | 150 | $_[1] = $string if $check; |
18586f54 | 151 | return $octets; |
4411f3b6 NIS |
152 | } |
153 | ||
b2704119 | 154 | sub decode($$;$) |
4411f3b6 | 155 | { |
18586f54 | 156 | my ($name,$octets,$check) = @_; |
0f7c507f | 157 | return undef unless defined $octets; |
b2704119 | 158 | $check ||=0; |
18586f54 | 159 | my $enc = find_encoding($name); |
10c5ecbb JH |
160 | unless(defined $enc){ |
161 | require Carp; | |
162 | Carp::croak("Unknown encoding '$name'"); | |
163 | } | |
18586f54 NIS |
164 | my $string = $enc->decode($octets,$check); |
165 | $_[1] = $octets if $check; | |
166 | return $string; | |
4411f3b6 NIS |
167 | } |
168 | ||
b2704119 | 169 | sub from_to($$$;$) |
4411f3b6 | 170 | { |
18586f54 | 171 | my ($string,$from,$to,$check) = @_; |
0f7c507f | 172 | return undef unless defined $string; |
b2704119 | 173 | $check ||=0; |
18586f54 | 174 | my $f = find_encoding($from); |
10c5ecbb JH |
175 | unless (defined $f){ |
176 | require Carp; | |
177 | Carp::croak("Unknown encoding '$from'"); | |
178 | } | |
18586f54 | 179 | my $t = find_encoding($to); |
10c5ecbb JH |
180 | unless (defined $t){ |
181 | require Carp; | |
182 | Carp::croak("Unknown encoding '$to'"); | |
183 | } | |
18586f54 NIS |
184 | my $uni = $f->decode($string,$check); |
185 | return undef if ($check && length($string)); | |
a999c27c | 186 | $string = $t->encode($uni,$check); |
18586f54 | 187 | return undef if ($check && length($uni)); |
3ef515df | 188 | return defined($_[0] = $string) ? length($string) : undef ; |
4411f3b6 NIS |
189 | } |
190 | ||
b2704119 | 191 | sub encode_utf8($) |
4411f3b6 | 192 | { |
18586f54 | 193 | my ($str) = @_; |
c731e18e | 194 | utf8::encode($str); |
18586f54 | 195 | return $str; |
4411f3b6 NIS |
196 | } |
197 | ||
c2cbba7d | 198 | sub decode_utf8($;$) |
4411f3b6 | 199 | { |
c2cbba7d RGS |
200 | my ($str, $check) = @_; |
201 | if ($check){ | |
202 | return decode("utf8", $str, $check); | |
203 | }else{ | |
204 | return undef unless utf8::decode($str); | |
205 | return $str; | |
206 | } | |
5ad8ef52 NIS |
207 | } |
208 | ||
b536bf57 | 209 | predefine_encodings(1); |
f2a2953c JH |
210 | |
211 | # | |
212 | # This is to restore %Encoding if really needed; | |
213 | # | |
10c5ecbb | 214 | |
f2a2953c | 215 | sub predefine_encodings{ |
10c5ecbb | 216 | use Encode::Encoding; |
b536bf57 DK |
217 | no warnings 'redefine'; |
218 | my $use_xs = shift; | |
6d1c0808 | 219 | if ($ON_EBCDIC) { |
f2a2953c JH |
220 | # was in Encode::UTF_EBCDIC |
221 | package Encode::UTF_EBCDIC; | |
10c5ecbb | 222 | push @Encode::UTF_EBCDIC::ISA, 'Encode::Encoding'; |
f2a2953c JH |
223 | *decode = sub{ |
224 | my ($obj,$str,$chk) = @_; | |
225 | my $res = ''; | |
226 | for (my $i = 0; $i < length($str); $i++) { | |
6d1c0808 | 227 | $res .= |
f2a2953c JH |
228 | chr(utf8::unicode_to_native(ord(substr($str,$i,1)))); |
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++) { | |
6d1c0808 | 237 | $res .= |
f2a2953c JH |
238 | chr(utf8::native_to_unicode(ord(substr($str,$i,1)))); |
239 | } | |
240 | $_[1] = '' if $chk; | |
241 | return $res; | |
242 | }; | |
6d1c0808 | 243 | $Encode::Encoding{Unicode} = |
c731e18e | 244 | bless {Name => "UTF_EBCDIC"} => "Encode::UTF_EBCDIC"; |
6d1c0808 | 245 | } else { |
f2a2953c | 246 | package Encode::Internal; |
10c5ecbb | 247 | push @Encode::Internal::ISA, 'Encode::Encoding'; |
f2a2953c JH |
248 | *decode = sub{ |
249 | my ($obj,$str,$chk) = @_; | |
250 | utf8::upgrade($str); | |
251 | $_[1] = '' if $chk; | |
252 | return $str; | |
253 | }; | |
254 | *encode = \&decode; | |
6d1c0808 | 255 | $Encode::Encoding{Unicode} = |
c731e18e | 256 | bless {Name => "Internal"} => "Encode::Internal"; |
f2a2953c JH |
257 | } |
258 | ||
259 | { | |
260 | # was in Encode::utf8 | |
261 | package Encode::utf8; | |
10c5ecbb | 262 | push @Encode::utf8::ISA, 'Encode::Encoding'; |
b536bf57 DK |
263 | # |
264 | if ($use_xs){ | |
8f139f4c | 265 | Encode::DEBUG and warn __PACKAGE__, " XS on"; |
b536bf57 DK |
266 | *decode = \&decode_xs; |
267 | *encode = \&encode_xs; | |
268 | }else{ | |
8f139f4c | 269 | Encode::DEBUG and warn __PACKAGE__, " XS off"; |
b536bf57 DK |
270 | *decode = sub{ |
271 | my ($obj,$octets,$chk) = @_; | |
272 | my $str = Encode::decode_utf8($octets); | |
273 | if (defined $str) { | |
274 | $_[1] = '' if $chk; | |
275 | return $str; | |
276 | } | |
277 | return undef; | |
278 | }; | |
279 | *encode = sub { | |
280 | my ($obj,$string,$chk) = @_; | |
281 | my $octets = Encode::encode_utf8($string); | |
282 | $_[1] = '' if $chk; | |
283 | return $octets; | |
284 | }; | |
285 | } | |
220e2d4e IH |
286 | *cat_decode = sub{ # ($obj, $dst, $src, $pos, $trm, $chk) |
287 | my ($obj, undef, undef, $pos, $trm) = @_; # currently ignores $chk | |
288 | my ($rdst, $rsrc, $rpos) = \@_[1,2,3]; | |
289 | use bytes; | |
290 | if ((my $npos = index($$rsrc, $trm, $pos)) >= 0) { | |
291 | $$rdst .= substr($$rsrc, $pos, $npos - $pos + length($trm)); | |
292 | $$rpos = $npos + length($trm); | |
293 | return 1; | |
294 | } | |
295 | $$rdst .= substr($$rsrc, $pos); | |
296 | $$rpos = length($$rsrc); | |
297 | return ''; | |
298 | }; | |
b7a5c9de | 299 | $Encode::Encoding{utf8} = |
c731e18e | 300 | bless {Name => "utf8"} => "Encode::utf8"; |
f2a2953c | 301 | } |
f2a2953c JH |
302 | } |
303 | ||
656753f8 NIS |
304 | 1; |
305 | ||
2a936312 NIS |
306 | __END__ |
307 | ||
4411f3b6 NIS |
308 | =head1 NAME |
309 | ||
310 | Encode - character encodings | |
311 | ||
312 | =head1 SYNOPSIS | |
313 | ||
314 | use Encode; | |
315 | ||
67d7b5ef JH |
316 | =head2 Table of Contents |
317 | ||
0ab8f81e | 318 | Encode consists of a collection of modules whose details are too big |
67d7b5ef | 319 | to fit in one document. This POD itself explains the top-level APIs |
6d1c0808 | 320 | and general topics at a glance. For other topics and more details, |
0ab8f81e | 321 | see the PODs below: |
67d7b5ef JH |
322 | |
323 | Name Description | |
324 | -------------------------------------------------------- | |
6d1c0808 | 325 | Encode::Alias Alias definitions to encodings |
67d7b5ef JH |
326 | Encode::Encoding Encode Implementation Base Class |
327 | Encode::Supported List of Supported Encodings | |
328 | Encode::CN Simplified Chinese Encodings | |
329 | Encode::JP Japanese Encodings | |
330 | Encode::KR Korean Encodings | |
331 | Encode::TW Traditional Chinese Encodings | |
332 | -------------------------------------------------------- | |
333 | ||
4411f3b6 NIS |
334 | =head1 DESCRIPTION |
335 | ||
47bfe92f | 336 | The C<Encode> module provides the interfaces between Perl's strings |
67d7b5ef JH |
337 | and the rest of the system. Perl strings are sequences of |
338 | B<characters>. | |
339 | ||
340 | The repertoire of characters that Perl can represent is at least that | |
341 | defined by the Unicode Consortium. On most platforms the ordinal | |
342 | values of the characters (as returned by C<ord(ch)>) is the "Unicode | |
343 | codepoint" for the character (the exceptions are those platforms where | |
344 | the legacy encoding is some variant of EBCDIC rather than a super-set | |
345 | of ASCII - see L<perlebcdic>). | |
346 | ||
0ab8f81e | 347 | Traditionally, computer data has been moved around in 8-bit chunks |
67d7b5ef JH |
348 | often called "bytes". These chunks are also known as "octets" in |
349 | networking standards. Perl is widely used to manipulate data of many | |
350 | types - not only strings of characters representing human or computer | |
0ab8f81e | 351 | languages but also "binary" data being the machine's representation of |
67d7b5ef JH |
352 | numbers, pixels in an image - or just about anything. |
353 | ||
0ab8f81e | 354 | When Perl is processing "binary data", the programmer wants Perl to |
67d7b5ef | 355 | process "sequences of bytes". This is not a problem for Perl - as a |
0ab8f81e | 356 | byte has 256 possible values, it easily fits in Perl's much larger |
67d7b5ef JH |
357 | "logical character". |
358 | ||
359 | =head2 TERMINOLOGY | |
4411f3b6 | 360 | |
7e19fb92 | 361 | =over 2 |
21938dfa | 362 | |
67d7b5ef JH |
363 | =item * |
364 | ||
365 | I<character>: a character in the range 0..(2**32-1) (or more). | |
366 | (What Perl's strings are made of.) | |
367 | ||
368 | =item * | |
369 | ||
370 | I<byte>: a character in the range 0..255 | |
371 | (A special case of a Perl character.) | |
372 | ||
373 | =item * | |
374 | ||
375 | I<octet>: 8 bits of data, with ordinal values 0..255 | |
0ab8f81e | 376 | (Term for bytes passed to or from a non-Perl context, e.g. a disk file.) |
67d7b5ef JH |
377 | |
378 | =back | |
4411f3b6 | 379 | |
67d7b5ef | 380 | =head1 PERL ENCODING API |
4411f3b6 | 381 | |
7e19fb92 | 382 | =over 2 |
4411f3b6 | 383 | |
b7a5c9de | 384 | =item $octets = encode(ENCODING, $string [, CHECK]) |
4411f3b6 | 385 | |
0ab8f81e | 386 | Encodes a string from Perl's internal form into I<ENCODING> and returns |
67d7b5ef | 387 | a sequence of octets. ENCODING can be either a canonical name or |
0ab8f81e JH |
388 | an alias. For encoding names and aliases, see L</"Defining Aliases">. |
389 | For CHECK, see L</"Handling Malformed Data">. | |
4411f3b6 | 390 | |
b7a5c9de | 391 | For example, to convert a string from Perl's internal format to |
6d1c0808 | 392 | iso-8859-1 (also known as Latin1), |
681a7c68 | 393 | |
b7a5c9de | 394 | $octets = encode("iso-8859-1", $string); |
7e19fb92 | 395 | |
b7a5c9de JH |
396 | B<CAVEAT>: When you run C<$octets = encode("utf8", $string)>, then $octets |
397 | B<may not be equal to> $string. Though they both contain the same data, the utf8 flag | |
7e19fb92 JH |
398 | for $octets is B<always> off. When you encode anything, utf8 flag of |
399 | the result is always off, even when it contains completely valid utf8 | |
400 | string. See L</"The UTF-8 flag"> below. | |
681a7c68 | 401 | |
4089adc4 JH |
402 | encode($valid_encoding, undef) is harmless but warns you for |
403 | C<Use of uninitialized value in subroutine entry>. | |
404 | encode($valid_encoding, '') is harmless and warnless. | |
405 | ||
b7a5c9de | 406 | =item $string = decode(ENCODING, $octets [, CHECK]) |
4411f3b6 | 407 | |
0ab8f81e JH |
408 | Decodes a sequence of octets assumed to be in I<ENCODING> into Perl's |
409 | internal form and returns the resulting string. As in encode(), | |
410 | ENCODING can be either a canonical name or an alias. For encoding names | |
411 | and aliases, see L</"Defining Aliases">. For CHECK, see | |
47bfe92f JH |
412 | L</"Handling Malformed Data">. |
413 | ||
b7a5c9de | 414 | For example, to convert ISO-8859-1 data to a string in Perl's internal format: |
681a7c68 | 415 | |
b7a5c9de | 416 | $string = decode("iso-8859-1", $octets); |
681a7c68 | 417 | |
b7a5c9de JH |
418 | B<CAVEAT>: When you run C<$string = decode("utf8", $octets)>, then $string |
419 | B<may not be equal to> $octets. Though they both contain the same data, | |
420 | the utf8 flag for $string is on unless $octets entirely consists of | |
7e19fb92 JH |
421 | ASCII data (or EBCDIC on EBCDIC machines). See L</"The UTF-8 flag"> |
422 | below. | |
47bfe92f | 423 | |
4089adc4 JH |
424 | decode($valid_encoding, undef) is harmless but warns you for |
425 | C<Use of uninitialized value in subroutine entry>. | |
426 | decode($valid_encoding, '') is harmless and warnless. | |
427 | ||
b7a5c9de | 428 | =item [$length =] from_to($octets, FROM_ENC, TO_ENC [, CHECK]) |
7e19fb92 | 429 | |
b7a5c9de JH |
430 | Converts B<in-place> data between two encodings. The data in $octets |
431 | must be encoded as octets and not as characters in Perl's internal | |
432 | format. For example, to convert ISO-8859-1 data to Microsoft's CP1250 encoding: | |
2b106fbe | 433 | |
b7a5c9de | 434 | from_to($octets, "iso-8859-1", "cp1250"); |
2b106fbe JH |
435 | |
436 | and to convert it back: | |
437 | ||
b7a5c9de | 438 | from_to($octets, "cp1250", "iso-8859-1"); |
4411f3b6 | 439 | |
ab97ca19 | 440 | Note that because the conversion happens in place, the data to be |
0ab8f81e | 441 | converted cannot be a string constant; it must be a scalar variable. |
ab97ca19 | 442 | |
b7a5c9de | 443 | from_to() returns the length of the converted string in octets on success, undef |
3ef515df JH |
444 | otherwise. |
445 | ||
b7a5c9de | 446 | B<CAVEAT>: The following operations look the same but are not quite so; |
7e19fb92 | 447 | |
b7a5c9de | 448 | from_to($data, "iso-8859-1", "utf8"); #1 |
7e19fb92 | 449 | $data = decode("iso-8859-1", $data); #2 |
4411f3b6 | 450 | |
b7a5c9de | 451 | Both #1 and #2 make $data consist of a completely valid UTF-8 string |
7e19fb92 | 452 | but only #2 turns utf8 flag on. #1 is equivalent to |
f2a2953c | 453 | |
7e19fb92 | 454 | $data = encode("utf8", decode("iso-8859-1", $data)); |
f2a2953c | 455 | |
7e19fb92 | 456 | See L</"The UTF-8 flag"> below. |
f2a2953c JH |
457 | |
458 | =item $octets = encode_utf8($string); | |
459 | ||
7e19fb92 | 460 | Equivalent to C<$octets = encode("utf8", $string);> The characters |
b7a5c9de JH |
461 | that comprise $string are encoded in Perl's internal format and the |
462 | result is returned as a sequence of octets. All possible | |
7e19fb92 JH |
463 | characters have a UTF-8 representation so this function cannot fail. |
464 | ||
f2a2953c JH |
465 | |
466 | =item $string = decode_utf8($octets [, CHECK]); | |
467 | ||
7e19fb92 | 468 | equivalent to C<$string = decode("utf8", $octets [, CHECK])>. |
b7a5c9de | 469 | The sequence of octets represented by |
7e19fb92 JH |
470 | $octets is decoded from UTF-8 into a sequence of logical |
471 | characters. Not all sequences of octets form valid UTF-8 encodings, so | |
472 | it is possible for this call to fail. For CHECK, see | |
473 | L</"Handling Malformed Data">. | |
f2a2953c JH |
474 | |
475 | =back | |
476 | ||
51ef4e11 NIS |
477 | =head2 Listing available encodings |
478 | ||
5129552c JH |
479 | use Encode; |
480 | @list = Encode->encodings(); | |
481 | ||
482 | Returns a list of the canonical names of the available encodings that | |
483 | are loaded. To get a list of all available encodings including the | |
484 | ones that are not loaded yet, say | |
485 | ||
486 | @all_encodings = Encode->encodings(":all"); | |
487 | ||
0ab8f81e | 488 | Or you can give the name of a specific module. |
5129552c | 489 | |
c731e18e JH |
490 | @with_jp = Encode->encodings("Encode::JP"); |
491 | ||
492 | When "::" is not in the name, "Encode::" is assumed. | |
51ef4e11 | 493 | |
c731e18e | 494 | @ebcdic = Encode->encodings("EBCDIC"); |
5d030b67 | 495 | |
0ab8f81e | 496 | To find out in detail which encodings are supported by this package, |
5d030b67 | 497 | see L<Encode::Supported>. |
51ef4e11 NIS |
498 | |
499 | =head2 Defining Aliases | |
500 | ||
0ab8f81e | 501 | To add a new alias to a given encoding, use: |
67d7b5ef | 502 | |
5129552c JH |
503 | use Encode; |
504 | use Encode::Alias; | |
a63c962f | 505 | define_alias(newName => ENCODING); |
51ef4e11 | 506 | |
3ef515df | 507 | After that, newName can be used as an alias for ENCODING. |
f2a2953c JH |
508 | ENCODING may be either the name of an encoding or an |
509 | I<encoding object> | |
51ef4e11 | 510 | |
fcb875d4 JH |
511 | But before you do so, make sure the alias is nonexistent with |
512 | C<resolve_alias()>, which returns the canonical name thereof. | |
513 | i.e. | |
514 | ||
515 | Encode::resolve_alias("latin1") eq "iso-8859-1" # true | |
516 | Encode::resolve_alias("iso-8859-12") # false; nonexistent | |
517 | Encode::resolve_alias($name) eq $name # true if $name is canonical | |
518 | ||
0ab8f81e JH |
519 | resolve_alias() does not need C<use Encode::Alias>; it can be |
520 | exported via C<use Encode qw(resolve_alias)>. | |
fcb875d4 | 521 | |
0ab8f81e | 522 | See L<Encode::Alias> for details. |
51ef4e11 | 523 | |
85982a32 | 524 | =head1 Encoding via PerlIO |
4411f3b6 | 525 | |
b7a5c9de | 526 | If your perl supports I<PerlIO> (which is the default), you can use a PerlIO layer to decode |
0ab8f81e JH |
527 | and encode directly via a filehandle. The following two examples |
528 | are totally identical in their functionality. | |
4411f3b6 | 529 | |
85982a32 JH |
530 | # via PerlIO |
531 | open my $in, "<:encoding(shiftjis)", $infile or die; | |
532 | open my $out, ">:encoding(euc-jp)", $outfile or die; | |
b7a5c9de | 533 | while(<$in>){ print $out $_; } |
8e86646e | 534 | |
85982a32 | 535 | # via from_to |
0ab8f81e JH |
536 | open my $in, "<", $infile or die; |
537 | open my $out, ">", $outfile or die; | |
b7a5c9de | 538 | while(<$in>){ |
0ab8f81e | 539 | from_to($_, "shiftjis", "euc-jp", 1); |
b7a5c9de | 540 | print $out $_; |
85982a32 | 541 | } |
4411f3b6 | 542 | |
b7a5c9de | 543 | Unfortunately, it may be that encodings are PerlIO-savvy. You can check |
0ab8f81e JH |
544 | if your encoding is supported by PerlIO by calling the C<perlio_ok> |
545 | method. | |
546 | ||
547 | Encode::perlio_ok("hz"); # False | |
548 | find_encoding("euc-cn")->perlio_ok; # True where PerlIO is available | |
549 | ||
550 | use Encode qw(perlio_ok); # exported upon request | |
551 | perlio_ok("euc-jp") | |
4411f3b6 | 552 | |
0ab8f81e | 553 | Fortunately, all encodings that come with Encode core are PerlIO-savvy |
b7a5c9de | 554 | except for hz and ISO-2022-kr. For gory details, see L<Encode::Encoding> and L<Encode::PerlIO>. |
4411f3b6 | 555 | |
85982a32 | 556 | =head1 Handling Malformed Data |
4411f3b6 | 557 | |
0ab8f81e JH |
558 | The I<CHECK> argument is used as follows. When you omit it, |
559 | the behaviour is the same as if you had passed a value of 0 for | |
560 | I<CHECK>. | |
47bfe92f | 561 | |
151b5d36 JH |
562 | =over 2 |
563 | ||
85982a32 | 564 | =item I<CHECK> = Encode::FB_DEFAULT ( == 0) |
47bfe92f | 565 | |
0ab8f81e JH |
566 | If I<CHECK> is 0, (en|de)code will put a I<substitution character> |
567 | in place of a malformed character. For UCM-based encodings, | |
b7a5c9de | 568 | E<lt>subcharE<gt> will be used. For Unicode, the code point C<0xFFFD> is used. |
0ab8f81e JH |
569 | If the data is supposed to be UTF-8, an optional lexical warning |
570 | (category utf8) is given. | |
e9692b5b | 571 | |
7e19fb92 | 572 | =item I<CHECK> = Encode::FB_CROAK ( == 1) |
e9692b5b | 573 | |
b7a5c9de | 574 | If I<CHECK> is 1, methods will die on error immediately with an error |
0ab8f81e JH |
575 | message. Therefore, when I<CHECK> is set to 1, you should trap the |
576 | fatal error with eval{} unless you really want to let it die on error. | |
47bfe92f | 577 | |
85982a32 | 578 | =item I<CHECK> = Encode::FB_QUIET |
47bfe92f | 579 | |
85982a32 | 580 | If I<CHECK> is set to Encode::FB_QUIET, (en|de)code will immediately |
0ab8f81e JH |
581 | return the portion of the data that has been processed so far when |
582 | an error occurs. The data argument will be overwritten with | |
583 | everything after that point (that is, the unprocessed part of data). | |
584 | This is handy when you have to call decode repeatedly in the case | |
585 | where your source data may contain partial multi-byte character | |
586 | sequences, for example because you are reading with a fixed-width | |
587 | buffer. Here is some sample code that does exactly this: | |
4411f3b6 | 588 | |
b7a5c9de | 589 | my $data = ''; my $utf8 = ''; |
85982a32 | 590 | while(defined(read $fh, $buffer, 256)){ |
0ab8f81e | 591 | # buffer may end in a partial character so we append |
85982a32 | 592 | $data .= $buffer; |
ee269af2 | 593 | $utf8 .= decode($encoding, $data, Encode::FB_QUIET); |
0ab8f81e | 594 | # $data now contains the unprocessed partial character |
85982a32 | 595 | } |
1768d7eb | 596 | |
85982a32 | 597 | =item I<CHECK> = Encode::FB_WARN |
67d7b5ef | 598 | |
0ab8f81e JH |
599 | This is the same as above, except that it warns on error. Handy when |
600 | you are debugging the mode above. | |
85982a32 JH |
601 | |
602 | =item perlqq mode (I<CHECK> = Encode::FB_PERLQQ) | |
603 | ||
af1f55d9 JH |
604 | =item HTML charref mode (I<CHECK> = Encode::FB_HTMLCREF) |
605 | ||
606 | =item XML charref mode (I<CHECK> = Encode::FB_XMLCREF) | |
607 | ||
85982a32 JH |
608 | For encodings that are implemented by Encode::XS, CHECK == |
609 | Encode::FB_PERLQQ turns (en|de)code into C<perlqq> fallback mode. | |
610 | ||
b7a5c9de JH |
611 | When you decode, C<\xI<HH>> will be inserted for a malformed character, |
612 | where I<HH> is the hex representation of the octet that could not be | |
613 | decoded to utf8. And when you encode, C<\x{I<HHHH>}> will be inserted, | |
614 | where I<HHHH> is the Unicode ID of the character that cannot be found | |
0ab8f81e | 615 | in the character repertoire of the encoding. |
85982a32 | 616 | |
af1f55d9 | 617 | HTML/XML character reference modes are about the same, in place of |
b7a5c9de JH |
618 | C<\x{I<HHHH>}>, HTML uses C<&#I<NNNN>>; where I<NNNN> is a decimal digit and |
619 | XML uses C<&#xI<HHHH>>; where I<HHHH> is the hexadecimal digit. | |
af1f55d9 | 620 | |
85982a32 JH |
621 | =item The bitmask |
622 | ||
0ab8f81e JH |
623 | These modes are actually set via a bitmask. Here is how the FB_XX |
624 | constants are laid out. You can import the FB_XX constants via | |
625 | C<use Encode qw(:fallbacks)>; you can import the generic bitmask | |
626 | constants via C<use Encode qw(:fallback_all)>. | |
85982a32 | 627 | |
b0b300a3 JH |
628 | FB_DEFAULT FB_CROAK FB_QUIET FB_WARN FB_PERLQQ |
629 | DIE_ON_ERR 0x0001 X | |
4089adc4 | 630 | WARN_ON_ERR 0x0002 X |
b0b300a3 JH |
631 | RETURN_ON_ERR 0x0004 X X |
632 | LEAVE_SRC 0x0008 | |
633 | PERLQQ 0x0100 X | |
b7a5c9de JH |
634 | HTMLCREF 0x0200 |
635 | XMLCREF 0x0400 | |
67d7b5ef | 636 | |
151b5d36 JH |
637 | =back |
638 | ||
0ab8f81e | 639 | =head2 Unimplemented fallback schemes |
67d7b5ef | 640 | |
0ab8f81e | 641 | In the future, you will be able to use a code reference to a callback |
f2a2953c | 642 | function for the value of I<CHECK> but its API is still undecided. |
67d7b5ef | 643 | |
982a4085 JH |
644 | The fallback scheme does not work on EBCDIC platforms. |
645 | ||
67d7b5ef JH |
646 | =head1 Defining Encodings |
647 | ||
648 | To define a new encoding, use: | |
649 | ||
b7a5c9de | 650 | use Encode qw(define_encoding); |
67d7b5ef JH |
651 | define_encoding($object, 'canonicalName' [, alias...]); |
652 | ||
653 | I<canonicalName> will be associated with I<$object>. The object | |
0ab8f81e | 654 | should provide the interface described in L<Encode::Encoding>. |
67d7b5ef | 655 | If more than two arguments are provided then additional |
b7a5c9de | 656 | arguments are taken as aliases for I<$object>. |
67d7b5ef | 657 | |
f2a2953c JH |
658 | See L<Encode::Encoding> for more details. |
659 | ||
7e19fb92 JH |
660 | =head1 The UTF-8 flag |
661 | ||
662 | Before the introduction of utf8 support in perl, The C<eq> operator | |
b7a5c9de JH |
663 | just compared the strings represented by two scalars. Beginning with |
664 | perl 5.8, C<eq> compares two strings with simultaneous consideration | |
665 | of I<the utf8 flag>. To explain why we made it so, I will quote page | |
666 | 402 of C<Programming Perl, 3rd ed.> | |
7e19fb92 JH |
667 | |
668 | =over 2 | |
669 | ||
670 | =item Goal #1: | |
671 | ||
672 | Old byte-oriented programs should not spontaneously break on the old | |
673 | byte-oriented data they used to work on. | |
674 | ||
675 | =item Goal #2: | |
676 | ||
677 | Old byte-oriented programs should magically start working on the new | |
678 | character-oriented data when appropriate. | |
679 | ||
680 | =item Goal #3: | |
681 | ||
682 | Programs should run just as fast in the new character-oriented mode | |
683 | as in the old byte-oriented mode. | |
684 | ||
685 | =item Goal #4: | |
686 | ||
687 | Perl should remain one language, rather than forking into a | |
688 | byte-oriented Perl and a character-oriented Perl. | |
689 | ||
690 | =back | |
691 | ||
692 | Back when C<Programming Perl, 3rd ed.> was written, not even Perl 5.6.0 | |
693 | was born and many features documented in the book remained | |
b7a5c9de JH |
694 | unimplemented for a long time. Perl 5.8 corrected this and the introduction |
695 | of the UTF-8 flag is one of them. You can think of this perl notion as of a | |
696 | byte-oriented mode (utf8 flag off) and a character-oriented mode (utf8 | |
7e19fb92 JH |
697 | flag on). |
698 | ||
699 | Here is how Encode takes care of the utf8 flag. | |
700 | ||
4bdf5738 | 701 | =over 2 |
7e19fb92 JH |
702 | |
703 | =item * | |
704 | ||
705 | When you encode, the resulting utf8 flag is always off. | |
706 | ||
151b5d36 | 707 | =item * |
7e19fb92 | 708 | |
b7a5c9de | 709 | When you decode, the resulting utf8 flag is on unless you can |
7e19fb92 JH |
710 | unambiguously represent data. Here is the definition of |
711 | dis-ambiguity. | |
712 | ||
b7a5c9de | 713 | After C<$utf8 = decode('foo', $octet);>, |
7e19fb92 JH |
714 | |
715 | When $octet is... The utf8 flag in $utf8 is | |
716 | --------------------------------------------- | |
717 | In ASCII only (or EBCDIC only) OFF | |
718 | In ISO-8859-1 ON | |
719 | In any other Encoding ON | |
720 | --------------------------------------------- | |
721 | ||
722 | As you see, there is one exception, In ASCII. That way you can assue | |
723 | Goal #1. And with Encode Goal #2 is assumed but you still have to be | |
724 | careful in such cases mentioned in B<CAVEAT> paragraphs. | |
725 | ||
726 | This utf8 flag is not visible in perl scripts, exactly for the same | |
727 | reason you cannot (or you I<don't have to>) see if a scalar contains a | |
728 | string, integer, or floating point number. But you can still peek | |
729 | and poke these if you will. See the section below. | |
730 | ||
731 | =back | |
732 | ||
733 | =head2 Messing with Perl's Internals | |
4411f3b6 | 734 | |
47bfe92f | 735 | The following API uses parts of Perl's internals in the current |
0ab8f81e | 736 | implementation. As such, they are efficient but may change. |
4411f3b6 | 737 | |
7e19fb92 | 738 | =over 2 |
4411f3b6 | 739 | |
a63c962f | 740 | =item is_utf8(STRING [, CHECK]) |
4411f3b6 | 741 | |
0ab8f81e | 742 | [INTERNAL] Tests whether the UTF-8 flag is turned on in the STRING. |
47bfe92f JH |
743 | If CHECK is true, also checks the data in STRING for being well-formed |
744 | UTF-8. Returns true if successful, false otherwise. | |
4411f3b6 | 745 | |
2c246b25 | 746 | As of perl 5.8.1, L<utf8> also has utf8::is_utf8(). |
b5ab1f6f | 747 | |
a63c962f | 748 | =item _utf8_on(STRING) |
4411f3b6 | 749 | |
0ab8f81e | 750 | [INTERNAL] Turns on the UTF-8 flag in STRING. The data in STRING is |
4411f3b6 NIS |
751 | B<not> checked for being well-formed UTF-8. Do not use unless you |
752 | B<know> that the STRING is well-formed UTF-8. Returns the previous | |
0ab8f81e JH |
753 | state of the UTF-8 flag (so please don't treat the return value as |
754 | indicating success or failure), or C<undef> if STRING is not a string. | |
4411f3b6 | 755 | |
a63c962f | 756 | =item _utf8_off(STRING) |
4411f3b6 | 757 | |
0ab8f81e JH |
758 | [INTERNAL] Turns off the UTF-8 flag in STRING. Do not use frivolously. |
759 | Returns the previous state of the UTF-8 flag (so please don't treat the | |
760 | return value as indicating success or failure), or C<undef> if STRING is | |
4411f3b6 NIS |
761 | not a string. |
762 | ||
763 | =back | |
764 | ||
765 | =head1 SEE ALSO | |
766 | ||
5d030b67 JH |
767 | L<Encode::Encoding>, |
768 | L<Encode::Supported>, | |
6d1c0808 | 769 | L<Encode::PerlIO>, |
5d030b67 | 770 | L<encoding>, |
6d1c0808 JH |
771 | L<perlebcdic>, |
772 | L<perlfunc/open>, | |
773 | L<perlunicode>, | |
774 | L<utf8>, | |
5d030b67 | 775 | the Perl Unicode Mailing List E<lt>perl-unicode@perl.orgE<gt> |
4411f3b6 | 776 | |
85982a32 | 777 | =head1 MAINTAINER |
aae85ceb DK |
778 | |
779 | This project was originated by Nick Ing-Simmons and later maintained | |
7e19fb92 JH |
780 | by Dan Kogai E<lt>dankogai@dan.co.jpE<gt>. See AUTHORS for a full |
781 | list of people involved. For any questions, use | |
b7a5c9de | 782 | E<lt>perl-unicode@perl.orgE<gt> so we can all share. |
aae85ceb | 783 | |
4411f3b6 | 784 | =cut |