1 package Encode::Encoding;
2 # Base class for classes which implement encodings
4 our $VERSION = do { my @r = (q$Revision: 2.2 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
12 my $canonical = shift;
13 $obj = bless { Name => $canonical },$obj unless ref $obj;
14 # warn "$canonical => $obj\n";
15 Encode::define_encoding($obj, $canonical, @_);
18 sub name { return shift->{'Name'} }
20 # sub renew { return $_[0] }
24 my $clone = bless { %$self } => ref($self);
25 $clone->{renewed}++; # so the caller can see it
26 DEBUG and warn $clone->{renewed};
30 sub renewed{ return $_[0]->{renewed} || 0 }
32 *new_sequence = \&renew;
34 sub needs_lines { 0 };
37 eval{ require PerlIO::encoding };
41 # (Temporary|legacy) methods
43 sub toUnicode { shift->decode(@_) }
44 sub fromUnicode { shift->encode(@_) }
47 # Needs to be overloaded or just croak
53 my $class = ref($obj) ? ref($obj) : $obj;
54 Carp::croak($class . "->encode() not defined!");
60 my $class = ref($obj) ? ref($obj) : $obj;
61 Carp::croak($class . "->encode() not defined!");
71 Encode::Encoding - Encode Implementation Base Class
75 package Encode::MyEncoding;
76 use base qw(Encode::Encoding);
78 __PACKAGE__->Define(qw(myCanonical myAlias));
82 As mentioned in L<Encode>, encodings are (in the current
83 implementation at least) defined as objects. The mapping of encoding
84 name to object is via the C<%Encode::Encoding> hash. Though you can
85 directly manipulate this hash, it is strongly encouraged to use this
86 base class module and add encode() and decode() methods.
88 =head2 Methods you should implement
90 You are strongly encouraged to implement methods below, at least
91 either encode() or decode().
95 =item -E<gt>encode($string [,$check])
97 MUST return the octet sequence representing I<$string>.
103 If I<$check> is true, it SHOULD modify I<$string> in place to remove
104 the converted part (i.e. the whole string unless there is an error).
105 If perlio_ok() is true, SHOULD becomes MUST.
109 If an error occurs, it SHOULD return the octet sequence for the
110 fragment of string that has been converted and modify $string in-place
111 to remove the converted part leaving it starting with the problem
112 fragment. If perlio_ok() is true, SHOULD becomes MUST.
116 If I<$check> is is false then C<encode> MUST make a "best effort" to
117 convert the string - for example, by using a replacement character.
121 =item -E<gt>decode($octets [,$check])
123 MUST return the string that I<$octets> represents.
129 If I<$check> is true, it SHOULD modify I<$octets> in place to remove
130 the converted part (i.e. the whole sequence unless there is an
131 error). If perlio_ok() is true, SHOULD becomes MUST.
135 If an error occurs, it SHOULD return the fragment of string that has
136 been converted and modify $octets in-place to remove the converted
137 part leaving it starting with the problem fragment. If perlio_ok() is
138 true, SHOULD becomes MUST.
142 If I<$check> is false then C<decode> should make a "best effort" to
143 convert the string - for example by using Unicode's "\x{FFFD}" as a
144 replacement character.
150 If you want your encoding to work with L<encoding> pragma, you should
151 also implement the method below.
155 =item -E<gt>cat_decode($destination, $octets, $offset, $terminator [,$check])
157 MUST decode I<$octets> with I<$offset> and concatenate it to I<$destination>.
158 Decoding will terminate when $terminator (a string) appears in output.
159 I<$offset> will be modified to the last $octets position at end of decode.
160 Returns true if $terminator appears output, else returns false.
164 =head2 Other methods defined in Encode::Encodings
166 You do not have to override methods shown below unless you have to.
174 sub name { return shift->{'Name'} }
176 MUST return the string representing the canonical name of the encoding.
184 my $clone = bless { %$self } => ref($self);
189 This method reconstructs the encoding object if necessary. If you need
190 to store the state during encoding, this is where you clone your object.
192 PerlIO ALWAYS calls this method to make sure it has its own private
199 sub renewed { $_[0]->{renewed} || 0 }
201 Tells whether the object is renewed (and how many times). Some
202 modules emit C<Use of uninitialized value in null operation> warning
203 unless the value is numeric so return 0 for false.
205 =item -E<gt>perlio_ok()
210 eval{ require PerlIO::encoding };
214 If your encoding does not support PerlIO for some reasons, just;
218 =item -E<gt>needs_lines()
222 sub needs_lines { 0 };
224 If your encoding can work with PerlIO but needs line buffering, you
225 MUST define this method so it returns true. 7bit ISO-2022 encodings
226 are one example that needs this. When this method is missing, false
231 =head2 Example: Encode::ROT13
233 package Encode::ROT13;
235 use base qw(Encode::Encoding);
237 __PACKAGE__->Define('rot13');
240 my ($obj, $str, $chk) = @_;
241 $str =~ tr/A-Za-z/N-ZA-Mn-za-m/;
242 $_[1] = '' if $chk; # this is what in-place edit means
246 # Jr pna or ynml yvxr guvf;
251 =head1 Why the heck Encode API is different?
253 It should be noted that the I<$check> behaviour is different from the
254 outer public API. The logic is that the "unchecked" case is useful
255 when the encoding is part of a stream which may be reporting errors
256 (e.g. STDERR). In such cases, it is desirable to get everything
257 through somehow without causing additional errors which obscure the
258 original one. Also, the encoding is best placed to know what the
259 correct replacement character is, so if that is the desired behaviour
260 then letting low level code do it is the most efficient.
262 By contrast, if I<$check> is true, the scheme above allows the
263 encoding to do as much as it can and tell the layer above how much
264 that was. What is lacking at present is a mechanism to report what
265 went wrong. The most likely interface will be an additional method
266 call to the object, or perhaps (to avoid forcing per-stream objects
267 on otherwise stateless encodings) an additional parameter.
269 It is also highly desirable that encoding classes inherit from
270 C<Encode::Encoding> as a base class. This allows that class to define
271 additional behaviour for all encoding objects.
273 package Encode::MyEncoding;
274 use base qw(Encode::Encoding);
276 __PACKAGE__->Define(qw(myCanonical myAlias));
278 to create an object with C<< bless {Name => ...}, $class >>, and call
279 define_encoding. They inherit their C<name> method from
282 =head2 Compiled Encodings
284 For the sake of speed and efficiency, most of the encodings are now
285 supported via a I<compiled form>: XS modules generated from UCM
286 files. Encode provides the enc2xs tool to achieve that. Please see
287 L<enc2xs> for more details.
291 L<perlmod>, L<enc2xs>
299 The fixup routine gets passed the remaining fragment of string being
300 processed. It modifies it in place to remove bytes/characters it can
301 understand and returns a string used to represent them. For example:
304 my $ch = substr($_[0],0,1,'');
305 return sprintf("\x{%02X}",ord($ch);
308 This scheme is close to how the underlying C code for Encode works,
309 but gives the fixup routine very little context.
313 The fixup routine gets passed the original string, an index into
314 it of the problem area, and the output string so far. It appends
315 what it wants to the output string and returns a new index into the
316 original string. For example:
319 # my ($s,$i,$d) = @_;
320 my $ch = substr($_[0],$_[1],1);
321 $_[2] .= sprintf("\x{%02X}",ord($ch);
325 This scheme gives maximal control to the fixup routine but is more
326 complicated to code, and may require that the internals of Encode be tweaked to
327 keep the original string intact.
331 Hybrids of the above.
333 Multiple return values rather than in-place modifications.
335 Index into the string could be C<pos($str)> allowing C<s/\G...//>.