2 our $VERSION = do { my @r = (q$Revision: 1.26 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
10 Carp::croak "encoding pragma does not support EBCDIC platforms";
18 $name ||= $ENV{PERL_ENCODING};
20 my $enc = find_encoding($name);
21 unless (defined $enc) {
23 Carp::croak "Unknown encoding '$name'";
25 unless ($arg{Filter}){
26 ${^ENCODING} = $enc; # this is all you need, actually.
27 for my $h (qw(STDIN STDOUT)){
29 unless (defined find_encoding($arg{h})) {
31 Carp::croak "Unknown encoding for $h, '$arg{$h}'";
33 eval qq{ binmode($h, ":encoding($arg{$h})") };
35 unless (exists $arg{$h}){
36 eval qq{ binmode($h, ":encoding($name)") };
45 defined(${^ENCODING}) and undef ${^ENCODING};
47 require Filter::Util::Call ;
48 Filter::Util::Call->import ;
49 binmode(STDIN, ":raw");
50 binmode(STDOUT, ":raw");
53 if (($status = filter_read()) > 0){
54 $_ = $enc->decode($_, 1);
60 # warn "Filter installed";
62 return 1; # I doubt if we need it, though
68 binmode(STDIN, ":raw");
69 binmode(STDOUT, ":raw");
70 if ($INC{"Filter/Util/Call.pm"}){
71 eval { filter_del() };
81 encoding - allows you to write your script in non-asii or non-utf8
85 use encoding "euc-jp"; # Jperl!
87 # or you can even do this if your shell supports euc-jp
89 > perl -Mencoding=euc-jp -e '...'
91 # or from the shebang line
93 #!/your/path/to/perl -Mencoding=euc-jp
97 # A simple euc-jp => utf-8 converter
98 use encoding "euc-jp", STDOUT => "utf8"; while(<>){print};
100 # "no encoding;" supported (but not scoped!)
103 # an alternate way, Filter
104 use encoding "euc-jp", Filter=>1;
106 # now you can use kanji identifiers -- in euc-jp!
110 Perl 5.6.0 has introduced Unicode support. You could apply
111 C<substr()> and regexes even to complex CJK characters -- so long as
112 the script was written in UTF-8. But back then text editors that
113 support UTF-8 was still rare and many users rather chose to writer
114 scripts in legacy encodings, given up whole new feature of Perl 5.6.
116 With B<encoding> pragma, you can write your script in any encoding you like
117 (so long as the C<Encode> module supports it) and still enjoy Unicode
118 support. You can write a code in EUC-JP as follows;
120 my $Rakuda = "\xF1\xD1\xF1\xCC"; # Camel in Kanji
121 #<-char-><-char-> # 4 octets
122 s/\bCamel\b/$Rakuda/;
124 And with C<use encoding "euc-jp"> in effect, it is the same thing as
125 the code in UTF-8 as follow.
127 my $Rakuda = "\x{99F1}\x{99DD}"; # who Unicode Characters
128 s/\bCamel\b/$Rakuda/;
130 The B<encoding> pragma also modifies the file handle disciplines of
131 STDIN, STDOUT, and STDERR to the specified encoding. Therefore,
133 use encoding "euc-jp";
134 my $message = "Camel is the symbol of perl.\n";
135 my $Rakuda = "\xF1\xD1\xF1\xCC"; # Camel in Kanji
136 $message =~ s/\bCamel\b/$Rakuda/;
139 Will print "\xF1\xD1\xF1\xCC is the symbol of perl.\n", not
140 "\x{99F1}\x{99DD} is the symbol of perl.\n".
142 You can override this by giving extra arguments. See below.
148 =item use encoding [I<ENCNAME>] ;
150 Sets the script encoding to I<ENCNAME> and file handle disciplines of
151 STDIN, STDOUT are set to ":encoding(I<ENCNAME>)". Note STDERR will not
154 If no encoding is specified, the environment variable L<PERL_ENCODING>
155 is consulted. If no encoding can be found, C<Unknown encoding 'I<ENCNAME>'>
156 error will be thrown.
158 Note that non-STD file handles remain unaffected. Use C<use open> or
159 C<binmode> to change disciplines of those.
161 =item use encoding I<ENCNAME> [ STDIN =E<gt> I<ENCNAME_IN> ...] ;
163 You can also individually set encodings of STDIN and STDOUT via
164 STDI<FH> =E<gt> I<ENCNAME_FH> form. In this case, you cannot omit the
165 first I<ENCNAME>. C<STDI<FH> =E<gt> undef> turns IO transcoding
170 Unsets the script encoding and the disciplines of STDIN, STDOUT are
179 The pragma is a per script, not a per block lexical. Only the last
180 C<use encoding> or C<matters, and it affects B<the whole script>.
181 Though <no encoding> pragma is supported and C<use encoding> can
182 appear as many times as you want in a given script, the multiple use
183 of this pragma is discouraged.
185 =head2 DO NOT MIX MULTIPLE ENCODINGS
187 Notice that only literals (string or regular expression) having only
188 legacy code points are affected: if you mix data like this
192 the data is assumed to be in (Latin 1 and) Unicode, not in your native
193 encoding. In other words, this will match in "greek":
199 "\xDF\x{100}" =~ /\x{3af}\x{100}/
201 since the C<\xDF> on the left will B<not> be upgraded to C<\x{3af}>
202 because of the C<\x{100}> on the left. You should not be mixing your
203 legacy data and Unicode in the same string.
205 This pragma also affects encoding of the 0x80..0xFF code point range:
206 normally characters in that range are left as eight-bit bytes (unless
207 they are combined with characters with code points 0x100 or larger,
208 in which case all characters need to become UTF-8 encoded), but if
209 the C<encoding> pragma is present, even the 0x80..0xFF range always
212 After all, the best thing about this pragma is that you don't have to
213 resort to \x... just to spell your name in native encoding. So feel
214 free to put your strings in your encoding in quotes and regexes.
216 =head1 NON-ASCII Identifiers and Filter option
218 The magic of C<use encoding> is not applied to the names of identifiers.
219 In order to make C<${"4eba"}++> ($man++, where man is a single ideograph)
220 work, you still need to write your script in UTF-8 or use a source filter.
222 In other words, the same restriction as Jperl applies.
224 If you dare experiment, however, you can try Fitlter option.
228 =item use encoding I<ENCNAME> Filter=E<gt>1;
230 This turns encoding pragma into source filter. While the default
231 approach just decodes interpolated literals (in qq() and qr()), this
232 will apply source filter to entire source code. In this case, STDIN
233 and STDOUT remain untouched.
237 What does this mean? Your source code behaves as if it is written
238 in UTF-8. So even if your editor only supports Shift_JIS, for
239 example. You can still try examples in Chapter 15 of
240 C<Programming Perl, 3rd Ed.> For instance, you can use UTF-8
243 This option is significantly slower and (as of this writing) non-ASCII
244 identifiers are not very stable WITHOUT this option and with the
245 source code written in UTF-8.
247 To make your script in legacy encoding work with minimum effort, do
248 not use Filter=E<gt>1
251 =head1 EXAMPLE - Greekperl
253 use encoding "iso 8859-7";
255 # The \xDF of ISO 8859-7 (Greek) is \x{3af} in Unicode.
260 printf "%#x\n", ord($a); # will print 0x3af, not 0xdf
264 # $c will be "\x{3af}\x{100}", not "\x{df}\x{100}".
266 # chr() is affected, and ...
268 print "mega\n" if ord(chr(0xdf)) == 0x3af;
270 # ... ord() is affected by the encoding pragma ...
272 print "tera\n" if ord(pack("C", 0xdf)) == 0x3af;
274 # ... as are eq and cmp ...
276 print "peta\n" if "\x{3af}" eq pack("C", 0xdf);
277 print "exa\n" if "\x{3af}" cmp pack("C", 0xdf) == 0;
279 # ... but pack/unpack C are not affected, in case you still
280 # want back to your native encoding
282 print "zetta\n" if unpack("C", (pack("C", 0xdf))) == 0xdf;
284 =head1 KNOWN PROBLEMS
286 For native multibyte encodings (either fixed or variable length)
287 the current implementation of the regular expressions may introduce
288 recoding errors for longer regular expression literals than 127 bytes.
290 The encoding pragma is not supported on EBCDIC platforms.
295 L<perlunicode>, L<Encode>, L<open>, L<Filter::Util::Call>,
297 Ch. 15 of C<Programming Perl (3rd Edition)>
298 by Larry Wall, Tom Christiansen, Jon Orwant;
299 O'Reilly & Associates; ISBN 0-596-00027-8