xref: /openbsd-src/gnu/usr.bin/perl/lib/utf8.pm (revision b2ea75c1b17e1a9a339660e7ed45cd24946b230e)
1package utf8;
2
3if (ord('A') != 193) { # make things more pragmatic for EBCDIC folk
4
5$utf8::hint_bits = 0x00800000;
6
7sub import {
8    $^H |= $utf8::hint_bits;
9    $enc{caller()} = $_[1] if $_[1];
10}
11
12sub unimport {
13    $^H &= ~$utf8::hint_bits;
14}
15
16sub AUTOLOAD {
17    require "utf8_heavy.pl";
18    goto &$AUTOLOAD if defined &$AUTOLOAD;
19    Carp::croak("Undefined subroutine $AUTOLOAD called");
20}
21
22}
23
241;
25__END__
26
27=head1 NAME
28
29utf8 - Perl pragma to enable/disable UTF-8 in source code
30
31=head1 SYNOPSIS
32
33    use utf8;
34    no utf8;
35
36=head1 DESCRIPTION
37
38WARNING: The implementation of Unicode support in Perl is incomplete.
39See L<perlunicode> for the exact details.
40
41The C<use utf8> pragma tells the Perl parser to allow UTF-8 in the
42program text in the current lexical scope.  The C<no utf8> pragma
43tells Perl to switch back to treating the source text as literal
44bytes in the current lexical scope.
45
46This pragma is primarily a compatibility device.  Perl versions
47earlier than 5.6 allowed arbitrary bytes in source code, whereas
48in future we would like to standardize on the UTF-8 encoding for
49source text.  Until UTF-8 becomes the default format for source
50text, this pragma should be used to recognize UTF-8 in the source.
51When UTF-8 becomes the standard source format, this pragma will
52effectively become a no-op.  This pragma already is a no-op on
53EBCDIC platforms (where it is alright to code perl in EBCDIC
54rather than UTF-8).
55
56Enabling the C<utf8> pragma has the following effects:
57
58=over
59
60=item *
61
62Bytes in the source text that have their high-bit set will be treated
63as being part of a literal UTF-8 character.  This includes most literals
64such as identifiers, string constants, constant regular expression patterns
65and package names.
66
67=item *
68
69In the absence of inputs marked as UTF-8, regular expressions within the
70scope of this pragma will default to using character semantics instead
71of byte semantics.
72
73    @bytes_or_chars = split //, $data;	# may split to bytes if data
74					# $data isn't UTF-8
75    {
76	use utf8;			# force char semantics
77	@chars = split //, $data;	# splits characters
78    }
79
80=head1 SEE ALSO
81
82L<perlunicode>, L<bytes>
83
84=cut
85