This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Re: maint@20974 or before broke mp2 ithreads test
[perl5.git] / lib / bytes.pm
CommitLineData
657b208b 1package bytes;
5bc28da9 2
b75c8c73
MS
3our $VERSION = '1.00';
4
d5448623
GS
5$bytes::hint_bits = 0x00000008;
6
5bc28da9 7sub import {
d5448623 8 $^H |= $bytes::hint_bits;
5bc28da9
NIS
9}
10
11sub unimport {
d5448623 12 $^H &= ~$bytes::hint_bits;
5bc28da9
NIS
13}
14
15sub AUTOLOAD {
657b208b 16 require "bytes_heavy.pl";
5bc28da9
NIS
17 goto &$AUTOLOAD;
18}
19
20sub length ($);
21
221;
23__END__
24
25=head1 NAME
26
657b208b 27bytes - Perl pragma to force byte semantics rather than character semantics
5bc28da9
NIS
28
29=head1 SYNOPSIS
30
657b208b
GS
31 use bytes;
32 no bytes;
5bc28da9
NIS
33
34=head1 DESCRIPTION
35
657b208b
GS
36The C<use bytes> pragma disables character semantics for the rest of the
37lexical scope in which it appears. C<no bytes> can be used to reverse
38the effect of C<use bytes> within the current lexical scope.
393fec97 39
5de28535
SC
40Perl normally assumes character semantics in the presence of character
41data (i.e. data that has come from a source that has been marked as
42being of a particular character encoding). When C<use bytes> is in
43effect, the encoding is temporarily ignored, and each string is treated
44as a series of bytes.
45
46As an example, when Perl sees C<$x = chr(400)>, it encodes the character
c26c758b 47in UTF-8 and stores it in $x. Then it is marked as character data, so,
5de28535
SC
48for instance, C<length $x> returns C<1>. However, in the scope of the
49C<bytes> pragma, $x is treated as a series of bytes - the bytes that make
50up the UTF8 encoding - and C<length $x> returns C<2>:
51
52 $x = chr(400);
53 print "Length is ", length $x, "\n"; # "Length is 1"
54 printf "Contents are %vd\n", $x; # "Contents are 400"
55 {
56 use bytes;
57 print "Length is ", length $x, "\n"; # "Length is 2"
58 printf "Contents are %vd\n", $x; # "Contents are 198.144"
59 }
60
61For more on the implications and differences between character
393fec97
GS
62semantics and byte semantics, see L<perlunicode>.
63
64=head1 SEE ALSO
65
66L<perlunicode>, L<utf8>
5bc28da9
NIS
67
68=cut