2 use bytes (); # for $bytes::hint_bits
3 $charnames::hint_bits = 0x20000;
5 my $fname = 'unicode/UnicodeData-Latest.txt';
8 # This is not optimized in any way yet
11 $txt = do "unicode/Name.pl" unless $txt;
13 if ($^H{charnames_full} and $txt =~ /\t\t$name$/m) {
14 @off = ($-[0], $+[0]);
17 if ($^H{charnames_short} and $name =~ /^(.*?):(.*)/s) {
18 my ($script, $cname) = ($1,$2);
19 my $case = ( $cname =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL");
20 if ($txt =~ m/\t\t\U$script\E (?:$case )?LETTER \U$cname$/m) {
21 @off = ($-[0], $+[0]);
26 my $case = ( $name =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL");
27 for ( @{$^H{charnames_scripts}} ) {
28 (@off = ($-[0], $+[0])), last
29 if $txt =~ m/\t\t$_ (?:$case )?LETTER \U$name$/m;
32 die "Unknown charname '$name'" unless @off;
34 my $ord = hex substr $txt, $off[0] - 4, 4;
35 if ($^H & $bytes::hint_bits) { # "use bytes" in effect?
37 return chr $ord if $ord <= 255;
38 my $hex = sprintf '%X=0%o', $ord, $ord;
39 my $fname = substr $txt, $off[0] + 2, $off[1] - $off[0] - 2;
40 die "Character 0x$hex with name '$fname' is above 0xFF";
47 die "`use charnames' needs explicit imports list" unless @_;
48 $^H |= $charnames::hint_bits;
49 $^H{charnames} = \&charnames ;
52 $^H{charnames_full} = delete $h{':full'};
53 $^H{charnames_short} = delete $h{':short'};
54 $^H{charnames_scripts} = [map uc, keys %h];
63 charnames - define character names for C<\N{named}> string literal escape.
67 use charnames ':full';
68 print "\N{GREEK SMALL LETTER SIGMA} is called sigma.\n";
70 use charnames ':short';
71 print "\N{greek:Sigma} is an upper-case sigma.\n";
73 use charnames qw(cyrillic greek);
74 print "\N{sigma} is Greek sigma, and \N{be} is Cyrillic b.\n";
78 Pragma C<use charnames> supports arguments C<:full>, C<:short> and
79 script names. If C<:full> is present, for expansion of
80 C<\N{CHARNAME}}> string C<CHARNAME> is first looked in the list of
81 standard Unicode names of chars. If C<:short> is present, and
82 C<CHARNAME> has the form C<SCRIPT:CNAME>, then C<CNAME> is looked up
83 as a letter in script C<SCRIPT>. If pragma C<use charnames> is used
84 with script name arguments, then for C<\N{CHARNAME}}> the name
85 C<CHARNAME> is looked up as a letter in the given scripts (in the
88 For lookup of C<CHARNAME> inside a given script C<SCRIPTNAME>
89 this pragma looks for the names
91 SCRIPTNAME CAPITAL LETTER CHARNAME
92 SCRIPTNAME SMALL LETTER CHARNAME
93 SCRIPTNAME LETTER CHARNAME
95 in the table of standard Unicode names. If C<CHARNAME> is lowercase,
96 then the C<CAPITAL> variant is ignored, otherwise the C<SMALL> variant is
99 =head1 CUSTOM TRANSLATORS
101 The mechanism of translation of C<\N{...}> escapes is general and not
102 hardwired into F<charnames.pm>. A module can install custom
103 translations (inside the scope which C<use>s the module) with the
104 following magic incantation:
106 use charnames (); # for $charnames::hint_bits
109 $^H |= $charnames::hint_bits;
110 $^H{charnames} = \&translator;
113 Here translator() is a subroutine which takes C<CHARNAME> as an
114 argument, and returns text to insert into the string instead of the
115 C<\N{CHARNAME}> escape. Since the text to insert should be different
116 in C<bytes> mode and out of it, the function should check the current
117 state of C<bytes>-flag as in:
119 use bytes (); # for $bytes::hint_bits
121 if ($^H & $bytes::hint_bits) {
122 return bytes_translator(@_);
125 return utf8_translator(@_);
131 Since evaluation of the translation function happens in a middle of
132 compilation (of a string literal), the translation function should not
133 do any C<eval>s or C<require>s. This restriction should be lifted in
134 a future version of Perl.