5 use bytes (); # for $bytes::hint_bits
7 $charnames::hint_bits = 0x20000;
11 # This is not optimized in any way yet
14 $txt = do "unicode/Name.pl" unless $txt;
16 if ($^H{charnames_full} and $txt =~ /\t\t$name$/m) {
17 @off = ($-[0], $+[0]);
20 if ($^H{charnames_short} and $name =~ /^(.*?):(.*)/s) {
21 my ($script, $cname) = ($1,$2);
22 my $case = ( $cname =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL");
23 if ($txt =~ m/\t\t\U$script\E (?:$case )?LETTER \U$cname$/m) {
24 @off = ($-[0], $+[0]);
29 my $case = ( $name =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL");
30 for ( @{$^H{charnames_scripts}} ) {
31 (@off = ($-[0], $+[0])), last
32 if $txt =~ m/\t\t$_ (?:$case )?LETTER \U$name$/m;
35 die "Unknown charname '$name'" unless @off;
38 $hexlen++ while substr($txt, $off[0] - $hexlen - 1, 1) =~ /[0-9a-f]/;
39 my $ord = hex substr $txt, $off[0] - $hexlen, $hexlen;
40 if ($^H & $bytes::hint_bits) { # "use bytes" in effect?
42 return chr $ord if $ord <= 255;
43 my $hex = sprintf '%X=0%o', $ord, $ord;
44 my $fname = substr $txt, $off[0] + 2, $off[1] - $off[0] - 2;
45 die "Character 0x$hex with name '$fname' is above 0xFF";
52 die "`use charnames' needs explicit imports list" unless @_;
53 $^H |= $charnames::hint_bits;
54 $^H{charnames} = \&charnames ;
57 $^H{charnames_full} = delete $h{':full'};
58 $^H{charnames_short} = delete $h{':short'};
59 $^H{charnames_scripts} = [map uc, keys %h];
60 if (warnings::enabled('utf8') && @{$^H{charnames_scripts}}) {
61 $txt = do "unicode/Name.pl" unless $txt;
62 for (@{$^H{charnames_scripts}}) {
63 warnings::warn('utf8', "No such script: '$_'") unless
64 $txt =~ m/\t\t$_ (?:CAPITAL |SMALL )?LETTER /;
75 charnames - define character names for C<\N{named}> string literal escape.
79 use charnames ':full';
80 print "\N{GREEK SMALL LETTER SIGMA} is called sigma.\n";
82 use charnames ':short';
83 print "\N{greek:Sigma} is an upper-case sigma.\n";
85 use charnames qw(cyrillic greek);
86 print "\N{sigma} is Greek sigma, and \N{be} is Cyrillic b.\n";
90 Pragma C<use charnames> supports arguments C<:full>, C<:short> and
91 script names. If C<:full> is present, for expansion of
92 C<\N{CHARNAME}}> string C<CHARNAME> is first looked in the list of
93 standard Unicode names of chars. If C<:short> is present, and
94 C<CHARNAME> has the form C<SCRIPT:CNAME>, then C<CNAME> is looked up
95 as a letter in script C<SCRIPT>. If pragma C<use charnames> is used
96 with script name arguments, then for C<\N{CHARNAME}}> the name
97 C<CHARNAME> is looked up as a letter in the given scripts (in the
100 For lookup of C<CHARNAME> inside a given script C<SCRIPTNAME>
101 this pragma looks for the names
103 SCRIPTNAME CAPITAL LETTER CHARNAME
104 SCRIPTNAME SMALL LETTER CHARNAME
105 SCRIPTNAME LETTER CHARNAME
107 in the table of standard Unicode names. If C<CHARNAME> is lowercase,
108 then the C<CAPITAL> variant is ignored, otherwise the C<SMALL> variant is
111 =head1 CUSTOM TRANSLATORS
113 The mechanism of translation of C<\N{...}> escapes is general and not
114 hardwired into F<charnames.pm>. A module can install custom
115 translations (inside the scope which C<use>s the module) with the
116 following magic incantation:
118 use charnames (); # for $charnames::hint_bits
121 $^H |= $charnames::hint_bits;
122 $^H{charnames} = \&translator;
125 Here translator() is a subroutine which takes C<CHARNAME> as an
126 argument, and returns text to insert into the string instead of the
127 C<\N{CHARNAME}> escape. Since the text to insert should be different
128 in C<bytes> mode and out of it, the function should check the current
129 state of C<bytes>-flag as in:
131 use bytes (); # for $bytes::hint_bits
133 if ($^H & $bytes::hint_bits) {
134 return bytes_translator(@_);
137 return utf8_translator(@_);
143 Since evaluation of the translation function happens in a middle of
144 compilation (of a string literal), the translation function should not
145 do any C<eval>s or C<require>s. This restriction should be lifted in
146 a future version of Perl.