package charnames;
use strict;
use warnings;
-use Carp;
use File::Spec;
-our $VERSION = '1.02';
+our $VERSION = '1.08';
use bytes (); # for $bytes::hint_bits
-$charnames::hint_bits = 0x20000; # HINT_LOCALIZE_HH
my %alias1 = (
# Icky 3.2 names with parentheses.
);
my $txt;
+sub croak
+{
+ require Carp; goto &Carp::croak;
+} # croak
+
+sub carp
+{
+ require Carp; goto &Carp::carp;
+} # carp
+
sub alias (@)
{
@_ or return %alias3;
## we know where it starts, so turn into number -
## the ordinal for the char.
- $ord = hex substr($txt, $hexstart, $off[0] - $hexstart);
+ $ord = CORE::hex substr($txt, $hexstart, $off[0] - $hexstart);
}
if ($^H & $bytes::hint_bits) { # "use bytes" in effect?
if (not @_) {
carp("`use charnames' needs explicit imports list");
}
- $^H |= $charnames::hint_bits;
$^H{charnames} = \&charnames ;
##
## fill %h keys with our @_ args.
##
my ($promote, %h, @args) = (0);
- while (@_ and $_ = shift) {
- if ($_ eq ":alias") {
+ while (my $arg = shift) {
+ if ($arg eq ":alias") {
@_ or
croak ":alias needs an argument in charnames";
my $alias = shift;
alias_file ($alias);
next;
}
- if (m/^:/ and ! ($_ eq ":full" || $_ eq ":short")) {
- warn "unsupported special '$_' in charnames";
+ if (substr($arg, 0, 1) eq ':' and ! ($arg eq ":full" || $arg eq ":short")) {
+ warn "unsupported special '$arg' in charnames";
next;
}
- push @args, $_;
+ push @args, $arg;
}
@args == 0 && $promote and @args = (":full");
@h{@args} = (1) x @args;
}
} # import
-require Unicode::UCD; # for Unicode::UCD::_getcode()
-
my %viacode;
sub viacode
{
if (@_ != 1) {
carp "charnames::viacode() expects one argument";
- return ()
+ return;
}
my $arg = shift;
- my $code = Unicode::UCD::_getcode($arg);
+ # this is derived from Unicode::UCD, where it is nearly the same as the
+ # function _getcode(), but it makes sure that even a hex argument has the
+ # proper number of leading zeros, which is critical in matching against $txt
+ # below
my $hex;
-
- if (defined $code) {
+ if ($arg =~ /^[1-9]\d*$/) {
$hex = sprintf "%04X", $arg;
+ } elsif ($arg =~ /^(?:[Uu]\+|0[xX])?([[:xdigit:]]+)$/) {
+ # Below is the line that differs from the _getcode() source
+ $hex = sprintf "%04X", hex $arg;
} else {
carp("unexpected arg \"$arg\" to charnames::viacode()");
return;
}
- if ($code > 0x10FFFF) {
- carp sprintf "Unicode characters only allocated up to U+10FFFF (you asked for U+%X)", $hex;
+ # checking the length first is slightly faster
+ if (length($hex) > 5 && hex($hex) > 0x10FFFF) {
+ carp "Unicode characters only allocated up to U+10FFFF (you asked for U+$hex)";
return;
}
$txt = do "unicore/Name.pl" unless $txt;
- if ($txt =~ m/^$hex\t\t(.+)/m) {
- return $viacode{$hex} = $1;
- } else {
- return;
- }
+ return unless $txt =~ m/^$hex\t\t(.+)/m;
+
+ $viacode{$hex} = $1;
} # viacode
my %vianame;
my $arg = shift;
- return chr hex $1 if $arg =~ /^U\+([0-9a-fA-F]+)$/;
+ return chr CORE::hex $1 if $arg =~ /^U\+([0-9a-fA-F]+)$/;
return $vianame{$arg} if exists $vianame{$arg};
$txt = do "unicore/Name.pl" unless $txt;
my $pos = index $txt, "\t\t$arg\n";
- if ($[ <= $pos) {
+ if (0 <= $pos) {
my $posLF = rindex $txt, "\n", $pos;
(my $code = substr $txt, $posLF + 1, 6) =~ tr/\t//d;
- return $vianame{$arg} = hex $code;
+ return $vianame{$arg} = CORE::hex $code;
- # If $pos is at the 1st line, $posLF must be $[ - 1 (not found);
- # then $posLF + 1 equals to $[ (at the beginning of $txt).
+ # If $pos is at the 1st line, $posLF must be -1 (not found);
+ # then $posLF + 1 equals to 0 (at the beginning of $txt).
# Otherwise $posLF is the position of "\n";
# then $posLF + 1 must be the position of the next to "\n"
# (the beginning of the line).
use charnames ":full", ":alias" => {
e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
- };
+ };
print "\N{e_ACUTE} is a small letter e with an acute.\n";
+ use charnames ();
print charnames::viacode(0x1234); # prints "ETHIOPIC SYLLABLE SEE"
printf "%04X", charnames::vianame("GOTHIC LETTER AHSA"); # prints "10330"
Pragma C<use charnames> supports arguments C<:full>, C<:short>, script
names and customized aliases. If C<:full> is present, for expansion of
-C<\N{CHARNAME}> string C<CHARNAME> is first looked in the list of
-standard Unicode names of chars. If C<:short> is present, and
+C<\N{CHARNAME}>, the string C<CHARNAME> is first looked up in the list of
+standard Unicode character names. If C<:short> is present, and
C<CHARNAME> has the form C<SCRIPT:CNAME>, then C<CNAME> is looked up
as a letter in script C<SCRIPT>. If pragma C<use charnames> is used
with script name arguments, then for C<\N{CHARNAME}> the name
Since the Unicode standard uses "U+HHHH", so can you: "\N{U+263a}"
is the Unicode smiley face, or "\N{WHITE SMILING FACE}".
-=head1 CUSTOM TRANSLATORS
+=head1 ALIASES
-The mechanism of translation of C<\N{...}> escapes is general and not
-hardwired into F<charnames.pm>. A module can install custom
-translations (inside the scope which C<use>s the module) with the
-following magic incantation:
+A few aliases have been defined for convenience: instead of having
+to use the official names
- use charnames (); # for $charnames::hint_bits
- sub import {
- shift;
- $^H |= $charnames::hint_bits;
- $^H{charnames} = \&translator;
- }
+ LINE FEED (LF)
+ FORM FEED (FF)
+ CARRIAGE RETURN (CR)
+ NEXT LINE (NEL)
-Here translator() is a subroutine which takes C<CHARNAME> as an
-argument, and returns text to insert into the string instead of the
-C<\N{CHARNAME}> escape. Since the text to insert should be different
-in C<bytes> mode and out of it, the function should check the current
-state of C<bytes>-flag as in:
+(yes, with parentheses) one can use
- use bytes (); # for $bytes::hint_bits
- sub translator {
- if ($^H & $bytes::hint_bits) {
- return bytes_translator(@_);
- }
- else {
- return utf8_translator(@_);
- }
- }
+ LINE FEED
+ FORM FEED
+ CARRIAGE RETURN
+ NEXT LINE
+ LF
+ FF
+ CR
+ NEL
+
+One can also use
+
+ BYTE ORDER MARK
+ BOM
+
+and
+
+ ZWNJ
+ ZWJ
+
+for ZERO WIDTH NON-JOINER and ZERO WIDTH JOINER.
+
+For backward compatibility one can use the old names for
+certain C0 and C1 controls
+
+ old new
+
+ HORIZONTAL TABULATION CHARACTER TABULATION
+ VERTICAL TABULATION LINE TABULATION
+ FILE SEPARATOR INFORMATION SEPARATOR FOUR
+ GROUP SEPARATOR INFORMATION SEPARATOR THREE
+ RECORD SEPARATOR INFORMATION SEPARATOR TWO
+ UNIT SEPARATOR INFORMATION SEPARATOR ONE
+ PARTIAL LINE DOWN PARTIAL LINE FORWARD
+ PARTIAL LINE UP PARTIAL LINE BACKWARD
+
+but the old names in addition to giving the character
+will also give a warning about being deprecated.
=head1 CUSTOM ALIASES
This version of charnames supports three mechanisms of adding local
-or customized aliases to standard Unicode naming conventions (:full)
+or customized aliases to standard Unicode naming conventions (:full).
+
+Note that an alias should not be something that is a legal curly
+brace-enclosed quantifier (see L<perlreref/QUANTIFIERS>). For example
+C<\N{123}> means to match 123 non-newline characters, and is not treated as an
+alias. Aliases are discouraged from beginning with anything other than an
+alphabetic character and from containing anything other than alphanumerics,
+spaces, dashes, colons, parentheses, and underscores. Currently they must be
+ASCII.
=head2 Anonymous hashes
This works only for the standard names, and does not yet apply
to custom translators.
-=head1 ALIASES
-
-A few aliases have been defined for convenience: instead of having
-to use the official names
-
- LINE FEED (LF)
- FORM FEED (FF)
- CARRIAGE RETURN (CR)
- NEXT LINE (NEL)
+=head1 CUSTOM TRANSLATORS
-(yes, with parentheses) one can use
+The mechanism of translation of C<\N{...}> escapes is general and not
+hardwired into F<charnames.pm>. A module can install custom
+translations (inside the scope which C<use>s the module) with the
+following magic incantation:
- LINE FEED
- FORM FEED
- CARRIAGE RETURN
- NEXT LINE
- LF
- FF
- CR
- NEL
+ sub import {
+ shift;
+ $^H{charnames} = \&translator;
+ }
-One can also use
+Here translator() is a subroutine which takes C<CHARNAME> as an
+argument, and returns text to insert into the string instead of the
+C<\N{CHARNAME}> escape. Since the text to insert should be different
+in C<bytes> mode and out of it, the function should check the current
+state of C<bytes>-flag as in:
- BYTE ORDER MARK
- BOM
+ use bytes (); # for $bytes::hint_bits
+ sub translator {
+ if ($^H & $bytes::hint_bits) {
+ return bytes_translator(@_);
+ }
+ else {
+ return utf8_translator(@_);
+ }
+ }
-and
+See L</CUSTOM ALIASES> above for restrictions on C<CHARNAME>.
- ZWNJ
- ZWJ
+=head1 ILLEGAL CHARACTERS
-for ZERO WIDTH NON-JOINER and ZERO WIDTH JOINER.
+If you ask by name for a character that does not exist, a warning is given and
+the Unicode I<replacement character> "\x{FFFD}" is returned.
-For backward compatibility one can use the old names for
-certain C0 and C1 controls
+If you ask by code for a character that is unassigned, no warning is
+given and C<undef> is returned. (Though if you ask for a code point
+past U+10FFFF you do get a warning.) See L</BUGS> below.
- old new
+=head1 BUGS
- HORIZONTAL TABULATION CHARACTER TABULATION
- VERTICAL TABULATION LINE TABULATION
- FILE SEPARATOR INFORMATION SEPARATOR FOUR
- GROUP SEPARATOR INFORMATION SEPARATOR THREE
- RECORD SEPARATOR INFORMATION SEPARATOR TWO
- UNIT SEPARATOR INFORMATION SEPARATOR ONE
- PARTIAL LINE DOWN PARTIAL LINE FORWARD
- PARTIAL LINE UP PARTIAL LINE BACKWARD
+viacode should return an empty string for unassigned in-range Unicode code
+points, as that is their correct current name.
-but the old names in addition to giving the character
-will also give a warning about being deprecated.
+viacode(0) doesn't return C<NULL>, but C<undef>
-=head1 ILLEGAL CHARACTERS
+vianame returns a chr if the input name is of the form C<U+...>, and an ord
+otherwise. It is planned to change this to always return an ord.
-If you ask by name for a character that does not exist, a warning is
-given and the Unicode I<replacement character> "\x{FFFD}" is returned.
+None of the functions work on almost all the Hangul syllable and CJK Unicode
+characters that have their code points as part of their names.
-If you ask by code for a character that does not exist, no warning is
-given and C<undef> is returned. (Though if you ask for a code point
-past U+10FFFF you do get a warning.)
+Names must be ASCII characters only.
-=head1 BUGS
+Unicode standard named sequences are not recognized, such as
+C<LATIN CAPITAL LETTER A WITH MACRON AND GRAVE>
+(which should mean C<LATIN CAPITAL LETTER A WITH MACRON> with an additional
+C<COMBINING GRAVE ACCENT>).
-Since evaluation of the translation function happens in a middle of
+Since evaluation of the translation function happens in the middle of
compilation (of a string literal), the translation function should not
do any C<eval>s or C<require>s. This restriction should be lifted in
a future version of Perl.