You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
36 lines
1.3 KiB
36 lines
1.3 KiB
#!/usr/bin/perl
|
|
#
|
|
# Split Unicode Data File
|
|
# (c) 1997--2003 Martin Mares <mj@atrey.karlin.mff.cuni.cz>
|
|
#
|
|
|
|
open(I, "unidata/UnicodeData.txt") || die "Unable to open UniCode data file";
|
|
open(C, ">misc/u-cat") || die "cat file open";
|
|
open(U, ">misc/u-upper") || die "upper file open";
|
|
open(L, ">misc/u-lower") || die "lower file open";
|
|
open(G, ">misc/u-ligatures") || die "lig file open";
|
|
while (<I>) {
|
|
chomp;
|
|
(/^$/ || /^#/) && next;
|
|
($code,$name,$cat,$comb,$bidir,$decomp,$d0,$d1,$n0,$mirr,$cmt1,$cmt2,$upper,$lower,$title) = split /;/;
|
|
$code =~ /^....$/ || next;
|
|
if ($cat =~ /^C/) { $ccat = "_U_CTRL"; }
|
|
elsif ($cat =~ /^Z/) { $ccat = "_U_SPACE"; }
|
|
elsif ($decomp =~ /<compat>/ && $name =~ / LIGATURE /) {
|
|
$ccat = "_U_LIGATURE";
|
|
print G "$code\n";
|
|
} elsif ($cat =~ /^Ll/) { $ccat = "_U_LLOWER"; }
|
|
elsif ($cat =~ /^Lu/) { $ccat = "_U_LUPPER"; }
|
|
elsif ($cat =~ /^L/) { $ccat = "_U_LETTER"; }
|
|
elsif ($code ge "0030" && $code le "0039") { $ccat = "_U_DIGIT | _U_XDIGIT"; }
|
|
else { $ccat = ""; }
|
|
if ($code ge "0041" && $code le "0046" || $code ge "0061" && $code le "0066") { $ccat = $ccat . "|_U_XDIGIT"; }
|
|
if ($ccat ne "") { print C "$code\t$ccat\n"; }
|
|
if ($upper ne "") { print U "$code\t0x$upper\n"; }
|
|
if ($lower ne "") { print L "$code\t0x$lower\n"; }
|
|
}
|
|
close I;
|
|
close C;
|
|
close U;
|
|
close L;
|
|
close G;
|
|
|