From: Mischa POSLAWSKY <perl@shiar.org>
Date: Fri, 21 Apr 2017 22:06:59 +0000 (+0200)
Subject: charset: rename u* preset for custom unicode requests
X-Git-Tag: v1.10~102
X-Git-Url: http://git.shiar.nl/sheet.git/commitdiff_plain/36f3ac1cd686b98270b6546da7a2ed399336354e

charset: rename u* preset for custom unicode requests

Replace special uHEX match by empty request with common range parameters.
---

diff --git a/charset.plp b/charset.plp
index e725c11..b84816a 100644
--- a/charset.plp
+++ b/charset.plp
@@ -98,11 +98,11 @@ sub tabinput {
 		baltic     => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
 		nordic     => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
 		cyrillic   => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855
-		                   U4 U2DE-2DF UA64-A69 U50-52 )], # MacUkrainian is broken
+		                   +400 +2DE0 +A640-A69F +500-52F )], # MacUkrainian is broken
 		arabic     => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
-		                   U6 U8A-8F+0-1F+40 U75-77 )],
-		greek      => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869  U37-3F U1F )],
-		hebrew     => [qw( iso-8859-8 cp1255 MacHebrew cp862  U59-5F )],
+		                   +600 +8A0-8BF+8E0 +750-77F )],
+		greek      => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869  +370 +1F00 )],
+		hebrew     => [qw( iso-8859-8 cp1255 MacHebrew cp862  +590 )],
 		thai       => [qw( iso-8859-11 cp874 MacThai )],
 		vietnamese => [qw( viscii cp1258 MacVietnamese )],
 		symbols    => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )],
@@ -261,21 +261,28 @@ sub tabinput {
 		eval "require Encode::\u$input";
 	}
 
-	if ($input =~ m{ \A u ([0-9a-f]+) (?:-([0-9a-f]+))? \z }ix) {
-		my $start = hex($1) << ($2 ? 4 : 8);
-		my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255;
-		$row{table} = join '', map { chr } $start .. $end;
+	if ($input eq '') {
+		$row{offset} = delete $row{startpoint};
+		$row{set} = 'Unicode characters';
+		my $block = $row{offset} >> 8;
+		$row{endpoint} ||= ($block + 1 << 8) - 1;
+		$block == ($row{endpoint} >> 8) or undef $block;
+
+		$row{table} = join '', map { chr } $row{offset} .. $row{endpoint};
 		utf8::upgrade($row{table});  # prevent latin1 output
-		$row{endpoint} = $end - $start;
-		$row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
-		$row{offset} = $start % 256;
-		$row{startpoint} = 0;
+
+		$row{endpoint} -= $row{offset};
+
+		if (defined $block) {
+			$row{set} = sprintf 'Unicode block U+%02Xxx', $block;
+			$row{offset} %= 0x100;
+		}
 	}
 	elsif (lc $input eq 'uu') {
-		$row{set} = 'Unicode planes';
 		$row{cell} = do 'charset-ucplanes.inc.pl'
 			or Alert('Table data could not be read', $@ || $!);
 		$row{endpoint} ||= 1023;
+		$row{set} = 'Unicode planes';
 	}
 	elsif (lc $input eq 'u') {
 		$row{cell} = do 'charset-unicode.inc.pl'