X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/12d4eef20f4fed251cec97be02abf6b0af9a929b..bf64871d2554e77756e10b896bb63589d15b4f73:/charset.plp diff --git a/charset.plp b/charset.plp index e1efd83..b1c680e 100644 --- a/charset.plp +++ b/charset.plp @@ -32,26 +32,32 @@ sub tabinput { state $ALIAS = { default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], - us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )], - ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )], + us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], + ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775 cp737 cp869 cp866 cp855 cp862 cp864 )], aix => [qw( cp1006 )], win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )], + mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami + MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )], + # languages westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )], centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )], 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 )], # MacUkrainian is broken - arabic => [qw( iso-8859-6 cp1006 cp864 cp1256 MacArabic )], # MacFarsi same as MacArabic? - greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )], - hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 )], + 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 + arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi + U6 U8A-8F+0-31+64 U75-77 )], + greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 U37-3F U1F )], + hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )], thai => [qw( iso-8859-11 cp874 MacThai )], vietnamese => [qw( viscii cp1258 MacVietnamese )], symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )], + # iso-code shorthand 1 => 'westeur', 2 => 'centeur', @@ -79,14 +85,14 @@ sub tabinput { use List::Util qw( first pairfirst pairs ); state $INHERIT = { - 'cp437' => ['cp850' => 0], # ascii range overridden later + 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later 'gsm0338' => ['ascii' => '0-127'], 'iso-8859-2' => ['iso-8859-1' => '160'], - 'iso-8859-3' => ['iso-8859-1' => '160'], + 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9 'iso-8859-4' => ['iso-8859-2' => '160'], 'iso-8859-5' => ['iso-8859-1' => '160'], - 'iso-8859-6' => ['iso-8859-1' => '160'], + 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'], 'iso-8859-7' => ['iso-8859-1' => '160'], 'iso-8859-8' => ['iso-8859-1' => '160'], 'iso-8859-9' => ['iso-8859-1' => '208-223+240'], @@ -119,21 +125,35 @@ sub tabinput { 'cp775' => ['cp850' => '128'], # partial cp437 'cp866' => ['cp437' => '128-175+224'], 'cp855' => ['cp437' => '128'], - 'cp1006' => ['cp437' => '128'], + 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'], 'cp737' => ['cp437' => '128-175+224'], 'cp869' => ['cp437' => '128'], 'cp862' => ['cp437' => '128-159'], - 'cp864' => ['cp437' => '128'], + 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants 'koi8-u' => ['koi8-r' => 128], 'koi8-f' => ['koi8-u' => 128], - 'MacRomanian' => ['MacRoman' => '160'], + 'MacRomanian' => ['MacRoman' => '160-191+208-223'], + 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'], 'MacCroatian' => ['MacRoman' => '160'], 'MacCentralEurRoman' => ['MacRoman' => '128'], + 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], #TODO: gaps at C/E 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned + 'MacSami' => ['MacIcelandic' => '144', 'MacRoman' => '144'], + 'MacGreek' => ['MacRoman' => '128'], 'MacCyrillic' => ['MacRoman' => '128'], - 'MacHebrew' => ['MacRoman' => '128'], + 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii + 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents + 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'], + + 'cp37' => ['posix-bc' => '0'], + 'posix-bc' => ['cp1047' => '64'], + 'cp500' => ['cp37' => '64-95+176-191'], + 'cp1047' => ['cp37' => '16-95+160-191'], #TODO: gap at 3/4 + 'cp1026' => ['cp37' => '64'], + 'cp875' => ['cp37' => '48'], + }; my @parents = @{ $INHERIT->{$input} || [] }; @@ -146,16 +166,21 @@ sub tabinput { } elsif (defined $visible->{ascii}) { $row{parent} = $parents[0]; - $params = $parents[1] || 128; + $params = $parents[1] // 128; $params = 128 if $params >= 128; # ascii offset at most } elsif (@parents) { $row{parent} = $parents[0]; + $params = $parents[1] if $parents[1] == 0; # apply ascii end } - $visible->{$_} //= 0 for $row{parent}; + $visible->{$_} //= 0 for $row{parent} || (); + } + else { + # manual option to double table width + $row{cols} *= 2 if $params =~ s/[+]\z//; } - if (defined $params) { + if (length $params) { $params =~ m{ \A (? \d+) (?: (?: [-] (? \d+) )? (?: [+] (? \d+) ) )? @@ -175,13 +200,14 @@ sub tabinput { } } - if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) { + if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) { my $start = hex($1) << ($2 ? 4 : 8); - my $end = $2 ? hex($2) << 4 : $start + 240; - $row{table} = join '', map { chr } $start .. $end+15; + my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255; + $row{table} = join '', map { chr } $start .. $end; utf8::upgrade($row{table}); # prevent latin1 output - $row{endpoint} = $end + 14 - $start; + $row{endpoint} = $end - $start; $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; + $row{offset} = $start % 256; } elsif ($input eq 'U') { $row{set} = 'Unicode planes'; @@ -204,7 +230,7 @@ sub tabinput { $row{endpoint} = 255; } else { - if ($row{set} eq 'MacHebrew') { + if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') { # array of possibly multiple characters per code point $row{table} = [ map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint