X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/f5b22bf2d6aef261464440bd548285b52908be3d..1526acc93e4a99f79545a62d5700d9020fe59b5b:/charset.plp diff --git a/charset.plp b/charset.plp index 8b42e62..309139e 100644 --- a/charset.plp +++ b/charset.plp @@ -31,19 +31,25 @@ sub tabinput { my $input = shift or return; state $ALIAS = { - default => [qw( unicode+0-192 utf-8 iso-8859-1 cp1252+128-159 iso-8859-15+160-191 cp437 cp850+128 )], - us => [qw( cp437 cp863+128 AdobeStandardEncoding gsm0338+0-127 )], - ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )], - westeur => [qw( iso-8859-1 iso-8859-15+160-191 cp1252+128-159 iso-8859-14+160 cp850+128 MacRoman+128 nextstep+128 hp-roman8+160 )], - centeur => [qw( iso-8859-2 iso-8859-16+160 cp1250+128 cp852+128 MacCentralEurRoman+128 MacCroatian+128 MacRomanian+128 )], # MacRumanian only for DB - turkish => [qw( iso-8859-3 iso-8859-9+128 cp857+128 cp1254+128 MacTurkish+128 )], - baltic => [qw( iso-8859-4 iso-8859-13+160 cp775+128 cp1257+128 )], - nordic => [qw( iso-8859-10 cp865+128 cp861+160-191 MacIcelandic+128 MacSami+160 )], - cyrillic => [qw( koi8-f koi8-r+128-192 koi8-u+160-192 iso-8859-5+128 cp1251+128 MacCyrillic+128 cp855+128 cp866+128 )], # MacUkrainian is broken - arabic => [qw( iso-8859-6 cp1006+160 cp864+128 cp1256+128 MacArabic )], # MacFarsi same as MacArabic? - greek => [qw( iso-8859-7 cp1253+128 MacGreek+128 cp737+128 cp869+128 )], - hebrew => [qw( iso-8859-8 cp1255+128 MacHebrew+128 cp862+128 )], - thai => [qw( iso-8859-11 cp874+128-159 MacThai+128 )], + default => [qw( unicode+0-192 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )], + ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )], + iso => [map {"iso-8859-$_"} 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 )], + + 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 )], + thai => [qw( iso-8859-11 cp874 MacThai )], vietnamese => [qw( viscii cp1258 MacVietnamese )], symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )], # iso-code shorthand @@ -64,28 +70,102 @@ sub tabinput { return; } + state $visible = {}; # all present tables my %row = (offset => 0, cols => 16); my $endpoint = 255; my $params = $input =~ s/[+](.*)\z// ? $1 : undef; + if (not defined $params) { + use List::Util qw( first pairfirst pairs ); + + state $INHERIT = { + 'cp437' => ['cp850' => 0], # ascii range overridden later + 'gsm0338' => ['ascii' => '0-127'], + + 'iso-8859-2' => ['iso-8859-1' => '160'], + 'iso-8859-3' => ['iso-8859-1' => '160'], + 'iso-8859-4' => ['iso-8859-2' => '160'], + 'iso-8859-5' => ['iso-8859-1' => '160'], + 'iso-8859-6' => ['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'], + 'iso-8859-10' => ['iso-8859-4' => '160'], + 'iso-8859-11' => ['iso-8859-1' => '160'], + 'iso-8859-13' => ['iso-8859-4' => '160'], + 'iso-8859-14' => ['iso-8859-1' => '160'], + 'iso-8859-15' => ['iso-8859-1' => '160-191'], + 'iso-8859-16' => ['iso-8859-2' => '160'], + 'hp-roman8' => ['iso-8859-1' => '160'], + + 'cp1252' => ['iso-8859-1' => '128-159'], + 'cp1250' => ['iso-8859-2' => '128-191'], + 'cp1254' => ['iso-8859-9' => '128-159'], + 'cp874' => ['iso-8859-11' => '128-159'], # windows-874 actually cp1162 + 'cp1257' => ['iso-8859-13' => '128'], + 'cp1251' => ['cp1252' => '128'], + 'cp1253' => ['cp1252' => '128'], + 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'], + 'cp1256' => ['cp1252' => '128'], + + 'cp850' => ['cp437' => '144'], + 'cp860' => ['cp437' => '128-175'], + 'cp861' => ['cp865' => '128-175'], + 'cp863' => ['cp437' => '128-175'], + 'cp865' => ['cp437' => '144-175'], + 'cp852' => ['cp850' => '128', 'cp437' => '128'], + 'cp857' => ['cp850' => '128-239', 'cp437' => '128'], + 'cp775' => ['cp850' => '128'], # partial cp437 + 'cp866' => ['cp437' => '128'], + 'cp855' => ['cp437' => '128'], + 'cp1006' => ['cp437' => '128'], + 'cp737' => ['cp437' => '128'], + 'cp869' => ['cp437' => '128'], + 'cp862' => ['cp437' => '128'], + 'cp864' => ['cp437' => '128'], + + 'koi8-u' => ['koi8-r' => 128], + 'koi8-f' => ['koi8-u' => 128], + + 'MacRomanian' => ['MacRoman' => '160'], + 'MacCroatian' => ['MacRoman' => '160'], + 'MacCentralEurRoman' => ['MacRoman' => '128'], + 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned + 'MacCyrillic' => ['MacRoman' => '128'], + 'MacHebrew' => ['MacRoman' => '128'], + }; + + my @parents = @{ $INHERIT->{$input} || [] }; + + if (my ($parent, $part) = pairfirst { $visible->{$a} } @parents) { + $row{parent} = $parent; + $params = $part; + } + elsif ($visible->{ascii}) { + $row{parent} = $parents[0] || 'ascii'; + $params = $parents[1] || 128; + $params = 128 if $params >= 128; # ascii offset at most + } + elsif (@parents) { + $row{parent} = $parents[0]; + } + } + if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) { $row{offset} = $1; $endpoint = -$2 if $2; } - if ($row{offset}) { - $row{setnote} = 'over cp437' if $input eq 'cp850'; - $row{setnote} = 'over iso-8859-1' if $input =~ /^iso-8859-|^cp125/; - } if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) { my $start = hex($1) << ($2 ? 4 : 8); my $end = $2 ? hex($2) << 4 : $start + 240; $row{table} = join '', map { chr } $start .. $end+15; utf8::upgrade($row{table}); # prevent latin1 output + $row{len} = length $row{table}; $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; } elsif ($input eq 'U') { - $row{table} = ' ' x 1024; + $row{len} = 1024; $row{set} = 'Unicode planes'; $row{cell} = do 'charset-ucplanes.inc.pl' or Alert('Table data could not be read', $@ || $!); @@ -93,23 +173,32 @@ sub tabinput { } elsif ($row{set} = Encode::resolve_alias($input)) { if ($row{set} eq 'Internal') { - $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192); + $row{len} = $endpoint < 255 ? 640 : 8192; $row{set} = 'Unicode BMP'; $row{cell} = do 'charset-unicode.inc.pl' or Alert('Table data could not be read', $@ || $!); } elsif ($row{set} eq 'utf-8-strict') { - $row{table} = undef; + $row{len} = 256; $row{set} = 'UTF-8'; $row{cell} = do 'charset-utf8.inc.pl' or Alert('Table data could not be read', $@ || $!); } else { - $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint); - # (~16x faster than decoding in loop; - # substr strings is twice as fast as splitting to an array) + if ($row{set} eq 'MacHebrew') { + # array of possibly multiple characters per code point + $row{table} = [ + map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $endpoint + ]; + } + else { + # ~16x faster than decoding in loop; + # substr strings is twice as fast as splitting to an array + $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint); + } + $row{len} = $endpoint - $row{offset}; - if ($row{set} eq 'cp437') { + if ($row{set} eq 'cp437' and !$row{offset}) { substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign substr($row{table}, 0, 32) = pack 'U*', map {hex} qw( 2007 263A 263B 2665 2666 2663 2660 2022 @@ -118,6 +207,9 @@ sub tabinput { 2191 2193 2192 2190 221F 2194 25B2 25BC ); } + + $visible->{ascii} = # assume common base + $visible->{ $row{set} } = 1; } } else { @@ -189,11 +281,12 @@ for my $row (@request) { my $rowdiv = 16 ** $coldigits; # row divide for column digits $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched my $offset = 0; - my $endpoint = $offset + (length($row->{table}) || 256) * $colsize; + my $endpoint = $offset + $row->{len} * $colsize; printf '
', !$row->{cell} && ' charmap'; my $title = $row->{set}; - $title .= " " for $row->{setnote} // (); + $title .= " " + for grep { $_ ne 'iso-8859-1' } $row->{parent} // (); printf '', $title; print '' x ($cols + 1); for my $section (qw{thead}) { @@ -220,7 +313,8 @@ for my $row (@request) { next; } - my $glyph = substr $row->{table}, $offset, 1; + my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] : + substr $row->{table}, $offset, 1; if ($glyph eq $NOCHAR) { print '
%s
'; next;