X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/10b15952ffe8cb59e1adf705b596f1d9a194c979..8e8998a5c2e51159db372eed7e4e293df83ccc91:/charset.plp diff --git a/charset.plp b/charset.plp index e725c11..1b672e9 100644 --- a/charset.plp +++ b/charset.plp @@ -12,7 +12,10 @@ Html({ reference common overview table '], stylesheet => [qw'light'], - data => [qw'charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl'], + data => [qw( + charset-encoding.inc.pl + charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl + )], }); my @tablist = split m{/+}, $Request || 'default'; @@ -73,6 +76,9 @@ use Shiar_Sheet::FormatChar; my $glyphs = Shiar_Sheet::FormatChar->new; my @request; +my $charsets = do 'charset-encoding.inc.pl' + or Alert('Encoding metadata could not be read', $@ || $!); + sub tabinput { # generate character table(s) my $input = shift or return; @@ -98,11 +104,11 @@ 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 )], @@ -130,89 +136,8 @@ sub tabinput { my $params = $input =~ s/[+](.*)\z// ? $1 : undef; if (not defined $params) { - state $INHERIT = { - 'ascii' => ['' => '00-7F'], - 'cp437' => ['cp850' => 0, 'ascii' => '00-1F+80'], # ascii range overridden later - 'gsm0338' => ['ascii' => '00-7F'], - 'dingbats' => ['' => '20-7F+A0'], - 'MacDingbats' => ['dingbats' => '80-9F'], - 'AdobeZdingbat'=> ['MacDingbats' => '80-9F'], # should be identical but maps to private use - 'symbol' => ['' => '20-7F+A0'], - 'AdobeSymbol' => ['symbol' => '20-7F+A0', '' => '20-7F+A0'], # minor differences, irrelevant except for different '€' - 'wingdings' => ['' => '20'], - 'wingdings2' => ['' => '20'], - 'wingdings3' => ['' => '20'], - 'webdings' => ['' => '20'], - - 'iso-8859-2' => ['iso-8859-1' => 'A0'], - 'iso-8859-3' => ['iso-8859-1' => 'A0'], #TODO: also apply to iso-8859-9 - 'iso-8859-4' => ['iso-8859-2' => 'A0'], - 'iso-8859-5' => ['iso-8859-1' => 'A0'], - 'iso-8859-6' => ['cp1256' => '80', 'iso-8859-1' => 'A0'], - 'iso-8859-7' => ['iso-8859-1' => 'A0'], - 'iso-8859-8' => ['iso-8859-1' => 'A0'], - 'iso-8859-9' => ['iso-8859-1' => 'D0-DF+F0'], - 'iso-8859-10' => ['iso-8859-4' => 'A0'], - 'iso-8859-11' => ['iso-8859-1' => 'A0'], - 'iso-8859-13' => ['iso-8859-4' => 'A0'], - 'iso-8859-14' => ['iso-8859-1' => 'A0'], - 'iso-8859-15' => ['iso-8859-1' => 'A0-BF'], - 'iso-8859-16' => ['iso-8859-2' => 'A0'], - 'hp-roman8' => ['iso-8859-1' => 'A0'], - - 'cp1252' => ['iso-8859-1' => '80-9F'], - 'cp1250' => ['iso-8859-2' => '80-BF', 'cp1252' => '80'], - 'cp1254' => ['iso-8859-9' => '80-9F', 'cp1252' => '80-9F+D0'], - 'cp874' => ['iso-8859-11' => '80-9F', 'cp1252' => '80'], # windows-874 actually cp1162 - 'cp1257' => ['iso-8859-13' => '80-9F+FF', 'cp1252' => '80'], - 'cp1251' => ['cp1252' => '80'], - 'cp1253' => ['cp1252' => '80'], - 'cp1255' => ['iso-8859-8' => '80-DF', 'cp1252' => '80'], - 'cp1256' => ['cp1252' => '80'], - 'cp1258' => ['cp1252' => '80-9F+C0'], - - 'cp850' => ['cp437' => '90'], - 'cp860' => ['cp437' => '80-AF'], - 'cp861' => ['cp865' => '80-AF'], - 'cp863' => ['cp437' => '80-AF'], - 'cp865' => ['cp437' => '90-AF'], - 'cp852' => ['cp850' => '80', 'cp437' => '80'], - 'cp857' => ['cp850' => '80-AF+D0-EF', 'cp437' => '80'], - 'cp775' => ['cp850' => '80'], # partial cp437 - 'cp866' => ['cp437' => '80-AF+E0'], - 'cp855' => ['cp437' => '80'], - 'cp1006' => ['iso-8859-6' => 'A0', 'cp437' => '80'], - 'cp737' => ['cp437' => '80-AF+E0'], - 'cp869' => ['cp437' => '80'], - 'cp862' => ['cp437' => '80-9F'], - 'cp864' => ['MacArabic' => '80', 'iso-8859-6' => '80', 'cp437' => '80'], #TODO: compare form variants - - 'koi8-u' => ['koi8-r' => '90-BF'], - 'koi8-f' => ['koi8-u' => '90-BF'], - - 'MacRomanian' => ['MacRoman' => 'A0-BF+D0-DF'], - 'MacRumanian' => ['MacRomanian' => 'A0-BF+D0-DF', 'MacRoman' => 'A0-BF+D0-DF'], - 'MacCroatian' => ['MacRoman' => 'A0'], - 'MacCentralEurRoman' => ['MacRoman' => '80'], - 'MacIcelandic'=> ['MacRoman' => 'A0-AF+D0-EF'], - 'MacTurkish' => ['MacRoman' => 'D0-DF'], # F5 is unassigned - 'MacSami' => ['MacIcelandic' => '90-BF+D0-DF+F0', 'MacRoman' => '90'], - 'MacGreek' => ['MacRoman' => '80'], - 'MacCyrillic' => ['MacRoman' => '80'], - 'MacHebrew' => ['iso-8859-8' => '80', 'MacRoman' => '80-8F+A0'], # partial ascii - 'MacArabic' => ['iso-8859-6' => '80', 'cp864' => '80', 'MacRoman' => '80'], #TODO: multiple parents - 'MacFarsi' => ['MacArabic' => 'B0-BF', 'MacRoman' => '80'], - - 'cp37' => ['posix-bc' => '00'], - 'posix-bc' => ['cp1047' => '40'], - 'cp500' => ['cp37' => '40-5F+B0-BF'], - 'cp1047' => ['cp37' => '10-2F+50-5F+A0-BF'], - 'cp1026' => ['cp37' => '40'], - 'cp875' => ['cp37' => '30'], - - }; - - my @parents = @{ $INHERIT->{$input} || [] }; + my $charset = $charsets->{$input} || {}; + my @parents = @{ $charset->{inherit} || [] }; if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) { $row{parent} = $parent; @@ -223,7 +148,7 @@ sub tabinput { elsif (defined $visible->{ascii}) { $row{parent} = $parents[0]; $params = $parents[1] // 80; - $params = 80 if hex $params >= 128; # ascii offset at most + $params = 80 if hex $params >= 0x80; # ascii offset at most } elsif (@parents) { $row{parent} = $parents[0]; @@ -261,27 +186,34 @@ sub tabinput { eval "require Encode::\u$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{endpoint} ||= 0x3FF; + $row{set} = 'Unicode planes'; } elsif (lc $input eq 'u') { $row{cell} = do 'charset-unicode.inc.pl' or Alert('Table data could not be read', $@ || $!); - $row{endpoint} ||= 8191; + $row{endpoint} ||= 0x1FFF; $row{set} = 'Unicode ' . ( $row{startpoint} < 0x1000 && $row{endpoint} < 0x1000 ? 'BMP' : $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' : @@ -292,11 +224,11 @@ sub tabinput { $row{set} = 'UTF-8'; $row{cell} = do 'charset-utf8.inc.pl' or Alert('Table data could not be read', $@ || $!); - $row{endpoint} = 255; + $row{endpoint} = 0xFF; } elsif ($row{set} = Encode::resolve_alias($input)) { $row{offset} = delete $row{startpoint}; - $row{endpoint} ||= 255; + $row{endpoint} ||= 0xFF; if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') { # array of possibly multiple characters per code point $row{table} = [