X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/55a1d332e52bfd24519ca9c0f0a479f5161cca4c..006a5267c67ca7076380d40d4562555fb5578514:/charset.plp diff --git a/charset.plp b/charset.plp index 82f37f9..7e115a8 100644 --- a/charset.plp +++ b/charset.plp @@ -78,8 +78,8 @@ sub tabinput { my $input = shift or return; state $ALIAS = { - default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], - unicode => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )], + default => [qw( u+0-27F utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + unicode => [qw( uu+cols=32+realsize u+0-FFF u+1000-18AF u+18B0-1FFF )], us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], @@ -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+0-31+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 )], + +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 )], @@ -131,83 +131,84 @@ sub tabinput { if (not defined $params) { state $INHERIT = { - 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later - 'gsm0338' => ['ascii' => '0-127'], - 'dingbats' => ['' => '32-127+160'], - 'MacDingbats' => ['dingbats' => '128-159'], - 'AdobeZdingbat'=> ['MacDingbats' => '128-159'], # should be identical but maps to private use - 'symbol' => ['' => '32-127+160'], - 'AdobeSymbol' => ['symbol' => '32-127+160', '' => '32-127+160'], # minor differences, irrelevant except for different '€' - 'wingdings' => ['' => '32'], - 'wingdings2' => ['' => '32'], - 'wingdings3' => ['' => '32'], - 'webdings' => ['' => '32'], - - 'iso-8859-2' => ['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' => ['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'], - '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', 'cp1252' => '128'], - 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'], - 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162 - 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'], - 'cp1251' => ['cp1252' => '128'], - 'cp1253' => ['cp1252' => '128'], - 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'], - 'cp1256' => ['cp1252' => '128'], - 'cp1258' => ['cp1252' => '128-159+192'], - - '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' => '128-175+208-239', 'cp437' => '128'], - 'cp775' => ['cp850' => '128'], # partial cp437 - 'cp866' => ['cp437' => '128-175+224'], - 'cp855' => ['cp437' => '128'], - 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'], - 'cp737' => ['cp437' => '128-175+224'], - 'cp869' => ['cp437' => '128'], - 'cp862' => ['cp437' => '128-159'], - 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants - - 'koi8-u' => ['koi8-r' => '144-191'], - 'koi8-f' => ['koi8-u' => '144-191'], - - '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'], - 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned - 'MacSami' => ['MacIcelandic' => '144-191+208-223+240', 'MacRoman' => '144'], - 'MacGreek' => ['MacRoman' => '128'], - 'MacCyrillic' => ['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' => '16-47+80-95+160-191'], - 'cp1026' => ['cp37' => '64'], - 'cp875' => ['cp37' => '48'], + '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'], }; @@ -216,17 +217,17 @@ sub tabinput { if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) { $row{parent} = $parent; $params = $part; - $params = 128 unless $visible->{$parent} + $params = 80 unless $visible->{$parent} or ($input eq 'MacCroatian' and defined $visible->{MacRomanian}); } elsif (defined $visible->{ascii}) { $row{parent} = $parents[0]; - $params = $parents[1] // 128; - $params = 128 if $params >= 128; # ascii offset at most + $params = $parents[1] // 80; + $params = 80 if hex $params >= 0x80; # ascii offset at most } elsif (@parents) { $row{parent} = $parents[0]; - $params = $parents[1] if $parents[1] == 0; # apply ascii end + $params = $parents[1] if hex $parents[1] == 0; # apply ascii end } $visible->{$_} //= 0 for $row{parent} || (); } @@ -238,18 +239,18 @@ sub tabinput { elsif ($param =~ m{ \A cols = (\d+) \z }x) { $row{cols} = $1; } - elsif ($param =~ m{ \A (? \d+) (?: [-] (? \d+) )? \z }x) { + elsif ($param =~ m{ \A (? \p{AHex}+) (?: [-] (? \p{AHex}+) )? \z }x) { if (defined $row{endpoint}) { # extend earlier range my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols}); - for ($skip + 1 .. ($+{start} / $row{cols}) - 1) { + for ($skip + 1 .. (hex($+{start}) / $row{cols}) - 1) { $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++; } } else { - $row{startpoint} = $+{start}; + $row{startpoint} = hex $+{start}; } - $row{endpoint} = $+{end} || 0; + $row{endpoint} = hex($+{end} || 0); } else { Alert("Unknown option $param for charset $input"); @@ -260,32 +261,37 @@ sub tabinput { eval "require Encode::\u$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{cell}->{colsize}; + $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} *= $row{cell}->{colsize}; - $row{startpoint} *= $row{cell}->{colsize}; + $row{endpoint} ||= 0x1FFF; $row{set} = 'Unicode ' . ( - $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' : - $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' : + $row{startpoint} < 0x1000 && $row{endpoint} < 0x1000 ? 'BMP' : + $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' : 'allocations' ); } @@ -293,11 +299,11 @@ 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} = [ @@ -364,7 +370,7 @@ sub range_cell { $len /= $colsize; $name //= $len <= 2 ? 'res' : 'reserved'; - if (my $part = ($offset - $info->{startpoint})/$colsize % $cols) { + if (my $part = ($offset/$colsize - $info->{startpoint}) % $cols) { # continued row my $rest = $cols - $part; # remaining $rest = $len if $len < $rest; #TODO: optimise @@ -424,7 +430,7 @@ for my $row (@request) { my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header my $rowdiv = 16 ** $coldigits; # row divide for column digits $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched - my $offset = $row->{startpoint} || 0; + my $offset = $row->{startpoint} * $colsize || 0; printf '
', !$row->{cell} && ' charmap'; my $title = $row->{set}; @@ -437,8 +443,9 @@ for my $row (@request) { printf ''; - while ($offset < $row->{endpoint}) { + while ($offset <= $row->{endpoint} * $colsize) { if ($row->{skip}->{$offset}) { $offset += $cols * $colsize; next;
%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1; print "\n"; } + print '