X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/d5ddd90537e6910cf1b2fd1c6181b98b59d39bd0..414f2bc754363a3052954e45a92c76db6bfc9c06:/charset.plp diff --git a/charset.plp b/charset.plp index f26b7bd..5e50004 100644 --- a/charset.plp +++ b/charset.plp @@ -61,6 +61,7 @@ print join " •\n", ( cyrillic => 0, hebrew => 0, ], + [ uc => 'Unicode' ], ); :>.

@@ -76,7 +77,8 @@ sub tabinput { my $input = shift or return; state $ALIAS = { - default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + default => [qw( unicode+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + uc => [qw( U+cols=32 unicode+0-4095 unicode+4096-6319 unicode+6320-8191 )], us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], @@ -124,7 +126,6 @@ sub tabinput { 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) { @@ -175,16 +176,16 @@ sub tabinput { 'cp862' => ['cp437' => '128-159'], 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants - 'koi8-u' => ['koi8-r' => 128], - 'koi8-f' => ['koi8-u' => 128], + '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'], #TODO: gaps at C/E + 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned - 'MacSami' => ['MacIcelandic' => '144', 'MacRoman' => '144'], + '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 @@ -194,7 +195,7 @@ sub tabinput { '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 + 'cp1047' => ['cp37' => '16-47+80-95+160-191'], 'cp1026' => ['cp37' => '64'], 'cp875' => ['cp37' => '48'], @@ -219,28 +220,29 @@ sub tabinput { } $visible->{$_} //= 0 for $row{parent} || (); } - else { - # manual option to double table width - $row{cols} *= 2 if $params =~ s/[+]\z//; - } - if (length $params) { - $params =~ m{ - \A (? \d+) - (?: (?: [-] (? \d+) )? (?: [+] (? \d+) ) )? - (?: [-] (? \d+) )? \z - }x or Alert([ - "Unknown range parameters for $input", - "$params is not in format start(-stop)(+restart(-end))", - ]); - - $row{offset} = $+{offset}; - $endpoint = $+{endpoint} if $+{endpoint}; - if (my $restart = $+{restart}) { - my $skip = int(($+{stop} || $row{offset}) / $row{cols}); - for ($skip + 1 .. ($restart / $row{cols}) - 1) { - $row{skip}->{ $_ * $row{cols} - $row{offset} }++; + for my $param (split /[+]+/, $params // '') { + if ($param eq 'realsize') { + $row{realsize}++; + } + elsif ($param =~ m{ \A cols = (\d+) \z }x) { + $row{cols} = $1; + } + elsif ($param =~ m{ \A (? \d+) (?: [-] (? \d+) )? \z }x) { + if (defined $row{endpoint}) { + # extend earlier range + my $skip = int(($row{endpoint} || $row{offset}) / $row{cols}); + for ($skip + 1 .. ($+{offset} / $row{cols}) - 1) { + $row{skip}->{ $_ * $row{cols} - $row{offset} }++; + } + } + else { + $row{offset} = $+{offset}; } + $row{endpoint} = $+{endpoint} || 0; + } + else { + Alert("Unknown option $param for charset $input"); } } @@ -257,15 +259,22 @@ sub tabinput { $row{set} = 'Unicode planes'; $row{cell} = do 'charset-ucplanes.inc.pl' or Alert('Table data could not be read', $@ || $!); - $row{cols} *= 2; $row{endpoint} = 1023 * $row{cell}->{colsize}; } elsif ($row{set} = Encode::resolve_alias($input)) { if ($row{set} eq 'Internal') { - $row{set} = 'Unicode BMP'; $row{cell} = do 'charset-unicode.inc.pl' or Alert('Table data could not be read', $@ || $!); - $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize}; + + $row{endpoint} ||= 8191; + $row{endpoint} *= $row{cell}->{colsize}; + $row{startpoint} = $row{cell}->{colsize} * $row{offset}; + $row{offset} = 0; + $row{set} = 'Unicode ' . ( + $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' : + $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' : + 'allocations' + ); } elsif ($row{set} eq 'utf-8-strict') { $row{set} = 'UTF-8'; @@ -274,27 +283,35 @@ sub tabinput { $row{endpoint} = 255; } else { + $row{endpoint} ||= 255; if ($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} .. $endpoint + map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{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); + $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint}); } - $row{endpoint} = $endpoint - $row{offset}; - - if ($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 - 25D8 25CB 25D9 2642 2640 266A 266B 263C - 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 - 2191 2193 2192 2190 221F 2194 25B2 25BC - ); + $row{endpoint} -= $row{offset}; + + if ($row{set} eq 'cp437') { + for my $phipos (237 - $row{offset}) { + next if $phipos < 0 or $phipos > $row{endpoint}; + # replace phi glyph + substr($row{table}, $phipos, 1) = pack 'U*', 0x3D5; + } + if ($row{offset} == 0) { + # replace control characters by visible variants + substr($row{table}, 0, 32) = pack 'U*', map {hex} qw( + 2007 263A 263B 2665 2666 2663 2660 2022 + 25D8 25CB 25D9 2642 2640 266A 266B 263C + 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 + 2191 2193 2192 2190 221F 2194 25B2 25BC + ); + } } $visible->{ascii} = # assume common base @@ -354,13 +371,15 @@ sub range_cell { $class .= ' joind'; } - # coalesce multiple rows - while ($rows > 3) { - $info->{skip}->{$offset += $rowsize}++; - $rows--; - } - if ($rows > 2) { - $info->{skip}->{$offset += $rowsize} = 0; + unless ($info->{realsize}) { + # coalesce multiple rows + while ($rows > 3) { + $info->{skip}->{$offset += $rowsize}++; + $rows--; + } + if ($rows > 2) { + $info->{skip}->{$offset += $rowsize} = 0; + } } $attr .= sprintf ' rowspan=%d', $rows; @@ -380,7 +399,7 @@ 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 = 0; + my $offset = $row->{startpoint} || 0; printf '
', !$row->{cell} && ' charmap'; my $title = $row->{set};