X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/211543b923759291c93630472bae959c6e8fb652..1436e579bd7ccf52d5e53d2db969eb74093df149:/charset.plp?ds=sidebyside diff --git a/charset.plp b/charset.plp index b0d3b83..51c391c 100644 --- a/charset.plp +++ b/charset.plp @@ -15,129 +15,285 @@ Html({ data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'], }); +my @tablist = split m{/+}, $Request || 'default'; + :>
$paramsis 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} - $row{offset} }++; + } } + } - if ($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{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; + if ($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{endpoint} = $end + 14 - $start; + $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; + } + elsif ($input eq 'U') { + $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) * $row{cell}->{colsize}; } - elsif ($input eq 'U') { - $row{table} = ' ' x 1024; - $row{set} = 'Unicode planes'; - $row{cell} = do 'charset-ucplanes.inc.pl' - or printf "
Table data could not be read: %s.
\n", $@ || $!; - @nibble = (map { $_.0, $_.8 } 0 .. 7); - $nibsize = 8; + elsif ($row{set} eq 'utf-8-strict') { + $row{set} = 'UTF-8'; + $row{cell} = do 'charset-utf8.inc.pl' + or Alert('Table data could not be read', $@ || $!); + $row{endpoint} = 255; } - elsif ($row{set} = resolve_alias($input)) { - if ($row{set} eq 'Internal') { - $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192); - $row{set} = 'Unicode BMP'; - $row{cell} = do 'charset-unicode.inc.pl' - or printf "Table data could not be read: %s.
\n", $@ || $!; - } - elsif ($row{set} eq 'utf-8-strict') { - $row{table} = undef; - $row{set} = 'UTF-8'; - $row{cell} = do 'charset-utf8.inc.pl' - or printf "Table data could not be read: %s.
\n", $@ || $!; + else { + if ($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 + ]; } else { - $row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint); + # ~16x faster than decoding in loop; + # substr strings is twice as fast as splitting to an array + $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $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 + ); + } + + $visible->{ascii} = # assume common base + $visible->{ $row{set} } = 1; } - else { - print "Encoding $input unknown
\n"; - } - \%row; } else { - (); + Alert("Encoding$inputunknown"); + return; } -} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ } - $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default'; -my $NOCHAR = chr 0xFFFD; - -for my $cp437 (grep {$request[$_]->{set} eq 'cp437'} 0 .. $#request) { - substr($request[$cp437]->{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign - substr($request[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw( - 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C - 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC - ); + push @request, \%row; } +tabinput($_) for @tablist; + +my $NOCHAR = chr 0xFFFD; sub range_cell { - my ($table, $offset) = @_; + my ($info, $offset) = @_; + my $table = $info->{cell} or return; my $def = $table->{$offset} or return; my ($len, $class, $name, $title) = @{$def}; + my $cols = $info->{cols}; + my $colsize = $table->{colsize} || 1; my $attr = ''; - $len /= $nibsize; + $len /= $colsize; $name //= $len <= 2 ? 'res' : 'reserved'; - if (my $part = $offset/$nibsize % 16) { + if (my $part = $offset/$colsize % $cols) { # continued row - my $cols = 16 - $part; # remaining - $cols = $len if $len < $cols; #TODO: optimise - if ($len -= $cols) { + my $rest = $cols - $part; # remaining + $rest = $len if $len < $rest; #TODO: optimise + if ($len -= $rest) { # continued on new row - $table->{$offset + $nibsize*$cols} = [$len*$nibsize, "$class joinu", $name, $title]; - $name = ''; + my @next = ($len * $colsize, "$class joinu"); + if ($len > $rest) { + # minority remains + push @next, $name, $title; + $title ||= $name; + $name = ''; + } + else { + # minority on next row + push @next, '"', $title || $name; + } + $table->{$offset + $colsize*$rest} //= \@next; $class .= ' joind'; } - $len = $cols; + $len = $rest; } - elsif (my $rows = $len >> 4) { + elsif (my $rows = int($len / $cols)) { # multiple full rows - if ($len -= $rows << 4) { + if ($len -= $rows * $cols) { # partial row remains - $table->{$offset + $nibsize*$rows * 16} = [$len*$nibsize, "$class joinu", '', $title]; + $table->{$offset + $colsize*$rows * $cols} //= [$len*$colsize, "$class joinu", '', $title]; $class .= ' joind'; } $attr .= sprintf ' rowspan=%d', $rows; - $len = 16; + $len = $cols; } $attr .= sprintf ' colspan=%d', $len unless $len == 1; @@ -148,48 +304,90 @@ sub range_cell { } for my $row (@request) { + my $cols = $row->{cols}; + my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1; + 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; + printf '
â±"; - print ' | ', $_ for @nibble; + print "<$section> | |
---|---|---|
", $rowdiv == 1 ? '+' : 'â±'; + printf ' | %0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1; print "\n"; } print ' | |
%X', ($msb + ($row->{offset} >> 4)) * $nibsize; - for my $lsb (0 .. $#nibble) { - my $val = ( ($msb<<4) + $lsb ) * $nibsize; + while ($offset < $row->{endpoint}) { + if ($row->{skip}->{$offset}) { + $offset += $cols * $colsize; + next; + } + + print ' | ||
'; + { + if (my $rowmod = $offset % $rowdiv) { + # offset in column units + printf '+%X', $rowmod; + } + else { + # divided row offset + printf '%X', ($offset + $row->{offset}) / $rowdiv; + } + } + for (1 .. $cols) { if ($row->{cell}) { - if (ref $row->{cell} eq 'CODE') { - print $row->{cell}->($val); - next; - } - print range_cell($row->{cell}, $val); + print range_cell($row, $offset); next; } - my $glyph = substr $row->{table}, $val, 1; + my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] : + substr $row->{table}, $offset, 1; if ($glyph eq $NOCHAR) { print ' | '; next; } + if (exists $get{compare}) { + state $visible = {}; + my $cp = $offset + $row->{offset}; + printf ' | %2$s', + $cp == ord $glyph ? 'l4' : + $row->{parent} && $glyph eq + Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' : + $visible->{$glyph} ? 'l2' : + 'l1', + $glyphs->glyph_html($glyph); + $visible->{$glyph}++; + next; + } + print "\n".$glyphs->glyph_cell($glyph); } + continue { + $offset += $colsize; + } print "\n"; } - print " |
unicode + | inherited + | existing + | original + | unassigned +<: } else { :> | control | whitespace | diacritic
|