X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/64823813136c216d4bfcf7df302983ffe408b5f7..b1db17617e16cb9468808f7e2aea0ad8c96c166e:/charset.plp diff --git a/charset.plp b/charset.plp index 54e3b45..05336c1 100644 --- a/charset.plp +++ b/charset.plp @@ -12,7 +12,7 @@ Html({ reference common overview table '], stylesheet => [qw'light'], - data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'], + data => [qw'charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl'], }); my @tablist = split m{/+}, $Request || 'default'; @@ -43,14 +43,16 @@ sub optionlink { print join " •\n", ( map { join " ·\n", pairmap { - optionlink($b || ucfirst $a, "/charset/$a?compare", $a eq $Request); + optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request); } @{$_} } [ iso => 'ISO', + win => 'Windows', dos => 'DOS', mac => 'Apple', ebcdic => 'EBCDIC', + $tablist[0] eq 'default' ? () : ('' => 'common'), ], [ westeur => 'West', @@ -61,7 +63,6 @@ print join " •\n", ( cyrillic => 0, hebrew => 0, ], - [ uc => 'Unicode' ], ); :>.

@@ -77,8 +78,8 @@ sub tabinput { my $input = shift or return; state $ALIAS = { - default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], - uc => [qw( U++ unicode+0-4095 unicode+4096-6319 unicode+6320-8191 )], + default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + unicode => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )], us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], @@ -104,7 +105,7 @@ sub tabinput { hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )], thai => [qw( iso-8859-11 cp874 MacThai )], vietnamese => [qw( viscii cp1258 MacVietnamese )], - symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )], + symbols => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )], # iso-code shorthand 1 => 'westeur', @@ -124,15 +125,23 @@ sub tabinput { return; } - state $visible = {}; # all present tables + state $visible = {'' => 1}; # all present tables my %row = (offset => 0, cols => 16); - my $endpoint = 255; my $params = $input =~ s/[+](.*)\z// ? $1 : undef; 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 @@ -177,16 +186,16 @@ 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 @@ -196,7 +205,7 @@ 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'], @@ -222,32 +231,36 @@ sub tabinput { $visible->{$_} //= 0 for $row{parent} || (); } - if (length $params) { - # 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{startpoint}) / $row{cols}); + for ($skip + 1 .. ($+{start} / $row{cols}) - 1) { + $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++; + } } + else { + $row{startpoint} = $+{start}; + } + $row{endpoint} = $+{end} || 0; + } + else { + Alert("Unknown option $param for charset $input"); } } - if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) { + if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) { + eval "require Encode::\u$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; @@ -255,60 +268,81 @@ sub tabinput { $row{endpoint} = $end - $start; $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; $row{offset} = $start % 256; + $row{startpoint} = 0; } - elsif ($input eq 'U') { + 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} ||= 1023; + $row{endpoint} *= $row{cell}->{colsize}; + $row{startpoint} *= $row{cell}->{colsize}; + } + 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{set} = 'Unicode ' . ( + $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' : + $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' : + 'allocations' + ); + } + elsif ($input =~ m/^utf-*8$/i) { + $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} = Encode::resolve_alias($input)) { - if ($row{set} eq 'Internal') { - $row{cell} = do 'charset-unicode.inc.pl' - or Alert('Table data could not be read', $@ || $!); - - $row{endpoint } = $row{cell}->{colsize} * ($endpoint || 8191); - $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'; - $row{cell} = do 'charset-utf8.inc.pl' - or Alert('Table data could not be read', $@ || $!); - $row{endpoint} = 255; + $row{offset} = delete $row{startpoint}; + $row{endpoint} ||= 255; + if ($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} .. $row{endpoint} + ]; } else { - if ($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 - ]; + # ~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} .. $row{endpoint}); + } + + if ($row{set} eq 'cp437') { + if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) { + # replace phi glyph + substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ'; } - 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} .. $endpoint); + if ($row{offset} < 0x20) { + # replace control characters by visible variants + my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset}; + substr($row{table}, 0, length $sub) = $sub; } - $row{endpoint} = $endpoint - $row{offset}; - - if ($row{set} eq 'cp437' and !$row{offset}) { - substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign - substr($row{table}, 0, 32) = pack 'U*', map {hex} qw( - 2007 263A 263B 2665 2666 2663 2660 2022 - 25D8 25CB 25D9 2642 2640 266A 266B 263C - 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 - 2191 2193 2192 2190 221F 2194 25B2 25BC - ); + } + elsif ($row{set} eq 'symbol') { + if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60) { + # replace radical extender by closest unicode equivalent + substr($row{table}, 0x60 - $row{offset}, 1) = '│'; + } + if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF) { + substr($row{table}, 0xBD - $row{offset}, 2) = '⏐⎯'; # arrow extenders + substr($row{table}, 0xD2 - $row{offset}, 3) = '®©™'; # serif variants + substr($row{table}, 0xE0 - $row{offset}, 1) = '◊'; # replace lookalike, should match AdobeSymbol + substr($row{table}, 0xE2 - $row{offset}, 3) = '®©™'; # sans-serif variants + substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪'; + substr($row{table}, 0xF0 - $row{offset}, 1) = '€'; + substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭'; } - - $visible->{ascii} = # assume common base - $visible->{ $row{set} } = 1; } + + $row{endpoint} -= $row{offset}; + + $visible->{ascii} = # assume common base + $visible->{ $row{set} } = 1; } else { Alert("Encoding $input unknown"); @@ -332,22 +366,23 @@ sub range_cell { $len /= $colsize; $name //= $len <= 2 ? 'res' : 'reserved'; - if (my $part = $offset/$colsize % $cols) { + if (my $part = ($offset - $info->{startpoint})/$colsize % $cols) { # continued row my $rest = $cols - $part; # remaining $rest = $len if $len < $rest; #TODO: optimise if ($len -= $rest) { # continued on new row my @next = ($len * $colsize, "$class joinu"); + my $separate = $cols - $len > $rest; # columns not on next row if ($len > $rest) { # minority remains push @next, $name, $title; $title ||= $name; - $name = ''; + $name = $separate && '…'; } else { # minority on next row - push @next, '"', $title || $name; + push @next, $separate && '"', $title || $name; } $table->{$offset + $colsize*$rest} //= \@next; $class .= ' joind'; @@ -363,13 +398,15 @@ 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 +417,7 @@ sub range_cell { $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//; $attr .= sprintf ' class="%s"', $class if $class; $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title; - return "$name"; + return "$name\n"; } for my $row (@request) { @@ -394,7 +431,7 @@ for my $row (@request) { printf '
', !$row->{cell} && ' charmap'; my $title = $row->{set}; $title .= " " - for grep { $_ ne 'iso-8859-1' } $row->{parent} // (); + for $row->{parent} || (); printf '', $title; print '' x ($cols + 1); for my $section (qw{thead}) { @@ -423,39 +460,39 @@ for my $row (@request) { printf '%X', ($offset + $row->{offset}) / $rowdiv; } } + say ''; + for (1 .. $cols) { if ($row->{cell}) { print range_cell($row, $offset); next; } + my $cp = $offset + $row->{offset}; my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] : substr $row->{table}, $offset, 1; - if ($glyph eq $NOCHAR) { - print '
%s
'; - next; - } + my ($cell, $name, $class) = $glyph eq $NOCHAR ? () : + $glyphs->glyph_html($glyph); if (exists $get{compare}) { state $visible = {}; - my $cp = $offset + $row->{offset}; - printf '%2$s', + $class = ( $cp == ord $glyph ? 'l4' : $row->{parent} && $glyph eq Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' : + !$class ? undef : $visible->{$glyph} ? 'l2' : - 'l1', - $glyphs->glyph_html($glyph); + 'l1' + ); $visible->{$glyph}++; - next; } - print "\n".$glyphs->glyph_cell($glyph); + say sprintf $class ? '%s' : '', + $name, $class, $cell; } continue { $offset += $colsize; } - print "\n"; } say '
'; }