X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/61672b1912da0a1b1669259718e04b268367f98e..a8a1b6ccabca261ce3335b1f233cf0146bd0fab1:/charset.plp diff --git a/charset.plp b/charset.plp index 73f9377..7794284 100644 --- a/charset.plp +++ b/charset.plp @@ -1,162 +1,237 @@ -<: -use utf8; -use strict; -use warnings; -use open IO => ':utf8'; - -our $VERSION = 'v1.0'; - -$header{content_type} = 'text/html; charset=utf-8'; +<(common.inc.plp)><: -:> - +Html({ + title => 'charset cheat sheet', + version => '1.0', + description => [ + "Reference sheet with all glyphs in common character encoding tables,", + "and an overview of Unicode ranges and UTF-8 bytes.", + ], + keywords => [qw' + charset codepage unicode ascii utf8 latin glyph character encoding + reference common overview table + '], + stylesheet => [qw'light'], + data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'], +}); - - -charset cheat sheet - - +my @tablist = split /[^\w-]+/, $Request || 'default'; - +:>

Character encoding

<: -my $diinfo = do 'digraphs.inc.pl'; -my %di = map { $diinfo->{$_}->[0] => $_ } grep { ref $diinfo->{$_} } - keys %$diinfo; - -use Encode qw(decode resolve_alias); -# generate character table(s) -# (~16x faster than decoding in loop; -# substr strings is twice as fast as splitting to an array) -my %ALIAS = ( -# default => [qw(unicode utf-8 iso-8859-1 cp437 -cp1252- --iso-8859-15- -koi8-f)], - default => [qw(unicode- utf-8 iso-8859-1 -cp1252- --iso-8859-15- cp437 -cp850)], - 0 => [qw(cp437 cp863)], - 1 => [qw(iso-8859-1 cp1252 MacRoman cp850)], - 2 => [qw(iso-8859-2 cp1250 cp852 MacCentralEurRoman MacCroatian MacRumanian)], - 5 => [qw(koi8-f iso-8859-5 cp1251 MacCyrillic cp855 cp866)], - 7 => [qw(iso-8859-7 cp1253 MacGreek cp737 cp869)], - 8 => [qw(iso-8859-8 cp1255 MacHebrew cp862)], -); -my @request = map { - if (my $input = $_) { - my %row = (offset => 0); - my $endpoint = 255; - if ($input =~ s/^--//) { - $row{offset} = $endpoint > 160 ? 160 : 48; - } - elsif ($input =~ s/^-//) { - $row{offset} = $endpoint > 128 ? 128 : 32; - } - if ($input =~ s/-$//) { - $endpoint = $row{offset} ? $row{offset} < 160 ? 159 : 191 : 127; - } +use POSIX qw( ceil ); +use Shiar_Sheet::FormatChar; +my $glyphs = Shiar_Sheet::FormatChar->new; - if ($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', $start >> 8; - } - elsif ($input eq 'U') { - $row{table} = ' ' x 512; - $row{set} = 'Unicode planes'; - $row{cell} = do 'charset-ucplanes.inc.pl'; +sub tabinput { + # generate character table(s) + my $input = shift or return; + + state $ALIAS = { + default => [qw( unicode+0-192 utf-8 iso-8859-1 cp1252+128-159 iso-8859-15+160-191 cp437 cp850+128 )], + us => [qw( cp437 cp863+128 AdobeStandardEncoding gsm0338+0-127 )], + ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )], + westeur => [qw( iso-8859-1 iso-8859-15+160-191 cp1252+128-159 iso-8859-14+160 cp850+128 MacRoman+128 nextstep+128 hp-roman8+160 )], + centeur => [qw( iso-8859-2 iso-8859-16+160 cp1250+128 cp852+128 MacCentralEurRoman+128 MacCroatian+128 MacRomanian+128 )], # MacRumanian only for DB + turkish => [qw( iso-8859-3 iso-8859-9+128 cp857+128 cp1254+128 MacTurkish+128 )], + baltic => [qw( iso-8859-4 iso-8859-13+160 cp775+128 cp1257+128 )], + nordic => [qw( iso-8859-10 cp865+128 cp861+160-191 MacIcelandic+128 MacSami+160 )], + cyrillic => [qw( koi8-f koi8-r+128-192 koi8-u+160-192 iso-8859-5+128 cp1251+128 MacCyrillic+128 cp855+128 cp866+128 )], # MacUkrainian is broken + arabic => [qw( iso-8859-6 cp1006+160 cp864+128 cp1256+128 MacArabic )], # MacFarsi same as MacArabic? + greek => [qw( iso-8859-7 cp1253+128 MacGreek+128 cp737+128 cp869+128 )], + hebrew => [qw( iso-8859-8 cp1255+128 MacHebrew+128 cp862+128 )], + thai => [qw( iso-8859-11 cp874+128-159 MacThai+128 )], + vietnamese => [qw( viscii cp1258 MacVietnamese )], + symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )], + # iso-code shorthand + 1 => 'westeur', + 2 => 'centeur', + 3 => 'turkish', + 4 => 'baltic', + 5 => 'cyrillic', + 6 => 'arabic', + 7 => 'greek', + 8 => 'hebrew', + 9 => 'turkish', + 10 => 'nordic', + 11 => 'thai', + }; + if (my $follow = $ALIAS->{$input}) { + return map { tabinput($_) } ref $follow ? @{$follow} : $follow; + } + + my %row = (offset => 0, cols => 16); + my $endpoint = 255; + my $params = $input =~ s/[+](.*)\z// ? $1 : undef; + + if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) { + $row{offset} = $1; + $endpoint = -$2 if $2; + } + if ($row{offset}) { + $row{setnote} = 'over cp437' if $input eq 'cp850'; + $row{setnote} = 'over iso-8859-1' if $input =~ /^iso-8859-|^cp125/; + } + + if ($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', $start >> 8; + } + elsif ($input eq 'U') { + $row{table} = ' ' x 1024; + $row{set} = 'Unicode planes'; + $row{cell} = do 'charset-ucplanes.inc.pl' + or Alert('Table data could not be read', $@ || $!); + $row{cols} *= 2; + } + elsif ($row{set} = Encode::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 Alert('Table data could not be read', $@ || $!); } - elsif ($row{set} = resolve_alias($input)) { - if ($row{set} eq 'Internal') { - $row{table} = ' ' x ($endpoint < 255 ? 640 : 4096); - $row{set} = 'Unicode BMP'; - $row{cell} = do 'charset-unicode.inc.pl'; - } - elsif ($row{set} eq 'utf-8-strict') { - $row{table} = undef; - $row{set} = 'UTF-8'; - $row{cell} = do 'charset-utf8.inc.pl'; - } - else { - $row{table} = decode($row{set}, pack 'C*', $row{offset} .. $endpoint); - } + elsif ($row{set} eq 'utf-8-strict') { + $row{table} = undef; + $row{set} = 'UTF-8'; + $row{cell} = do 'charset-utf8.inc.pl' + or Alert('Table data could not be read', $@ || $!); } else { - print "

Encoding $input unknown

\n"; + $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint); + # (~16x faster than decoding in loop; + # substr strings is twice as fast as splitting to an array) + + if ($row{set} eq 'cp437') { + 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; } else { - (); + Alert("Encoding $input unknown"); + return; } -} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ } - $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default'; + return \%row; +} +my @request = map { tabinput($_) } @tablist; + 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[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw( - 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C - 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC - ); -} +sub range_cell { + my ($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 /= $colsize; + $name //= $len <= 2 ? 'res' : 'reserved'; + + if (my $part = $offset/$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"); + 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 = $rest; + } + elsif (my $rows = int($len / $cols)) { + # multiple full rows + if ($len -= $rows * $cols) { + # partial row remains + $table->{$offset + $colsize*$rows * $cols} //= [$len*$colsize, "$class joinu", '', $title]; + $class .= ' joind'; + } + $attr .= sprintf ' rowspan=%d', $rows; + $len = $cols; + } -sub quote { - local $_ = shift; - s/"/"/g; - s//>/g; - return $_; + $attr .= sprintf ' colspan=%d', $len unless $len == 1; + $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"; } -my @nibble = (0..9, 'A'..'F'); 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; + my $endpoint = $offset + (length($row->{table}) || 256) * $colsize; + printf '
', !$row->{cell} && ' charmap'; - printf '', $row->{set}; - print '' x 17; + my $title = $row->{set}; + $title .= " " for $row->{setnote} // (); + printf '', $title; + print '' x ($cols + 1); for my $section (qw{thead}) { - print "<$section>'; - for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) { - printf '
%s%s
↱"; - print '', $_ for @nibble; + print "<$section>
", $rowdiv == 1 ? '+' : '↱'; + printf '%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1; print "\n"; } print '
%X', $msb + ($row->{offset} >> 4); - for my $lsb (0 .. $#nibble) { + while ($offset < $endpoint - 1) { + 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}) { - print $row->{cell}->(($msb<<4) + $lsb); + print range_cell($row, $offset); next; } - my $glyph = substr $row->{table}, ($msb<<4) + $lsb, 1; + my $glyph = substr $row->{table}, $offset, 1; if ($glyph eq $NOCHAR) { print ''; next; } - my $info = [ord $glyph]; - if (defined (my $mnem = $di{ord $glyph})) { - $info = $diinfo->{$mnem}; - } - else { - require Unicode::UCD; - my $fullinfo = Unicode::UCD::charinfo(ord $glyph); - $info = [@$fullinfo{qw/code name category script string/}] if $fullinfo; - } - my ($codepoint, $name, $prop, $script, $string) = @$info; - - $glyph = quote($string || $glyph); - my $desc = sprintf 'U+%04X%s', $codepoint, $name && " ($name)"; - my @class = ('X', grep {$_} $prop, $script); - - $glyph = "$glyph" if $prop eq 'Zs'; - - printf "\n".'%s', - join(' ', @class), quote($desc), $glyph; + print "\n".$glyphs->glyph_cell($glyph); + } + continue { + $offset += $colsize; } print "\n"; } - print "
\n"; + say ''; } :> @@ -195,7 +270,7 @@ for my $row (@request) { -
unicode 5.0 + unicode 7.0 proposed deprecated unassigned @@ -203,14 +278,3 @@ for my $row (@request) {
- - -