X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/92240b0d1e4434e0981b3326ca4d7ad8673dbf71..4e051d54eec34ed62a4fa8982c2b34493a61e016:/charset.plp diff --git a/charset.plp b/charset.plp index d3ddea7..73b3740 100644 --- a/charset.plp +++ b/charset.plp @@ -1,80 +1,251 @@ -<: -use utf8; -use strict; -use warnings; -use open IO => ':utf8'; - -our $VERSION = '1.0'; - -$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 - - - - - +:>

Character encoding

<: -my $diinfo = do 'digraphs.inc.pl'; -my %di = map { $diinfo->{$_}->[0] => $_ } grep { ref $diinfo->{$_} } - keys %$diinfo; +use POSIX qw( ceil ); +use Shiar_Sheet::FormatChar; +my $glyphs = Shiar_Sheet::FormatChar->new; +my $cols = 16; # columns -use Encode qw(decode); # generate character table(s) # (~16x faster than decoding in loop; # substr strings is twice as fast as splitting to an array) -my @tables = map { decode($_, pack 'C*', 0..255) } 'iso-8859-1'; +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} ? $row{offset} < 160 ? 159 : 191 : 127; + } + if ($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', $@ || $!); + $cols *= 2; + } + elsif ($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} 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 { + $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint); + } + } + else { + Alert("Encoding $input unknown"); + } + $row{set} ? \%row : (); + } + else { + (); + } +} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ } + $Request =~ /\w/ ? split(m{[/+\s]}, $Request) : 'default'; my $NOCHAR = chr 0xFFFD; -sub quote { - local $_ = shift; - s/"/"/g; - s//>/g; - return $_; +for my $cp437 (grep {$request[$_]->{set} eq 'cp437'} 0 .. $#request) { + substr($request[$cp437]->{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign + substr($request[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw( + 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C + 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC + ); +} + +sub range_cell { + my ($table, $offset) = @_; + my $def = $table->{$offset} or return; + my ($len, $class, $name, $title) = @{$def}; + + 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; + } + + $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 $table (@tables) { - print ''; - for my $section (qw{thead tfoot}) { - print "<$section>
↳"; - print '', $_ for @nibble; - print " \n"; +for my $row (@request) { + my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1; + my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header + my $rowdiv = 16 ** $coldigits; # divider of row headers + + printf '
', !$row->{cell} && ' charmap'; + 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 .. $#nibble) { - print '
%s
↱"; + printf '%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1; + print "\n"; } print '
', $nibble[$msb]; - for my $lsb (0 .. $#nibble) { - my $glyph = substr $table, ($msb<<4) + $lsb, 1; + for my $msb (0 .. ((length($row->{table}) || 256) - 1) / $cols) { + print '
'; + { + my $rowlabel = ($msb + int($row->{offset} / $cols)) * $cols * $colsize; + if (my $rowmod = $rowlabel % $rowdiv) { + # offset in column units + printf '+%X', $rowmod; + } + else { + # divided row offset + printf '%X', $rowlabel / $rowdiv; + } + } + for my $lsb (0 .. $cols - 1) { + my $val = ( ($msb * $cols) + $lsb ) * $colsize; + if ($row->{cell}) { + print range_cell($row->{cell}, $val); + next; + } + + my $glyph = substr $row->{table}, $val, 1; if ($glyph eq $NOCHAR) { print ''; next; } - my $info = [ord $glyph]; - if (defined (my $mnem = $di{ord $glyph})) { - $info = $diinfo->{$mnem}; - } - 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); } - print "\n", $nibble[$msb], "\n"; + print "\n"; } - print "
\n"; + say '
'; } +:> +
+ +
+ +
control + whitespace + diacritic +
letter +
+ 		punctuation +
quote +
+ 		symbol +
math + currency +
+ 		numeric + greek +
latin + cyrillic +
+ 		aramaic +
brahmic + arabic +
+ 		syllabic +
african + japanese + cjk + chinese +
+ 		alphabetic +
+ + +
unicode 7.0 + proposed + deprecated + unassigned + invalid +
+
+