X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/99e52849c6eebe4caa1018e960054d663352dd5d..8e8998a5c2e51159db372eed7e4e293df83ccc91:/charset.plp diff --git a/charset.plp b/charset.plp index c89c924..1b672e9 100644 --- a/charset.plp +++ b/charset.plp @@ -12,7 +12,10 @@ Html({ reference common overview table '], stylesheet => [qw'light'], - data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'], + data => [qw( + charset-encoding.inc.pl + charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl + )], }); my @tablist = split m{/+}, $Request || 'default'; @@ -73,13 +76,16 @@ use Shiar_Sheet::FormatChar; my $glyphs = Shiar_Sheet::FormatChar->new; my @request; +my $charsets = do 'charset-encoding.inc.pl' + or Alert('Encoding metadata could not be read', $@ || $!); + sub tabinput { # generate character table(s) my $input = shift or return; state $ALIAS = { - default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], - unicode => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )], + default => [qw( u+0-27F utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + unicode => [qw( uu+cols=32+realsize u+0-FFF u+1000-18AF u+18B0-1FFF )], us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], @@ -98,14 +104,14 @@ sub tabinput { baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )], nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )], cyrillic => [qw( koi8-r koi8-u koi8-f iso-8859-5 cp1251 MacCyrillic cp866 cp855 - U4 U2DE-2DF UA64-A69 U50-52 )], # MacUkrainian is broken + +400 +2DE0 +A640-A69F +500-52F )], # MacUkrainian is broken arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi - U6 U8A-8F+0-31+64 U75-77 )], - greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 U37-3F U1F )], - hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 U59-5F )], + +600 +8A0-8BF+8E0 +750-77F )], + greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 +370 +1F00 )], + hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 +590 )], 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', @@ -125,99 +131,28 @@ sub tabinput { return; } - state $visible = {}; # all present tables + state $visible = {'' => 1}; # all present tables my %row = (offset => 0, cols => 16); 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'], - - 'iso-8859-2' => ['iso-8859-1' => '160'], - 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9 - 'iso-8859-4' => ['iso-8859-2' => '160'], - 'iso-8859-5' => ['iso-8859-1' => '160'], - 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'], - 'iso-8859-7' => ['iso-8859-1' => '160'], - 'iso-8859-8' => ['iso-8859-1' => '160'], - 'iso-8859-9' => ['iso-8859-1' => '208-223+240'], - 'iso-8859-10' => ['iso-8859-4' => '160'], - 'iso-8859-11' => ['iso-8859-1' => '160'], - 'iso-8859-13' => ['iso-8859-4' => '160'], - 'iso-8859-14' => ['iso-8859-1' => '160'], - 'iso-8859-15' => ['iso-8859-1' => '160-191'], - 'iso-8859-16' => ['iso-8859-2' => '160'], - 'hp-roman8' => ['iso-8859-1' => '160'], - - 'cp1252' => ['iso-8859-1' => '128-159'], - 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'], - 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'], - 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162 - 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'], - 'cp1251' => ['cp1252' => '128'], - 'cp1253' => ['cp1252' => '128'], - 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'], - 'cp1256' => ['cp1252' => '128'], - 'cp1258' => ['cp1252' => '128-159+192'], - - 'cp850' => ['cp437' => '144'], - 'cp860' => ['cp437' => '128-175'], - 'cp861' => ['cp865' => '128-175'], - 'cp863' => ['cp437' => '128-175'], - 'cp865' => ['cp437' => '144-175'], - 'cp852' => ['cp850' => '128', 'cp437' => '128'], - 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'], - 'cp775' => ['cp850' => '128'], # partial cp437 - 'cp866' => ['cp437' => '128-175+224'], - 'cp855' => ['cp437' => '128'], - 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'], - 'cp737' => ['cp437' => '128-175+224'], - 'cp869' => ['cp437' => '128'], - 'cp862' => ['cp437' => '128-159'], - 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants - - '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'], - 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned - '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 - 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents - 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'], - - 'cp37' => ['posix-bc' => '0'], - 'posix-bc' => ['cp1047' => '64'], - 'cp500' => ['cp37' => '64-95+176-191'], - 'cp1047' => ['cp37' => '16-47+80-95+160-191'], - 'cp1026' => ['cp37' => '64'], - 'cp875' => ['cp37' => '48'], - - }; - - my @parents = @{ $INHERIT->{$input} || [] }; + my $charset = $charsets->{$input} || {}; + my @parents = @{ $charset->{inherit} || [] }; if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) { $row{parent} = $parent; $params = $part; - $params = 128 unless $visible->{$parent} + $params = 80 unless $visible->{$parent} or ($input eq 'MacCroatian' and defined $visible->{MacRomanian}); } elsif (defined $visible->{ascii}) { $row{parent} = $parents[0]; - $params = $parents[1] // 128; - $params = 128 if $params >= 128; # ascii offset at most + $params = $parents[1] // 80; + $params = 80 if hex $params >= 0x80; # ascii offset at most } elsif (@parents) { $row{parent} = $parents[0]; - $params = $parents[1] if $parents[1] == 0; # apply ascii end + $params = $parents[1] if hex $parents[1] == 0; # apply ascii end } $visible->{$_} //= 0 for $row{parent} || (); } @@ -229,50 +164,59 @@ sub tabinput { elsif ($param =~ m{ \A cols = (\d+) \z }x) { $row{cols} = $1; } - elsif ($param =~ m{ \A (? \d+) (?: [-] (? \d+) )? \z }x) { + elsif ($param =~ m{ \A (? \p{AHex}+) (?: [-] (? \p{AHex}+) )? \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} }++; + my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols}); + for ($skip + 1 .. (hex($+{start}) / $row{cols}) - 1) { + $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++; } } else { - $row{offset} = $+{offset}; + $row{startpoint} = hex $+{start}; } - $row{endpoint} = $+{endpoint} || 0; + $row{endpoint} = hex($+{end} || 0); } else { Alert("Unknown option $param for charset $input"); } } - if ($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 .. $end; + if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) { + eval "require Encode::\u$input"; + } + + if ($input eq '') { + $row{offset} = delete $row{startpoint}; + $row{set} = 'Unicode characters'; + my $block = $row{offset} >> 8; + $row{endpoint} ||= ($block + 1 << 8) - 1; + $block == ($row{endpoint} >> 8) or undef $block; + + $row{table} = join '', map { chr } $row{offset} .. $row{endpoint}; utf8::upgrade($row{table}); # prevent latin1 output - $row{endpoint} = $end - $start; - $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; - $row{offset} = $start % 256; + + $row{endpoint} -= $row{offset}; + + if (defined $block) { + $row{set} = sprintf 'Unicode block U+%02Xxx', $block; + $row{offset} %= 0x100; + } } 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} ||= 0x3FF; + $row{set} = 'Unicode planes'; } 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{offset}; - $row{offset} = 0; + $row{endpoint} ||= 0x1FFF; $row{set} = 'Unicode ' . ( - $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' : - $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' : + $row{startpoint} < 0x1000 && $row{endpoint} < 0x1000 ? 'BMP' : + $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' : 'allocations' ); } @@ -280,10 +224,11 @@ sub tabinput { $row{set} = 'UTF-8'; $row{cell} = do 'charset-utf8.inc.pl' or Alert('Table data could not be read', $@ || $!); - $row{endpoint} = 255; + $row{endpoint} = 0xFF; } elsif ($row{set} = Encode::resolve_alias($input)) { - $row{endpoint} ||= 255; + $row{offset} = delete $row{startpoint}; + $row{endpoint} ||= 0xFF; if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') { # array of possibly multiple characters per code point $row{table} = [ @@ -295,24 +240,35 @@ sub tabinput { # substr strings is twice as fast as splitting to an array $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint}); } - $row{endpoint} -= $row{offset}; if ($row{set} eq 'cp437') { - for my $phipos (237 - $row{offset}) { - next if $phipos < 0 or $phipos > $row{endpoint}; + if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) { # replace phi glyph - substr($row{table}, $phipos, 1) = pack 'U*', 0x3D5; + substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ'; } - if ($row{offset} == 0) { + if ($row{offset} < 0x20) { # replace control characters by visible variants - substr($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 - ); + my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset}; + substr($row{table}, 0, length $sub) = $sub; } } + elsif ($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) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭'; + } + } + + $row{endpoint} -= $row{offset}; $visible->{ascii} = # assume common base $visible->{ $row{set} } = 1; @@ -339,22 +295,23 @@ sub range_cell { $len /= $colsize; $name //= $len <= 2 ? 'res' : 'reserved'; - if (my $part = $offset/$colsize % $cols) { + if (my $part = ($offset/$colsize - $info->{startpoint}) % $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'; @@ -398,12 +355,12 @@ 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 = $row->{startpoint} || 0; + my $offset = $row->{startpoint} * $colsize || 0; printf '
', !$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}) { @@ -411,8 +368,9 @@ for my $row (@request) { printf ''; - while ($offset < $row->{endpoint}) { + while ($offset <= $row->{endpoint} * $colsize) { if ($row->{skip}->{$offset}) { $offset += $cols * $colsize; next;
%s
%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1; print "\n"; } + print '