+<: +if ($tablist[0] eq 'default') { + say "Overview of Unicode allocation and common latin code pages."; + say "Compare alternate charsets:"; +} +else { + say "Charset comparison:"; +} + +sub optionlink { + my ($title, $href, $selected) = @_; + return sprintf( + $selected ? '%s' : '%s', + EscapeHTML($title), $href + ); +} + +print join " â¢\n", ( + map { + join " ·\n", pairmap { + optionlink($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', + centeur => 'Central', + norteur => 'North European', + turkish => 0, + greek => 0, + cyrillic => 0, + hebrew => 0, + ], +); +:>. +
<: use POSIX qw( ceil ); @@ -31,9 +78,10 @@ sub tabinput { my $input = shift or return; state $ALIAS = { - default => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], - us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )], - ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )], + default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], + unicode => [qw( uu+cols=32+realsize u+0-4095 u+4096-6319 u+6320-8191 )], + us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], + ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775 cp737 cp869 cp866 cp855 cp862 cp864 )], @@ -46,12 +94,15 @@ sub tabinput { westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )], centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )], + norteur => [qw( baltic nordic )], 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 )], # MacUkrainian is broken - arabic => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi )], - greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )], - hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 )], + 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 + 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 )], thai => [qw( iso-8859-11 cp874 MacThai )], vietnamese => [qw( viscii cp1258 MacVietnamese )], symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )], @@ -76,14 +127,11 @@ sub tabinput { state $visible = {}; # all present tables my %row = (offset => 0, cols => 16); - my $endpoint = 255; my $params = $input =~ s/[+](.*)\z// ? $1 : undef; if (not defined $params) { - use List::Util qw( first pairfirst pairs ); - state $INHERIT = { - 'cp437' => ['cp850' => 0], # ascii range overridden later + 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later 'gsm0338' => ['ascii' => '0-127'], 'iso-8859-2' => ['iso-8859-1' => '160'], @@ -129,21 +177,29 @@ 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 '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} || [] }; @@ -156,79 +212,99 @@ sub tabinput { } elsif (defined $visible->{ascii}) { $row{parent} = $parents[0]; - $params = $parents[1] || 128; + $params = $parents[1] // 128; $params = 128 if $params >= 128; # ascii offset at most } elsif (@parents) { $row{parent} = $parents[0]; + $params = $parents[1] if $parents[1] == 0; # apply ascii end } - $visible->{$_} //= 0 for $row{parent}; + $visible->{$_} //= 0 for $row{parent} || (); } - if (defined $params) { - $params =~ m{ - \A (?$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} }++; + 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 (?
$paramfor charset $input"); } } - if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) { + 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 : $start + 240; - $row{table} = join '', map { chr } $start .. $end+15; + my $end = $2 ? (hex($2) << 4) + $row{cols} - 1 : $start + 255; + $row{table} = join '', map { chr } $start .. $end; utf8::upgrade($row{table}); # prevent latin1 output - $row{endpoint} = $end + 14 - $start; + $row{endpoint} = $end - $start; $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; + $row{offset} = $start % 256; } - elsif ($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{cols} *= 2; $row{endpoint} = 1023 * $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{offset}; + $row{offset} = 0; + $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{set} = 'Unicode BMP'; - $row{cell} = do 'charset-unicode.inc.pl' - or Alert('Table data could not be read', $@ || $!); - $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize}; - } - elsif ($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; + $row{endpoint} ||= 255; + if ($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 - ]; - } - 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); - } - $row{endpoint} = $endpoint - $row{offset}; + # ~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}); + } + $row{endpoint} -= $row{offset}; - if ($row{set} eq 'cp437' and !$row{offset}) { - substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign + if ($row{set} eq 'cp437') { + for my $phipos (237 - $row{offset}) { + next if $phipos < 0 or $phipos > $row{endpoint}; + # replace phi glyph + substr($row{table}, $phipos, 1) = pack 'U*', 0x3D5; + } + if ($row{offset} == 0) { + # 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 @@ -236,10 +312,10 @@ sub tabinput { 2191 2193 2192 2190 221F 2194 25B2 25BC ); } - - $visible->{ascii} = # assume common base - $visible->{ $row{set} } = 1; } + + $visible->{ascii} = # assume common base + $visible->{ $row{set} } = 1; } else { Alert("Encoding
$inputunknown"); @@ -287,11 +363,24 @@ sub range_cell { } elsif (my $rows = int($len / $cols)) { # multiple full rows + my $rowsize = $colsize * $cols; if ($len -= $rows * $cols) { # partial row remains - $table->{$offset + $colsize*$rows * $cols} //= [$len*$colsize, "$class joinu", '', $title]; + $table->{$offset + $rowsize * $rows} //= [$len*$colsize, "$class joinu", '', $title]; $class .= ' joind'; } + + unless ($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; $len = $cols; } @@ -300,7 +389,7 @@ 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 "
| '; - { + if (defined $row->{skip}->{$offset}) { + print 'â®'; + } + else { if (my $rowmod = $offset % $rowdiv) { # offset in column units printf '+%X', $rowmod; @@ -340,39 +432,39 @@ 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 $NOCHAR) { - print ' | '; - next; - } + my ($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 ' |
|---|