X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/92240b0d1e4434e0981b3326ca4d7ad8673dbf71..e98b059d1f6bbd6ac8ca06de152959536e4b98b0:/charset.plp diff --git a/charset.plp b/charset.plp index d3ddea7..501e26c 100644 --- a/charset.plp +++ b/charset.plp @@ -4,7 +4,7 @@ use strict; use warnings; use open IO => ':utf8'; -our $VERSION = '1.0'; +our $VERSION = 'v1.0'; $header{content_type} = 'text/html; charset=utf-8'; @@ -13,8 +13,8 @@ $header{content_type} = 'text/html; charset=utf-8'; + charset cheat sheet - @@ -26,13 +26,81 @@ my $diinfo = do 'digraphs.inc.pl'; my %di = map { $diinfo->{$_}->[0] => $_ } grep { ref $diinfo->{$_} } keys %$diinfo; -use Encode qw(decode); +use Encode qw(decode resolve_alias); # generate character table(s) # (~16x faster than decoding in loop; # substr strings is twice as fast as splitting to an array) -my @tables = map { decode($_, 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} ? $row{offset} < 160 ? 159 : 191 : 127; + } + + if ($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'; + $row{cell} = do 'charset-ucplanes.inc.pl'; + } + elsif ($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); + } + } + else { + print "

Encoding $input unknown

\n"; + } + \%row; + } + else { + (); + } +} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ } + $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default'; 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 quote { local $_ = shift; s/"/"/g; @@ -41,27 +109,42 @@ sub quote { return $_; } +print "