X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/a180fb63ca5b34948194aad96395f3ba8ec5b22d..HEAD:/charset.plp diff --git a/charset.plp b/charset.plp index b6b61b9..bbab1d6 100644 --- a/charset.plp +++ b/charset.plp @@ -1,8 +1,12 @@ <(common.inc.plp)><: +my $mode = exists $get{compare}; +my @tablist = split m{/+}, $Request || 'default'; + Html({ title => 'charset cheat sheet', - version => '1.0', + version => '1.3', + canonical => "/charset/$Request" . ($mode && '?compare'), description => [ "Reference sheet with all glyphs in common character encoding tables,", "and an overview of Unicode ranges and UTF-8 bytes.", @@ -11,12 +15,13 @@ Html({ charset codepage unicode ascii utf8 latin glyph character encoding reference common overview table '], - stylesheet => [qw'light'], - data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'], + (stylesheet => [qw'light']) x !$mode, + data => [qw( + charset.inc.pl charset-encoding.inc.pl + charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl + )], }); -my @tablist = split m{/+}, $Request || 'default'; - use List::Util qw( first pairmap pairfirst pairs ); :> @@ -24,33 +29,34 @@ use List::Util qw( first pairmap pairfirst pairs );

<: +if ($tablist[0] eq 'unicode') { + say "Detailed allocation of Unicode blocks."; + print "See ", showlink("charsets", '/charset'), " to compare specific encodings"; +} +else { + if ($tablist[0] eq 'default') { - say "Overview of Unicode allocation and common latin code pages."; + say "Overview of Unicode ", showlink("allocation", '/charset/unicode'); + say "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), $href - ); -} - print join " •\n", ( map { join " ·\n", pairmap { - optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request); + showlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request); } @{$_} } [ iso => 'ISO', + win => 'Windows', dos => 'DOS', mac => 'Apple', ebcdic => 'EBCDIC', + legacy => 'legacy', $tablist[0] eq 'default' ? () : ('' => 'common'), ], [ @@ -63,6 +69,8 @@ print join " •\n", ( hebrew => 0, ], ); + +} :>.

@@ -72,151 +80,43 @@ use Shiar_Sheet::FormatChar; my $glyphs = Shiar_Sheet::FormatChar->new; my @request; +my $charsets = Data('charset'); + sub tabinput { # generate character table(s) my $input = shift or return; + my $params = $input =~ s/[+](.*)\z// ? $1 : undef; + my $charset = $charsets->{lc $input} || {}; - state $ALIAS = { - default => [qw( u+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )], - unicode => [qw( uu+cols=32 u+0-4095 u+4096-6319 u+6320-8191 )], - us => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )], - ebcdic => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )], - iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16], - dos => [qw( cp437 cp865 cp861 cp860 cp863 cp850 cp857 cp852 cp775 - cp737 cp869 cp866 cp855 cp862 cp864 )], - aix => [qw( cp1006 )], - win => [qw( cp1252 cp1250 cp1254 cp1257 cp1258 cp1253 cp1251 cp1255 cp1256 cp874 )], - mac => [qw( MacRoman MacRomanian MacRumanian MacCroatian MacCentralEurRoman MacTurkish MacIcelandic MacSami - MacGreek MacCyrillic MacHebrew MacArabic MacFarsi MacThai )], - - # languages - 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 - 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 )], - - # 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}) { - tabinput($_) for ref $follow ? @{$follow} : $follow; + if (ref $charset ne 'HASH') { + $params and Alert("Parameters ignored for $input", + "Cannot apply $params to multiple charsets.", + ); + tabinput($_) for ref $charset ? @{$charset} : $charset; 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; + $row{$_} = $charset->{$_} for qw( note table ); # copy metadata 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 @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 $row{parent} || (); } @@ -228,62 +128,36 @@ sub tabinput { elsif ($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} }++; } } 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 .. $end; - utf8::upgrade($row{table}); # prevent latin1 output - $row{endpoint} = $end - $start; - $row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8; - $row{offset} = $start % 256; - } - 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 * $row{cell}->{colsize}; + if ($charset->{setup}) { + eval { $charset->{setup}->(\%row) } + or Alert("Incomplete setup of $input", $@); } - 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; + $row{endpoint} ||= 0xFF; + + if (defined $row{table} or defined $row{cell}) { + $row{set} //= $input; } - elsif ($row{set} = Encode::resolve_alias($input)) { - $row{endpoint} ||= 255; - if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') { + elsif ($row{set} = Encode::resolve_alias($charset->{set} // $input)) { + $row{offset} = delete $row{startpoint}; + if ($charset->{varchar}) { # array of possibly multiple characters per code point $row{table} = [ map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint} @@ -294,33 +168,49 @@ 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 > $row{endpoint}; - # replace phi glyph - substr($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 - 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 - 2191 2193 2192 2190 221F 2194 25B2 25BC - ); - } - } - $visible->{ascii} = # assume common base - $visible->{ $row{set} } = 1; + $row{endpoint} -= $row{offset}; + $visible->{ascii}++; # assume common base + $row{set} = $input if $charset->{set}; # base override } else { Alert("Encoding $input unknown"); return; } + + if (my $replace = $charset->{replace}) { + while (my ($offset, $sub) = each %{$replace}) { + $offset -= $row{offset}; + + if (ref $row{table} eq 'ARRAY') { + $row{table}->[$offset] = $sub + if $offset >= 0 and $offset <= $row{endpoint}; + next; + } + + my $length = length $sub; + + if ($offset < 0) { + $offset > -$length or next; # at least one character after start + # trim leftmost part to start at offset + substr($sub, 0, -$offset) = ''; + $length += $offset; + $offset = 0; + } + + if ((my $excess = $row{endpoint} - $offset - $length + 1) < 0) { + $excess > -$length or next; + # trim rightmost part to prevent overflow + substr($sub, $excess) = ''; + $length += $excess; + } + + substr($row{table}, $offset, $length) = $sub; + } + } + push @request, \%row; + $visible->{ $row{set} } = 1 if $row{table}; } tabinput($_) for @tablist; @@ -338,22 +228,23 @@ 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'; @@ -397,12 +288,12 @@ 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} // (); + $title .= " " for $row->{parent} || (); + $title .= " " for $row->{note} || (); printf '', $title; print '' x ($cols + 1); for my $section (qw{thead}) { @@ -410,15 +301,16 @@ for my $row (@request) { printf ''; - while ($offset < $row->{endpoint}) { - if ($row->{skip}->{$offset}) { + while ($offset <= $row->{endpoint} * $colsize) { + if ($row->{skip}->{$offset + $row->{offset}}) { $offset += $cols * $colsize; next; } print '
%s
%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1; print "\n"; } + print '
'; - if (defined $row->{skip}->{$offset}) { + if (defined $row->{skip}->{$offset + $row->{offset}}) { print '⋮'; } else { @@ -439,28 +331,27 @@ for my $row (@request) { 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) = !defined $glyph || $glyph eq $NOCHAR ? () : + $glyphs->glyph_html($glyph); - if (exists $get{compare}) { + if ($mode) { 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' : + !defined $cell ? undef : $visible->{$glyph} ? 'l2' : - 'l1', - $glyphs->glyph_html($glyph); + 'l1' + ); $visible->{$glyph}++; - next; } - say $glyphs->glyph_cell($glyph); + printf '%s', $class, $cell) : '>'; } continue { $offset += $colsize; @@ -473,7 +364,7 @@ for my $row (@request) {
- <: if (exists $get{compare}) { :> +
<: if ($mode) { :>
unicode inherited existing @@ -511,7 +402,7 @@ for my $row (@request) {
-
unicode 7.0 + unicode 10.0 proposed deprecated unassigned