common: bump version to 1.11
[sheet.git] / charset.plp
index 7e115a8..f9c50d8 100644 (file)
@@ -1,8 +1,11 @@
 <(common.inc.plp)><:
 
+my $mode = exists $get{compare};
+my @tablist = split m{/+}, $Request || 'default';
+
 Html({
        title => 'charset cheat sheet',
-       version => '1.0',
+       version => '1.2',
        description => [
                "Reference sheet with all glyphs in common character encoding tables,",
                "and an overview of Unicode ranges and UTF-8 bytes.",
@@ -12,11 +15,12 @@ Html({
                reference common overview table
        '],
        stylesheet => [qw'light'],
-       data => [qw'charset-unicode.inc.pl charset-ucplanes.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';
-
 use List::Util qw( first pairmap pairfirst pairs );
 
 :>
@@ -24,26 +28,25 @@ use List::Util qw( first pairmap pairfirst pairs );
 
 <p>
 <:
+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(
-               $selected ? '<strong>%s</strong>' : '<a href="%2$s">%s</a>',
-               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 $Request);
                } @{$_}
        }
        [
@@ -64,6 +67,8 @@ print join " •\n", (
                hebrew   => 0,
        ],
 );
+
+}
 :>.
 </p>
 
@@ -73,146 +78,28 @@ 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;
+       my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
+       my $charset = $charsets->{lc $input} || {};
 
-       state $ALIAS = {
-               default    => [qw( u+0-27F utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
-               unicode    => [qw( uu+cols=32+realsize u+0-FFF u+1000-18AF u+18B0-1FFF )],
-               us         => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
-               ebcdic     => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
-               iso        => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
-               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
-                                  +400 +2DE0 +A640-A69F +500-52F )], # MacUkrainian is broken
-               arabic     => [qw( iso-8859-6 cp1256 MacArabic cp864 cp1006 MacFarsi
-                                  +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 )],
-               symbols    => [qw( symbol dingbats MacDingbats wingdings wingdings2 wingdings3 webdings )],
-
-               # 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 $input",
+                       "Cannot apply <q>$params</q> to multiple charsets.",
+               );
+               tabinput($_) for ref $charset ? @{$charset} : $charset;
                return;
        }
 
        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 = {
-                       'ascii'       => ['' => '00-7F'],
-                       'cp437'       => ['cp850' => 0, 'ascii' => '00-1F+80'], # ascii range overridden later
-                       'gsm0338'     => ['ascii' => '00-7F'],
-                       'dingbats'    => ['' => '20-7F+A0'],
-                       'MacDingbats' => ['dingbats' => '80-9F'],
-                       'AdobeZdingbat'=> ['MacDingbats' => '80-9F'], # should be identical but maps to private use
-                       'symbol'      => ['' => '20-7F+A0'],
-                       'AdobeSymbol' => ['symbol' => '20-7F+A0', '' => '20-7F+A0'], # minor differences, irrelevant except for different '€'
-                       'wingdings'   => ['' => '20'],
-                       'wingdings2'  => ['' => '20'],
-                       'wingdings3'  => ['' => '20'],
-                       'webdings'    => ['' => '20'],
-
-                       'iso-8859-2'  => ['iso-8859-1' => 'A0'],
-                       'iso-8859-3'  => ['iso-8859-1' => 'A0'], #TODO: also apply to iso-8859-9
-                       'iso-8859-4'  => ['iso-8859-2' => 'A0'],
-                       'iso-8859-5'  => ['iso-8859-1' => 'A0'],
-                       'iso-8859-6'  => ['cp1256' => '80', 'iso-8859-1' => 'A0'],
-                       'iso-8859-7'  => ['iso-8859-1' => 'A0'],
-                       'iso-8859-8'  => ['iso-8859-1' => 'A0'],
-                       'iso-8859-9'  => ['iso-8859-1' => 'D0-DF+F0'],
-                       'iso-8859-10' => ['iso-8859-4' => 'A0'],
-                       'iso-8859-11' => ['iso-8859-1' => 'A0'],
-                       'iso-8859-13' => ['iso-8859-4' => 'A0'],
-                       'iso-8859-14' => ['iso-8859-1' => 'A0'],
-                       'iso-8859-15' => ['iso-8859-1' => 'A0-BF'],
-                       'iso-8859-16' => ['iso-8859-2' => 'A0'],
-                       'hp-roman8'   => ['iso-8859-1' => 'A0'],
-
-                       'cp1252'      => ['iso-8859-1' => '80-9F'],
-                       'cp1250'      => ['iso-8859-2' => '80-BF', 'cp1252' => '80'],
-                       'cp1254'      => ['iso-8859-9' => '80-9F', 'cp1252' => '80-9F+D0'],
-                       'cp874'       => ['iso-8859-11' => '80-9F', 'cp1252' => '80'], # windows-874 actually cp1162
-                       'cp1257'      => ['iso-8859-13' => '80-9F+FF', 'cp1252' => '80'],
-                       'cp1251'      => ['cp1252' => '80'],
-                       'cp1253'      => ['cp1252' => '80'],
-                       'cp1255'      => ['iso-8859-8' => '80-DF', 'cp1252' => '80'],
-                       'cp1256'      => ['cp1252' => '80'],
-                       'cp1258'      => ['cp1252' => '80-9F+C0'],
-
-                       'cp850'       => ['cp437' => '90'],
-                       'cp860'       => ['cp437' => '80-AF'],
-                       'cp861'       => ['cp865' => '80-AF'],
-                       'cp863'       => ['cp437' => '80-AF'],
-                       'cp865'       => ['cp437' => '90-AF'],
-                       'cp852'       => ['cp850' => '80', 'cp437' => '80'],
-                       'cp857'       => ['cp850' => '80-AF+D0-EF', 'cp437' => '80'],
-                       'cp775'       => ['cp850' => '80'],  # partial cp437
-                       'cp866'       => ['cp437' => '80-AF+E0'],
-                       'cp855'       => ['cp437' => '80'],
-                       'cp1006'      => ['iso-8859-6' => 'A0', 'cp437' => '80'],
-                       'cp737'       => ['cp437' => '80-AF+E0'],
-                       'cp869'       => ['cp437' => '80'],
-                       'cp862'       => ['cp437' => '80-9F'],
-                       'cp864'       => ['MacArabic' => '80', 'iso-8859-6' => '80', 'cp437' => '80'], #TODO: compare form variants
-
-                       'koi8-u'      => ['koi8-r' => '90-BF'],
-                       'koi8-f'      => ['koi8-u' => '90-BF'],
-
-                       'MacRomanian' => ['MacRoman' => 'A0-BF+D0-DF'],
-                       'MacRumanian' => ['MacRomanian' => 'A0-BF+D0-DF', 'MacRoman' => 'A0-BF+D0-DF'],
-                       'MacCroatian' => ['MacRoman' => 'A0'],
-                       'MacCentralEurRoman' => ['MacRoman' => '80'],
-                       'MacIcelandic'=> ['MacRoman' => 'A0-AF+D0-EF'],
-                       'MacTurkish'  => ['MacRoman' => 'D0-DF'], # F5 is unassigned
-                       'MacSami'     => ['MacIcelandic' => '90-BF+D0-DF+F0', 'MacRoman' => '90'],
-                       'MacGreek'    => ['MacRoman' => '80'],
-                       'MacCyrillic' => ['MacRoman' => '80'],
-                       'MacHebrew'   => ['iso-8859-8' => '80', 'MacRoman' => '80-8F+A0'], # partial ascii
-                       'MacArabic'   => ['iso-8859-6' => '80', 'cp864' => '80', 'MacRoman' => '80'], #TODO: multiple parents
-                       'MacFarsi'    => ['MacArabic' => 'B0-BF', 'MacRoman' => '80'],
-
-                       'cp37'        => ['posix-bc' => '00'],
-                       'posix-bc'    => ['cp1047' => '40'],
-                       'cp500'       => ['cp37' => '40-5F+B0-BF'],
-                       'cp1047'      => ['cp37' => '10-2F+50-5F+A0-BF'],
-                       'cp1026'      => ['cp37' => '40'],
-                       'cp875'       => ['cp37' => '30'],
-
-               };
-
-               my @parents = @{ $INHERIT->{$input} || [] };
+               my @parents = @{ $charset->{inherit} || [] };
 
                if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
                        $row{parent} = $parent;
@@ -257,54 +144,18 @@ sub tabinput {
                }
        }
 
-       if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) {
-               eval "require Encode::\u$input";
+       if ($charset->{setup}) {
+               eval { $charset->{setup}->(\%row) }
+                       or Alert("Incomplete setup of $input", $@);
        }
+       $row{endpoint} ||= 0xFF;
 
-       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} -= $row{offset};
-
-               if (defined $block) {
-                       $row{set} = sprintf 'Unicode block U+%02Xxx', $block;
-                       $row{offset} %= 0x100;
-               }
-       }
-       elsif (lc $input eq 'uu') {
-               $row{cell} = do 'charset-ucplanes.inc.pl'
-                       or Alert('Table data could not be read', $@ || $!);
-               $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} ||= 0x1FFF;
-               $row{set} = 'Unicode ' . (
-                       $row{startpoint} <  0x1000 && $row{endpoint} < 0x1000 ? 'BMP' :
-                       $row{startpoint} >= 0x1000 && $row{endpoint} < 0x2000 ? 'SMP' :
-                       'allocations'
-               );
-       }
-       elsif ($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} = 0xFF;
+       if (defined $row{table} or defined $row{cell}) {
+               $row{set} //= $input;
        }
        elsif ($row{set} = Encode::resolve_alias($input)) {
                $row{offset} = delete $row{startpoint};
-               $row{endpoint} ||= 0xFF;
-               if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
+               if ($charset->{varchar}) {
                        # array of possibly multiple characters per code point
                        $row{table} = [
                                map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{endpoint}
@@ -316,43 +167,47 @@ sub tabinput {
                        $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
                }
 
-               if ($row{set} eq 'cp437') {
-                       if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) {
-                               # replace phi glyph
-                               substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
-                       }
-                       if ($row{offset} < 0x20) {
-                               # replace control characters by visible variants
-                               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;
+               $visible->{ascii}++;  # assume common base
        }
        else {
                Alert("Encoding <q>$input</q> 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;
 
@@ -476,10 +331,10 @@ for my $row (@request) {
                        my $cp = $offset + $row->{offset};
                        my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
                                substr $row->{table}, $offset, 1;
-                       my ($cell, $name, $class) = $glyph eq $NOCHAR ? () :
+                       my ($cell, $name, $class) = !defined $glyph || $glyph eq $NOCHAR ? () :
                                $glyphs->glyph_html($glyph);
 
-                       if (exists $get{compare}) {
+                       if ($mode) {
                                state $visible = {};
                                $class = (
                                        $cp == ord $glyph ? 'l4' :
@@ -506,7 +361,7 @@ for my $row (@request) {
 <hr>
 
 <div class="legend">
-       <table class="glyphs"><tr><: if (exists $get{compare}) { :>
+       <table class="glyphs"><tr><: if ($mode) { :>
        <td class="X l4">unicode
        <td class="X l3">inherited
        <td class="X l2">existing
@@ -544,7 +399,7 @@ for my $row (@request) {
        </table>
 
        <table class="glyphs"><tr>
-       <td class="X">unicode 7.0
+       <td class="X">unicode 10.0
        <td class="X Xr">proposed
        <td class="X Xd">deprecated
        <td class="">unassigned