termcol: fix index numbers of 88-colour greyscale ramp
[sheet.git] / charset.plp
index e9c4fcc7c0d323f2b34895096ad73aacae08339a..440cabe56cf0b0a384a391517487542bdab8c682 100644 (file)
@@ -1,24 +1,20 @@
-<:
-use utf8;
-use strict;
-use warnings;
-use open IO => ':utf8';
-
-our $VERSION = '1.0';
-
-$header{content_type} = 'text/html; charset=utf-8';
+<(common.inc.plp)><:
+       our $VERSION = 'v1.0';
 
 :><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
  "http://www.w3.org/TR/html4/loose.dtd">
-<html>
+<html lang="en">
 
 <head>
+<meta http-equiv="content-type" content="<:= $header{content_type} :>">
 <title>charset cheat sheet</title>
-<meta http-equiv="content-type" content="utf-8">
-<link rel="stylesheet" type="text/css" media="all" href="/base.css">
+<meta name="description" content="Reference sheet with all glyphs in common character encoding tables, and an overview of Unicode ranges and UTF-8 bytes.">
+<meta name="keywords" content="charset, codepage, unicode, ascii, utf8, latin, glyph, character, encoding, reference, common, overview, table">
+<:= stylesheet(qw'light') :>
+<link rel="icon" type="image/png" href="/clip.png">
 </head>
 
-<body>
+<body id="charset">
 <h1>Character encoding</h1>
 
 <:
@@ -31,7 +27,8 @@ use Encode qw(decode resolve_alias);
 # (~16x faster than decoding in loop;
 #  substr strings is twice as fast as splitting to an array)
 my %ALIAS = (
-       default => [qw(utf-8 iso-8859-1 cp437)],
+#      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)],
@@ -39,149 +36,169 @@ my %ALIAS = (
        7 => [qw(iso-8859-7 cp1253 MacGreek cp737 cp869)],
        8 => [qw(iso-8859-8 cp1255 MacHebrew cp862)],
 );
-my @request = grep { defined } map {
-       $_ ? (resolve_alias($_) or print("Encoding $_ unknown") && ()) : ();
+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 "<p>Encoding $input unknown</p>\n";
+               }
+               \%row;
+       }
+       else {
+               ();
+       }
 } map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
        $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
-my @tables = map { $_ eq 'utf-8-strict' ? undef : decode($_, pack 'C*', 0..255) } @request;
 my $NOCHAR = chr 0xFFFD;
 
-for my $cp437 (grep {$request[$_] eq 'cp437'} 0 .. $#request) {
-       substr($tables[$cp437], 237, 1) = pack 'U*', 0x3D5; # phi sign
-       substr($tables[$cp437], 0, 32) = pack 'U*', map {hex} qw(
+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/"/&quot;/g;
-       s/</&lt;/g;
-       s/>/&gt;/g;
-       return $_;
-}
-
-sub printcell_utf8 {
-       my ($value) = @_;
-       if ($value <= 0x7F) {
-               print '<td rowspan="8" colspan="16" class="X di-a"',
-                         ' title="U+0000 – U+007F">Single byte ASCII'
-                       if $value == 0;
-       }
-       elsif ($value <= 0xBF) {
-               print '<td rowspan="4" colspan="16" class="X di-d"',
-                         '>Multi-byte continuation'
-                       if $value == 0x80;
-       }
-       elsif ($value <= 0xC1) {
-               print '<td colspan="2" class="X di-b" style="border-right:none; border-bottom:none"',
-                         ' title="U+0000 – U+007F">(Overl.)'
-                       if $value == 0xC0;
-       }
-       elsif ($value <= 0xDF) {
-               print '<td rowspan="2" colspan="14" class="X di-prop" style="border-left:none"',
-                         ' title="U+0080 – U+03FF">2-byte sequence start'
-                       if $value == 0xC2;
-               print '<td rowspan="1" colspan="16" class="X di-prop" style="border-top:none"',
-                         ' title="U+0400 – U+07FF">'
-                       if $value == 0xD0;
-       }
-       elsif ($value <= 0xEF) {
-               print '<td colspan="16" class="X di-prop"',
-                         ' title="U+0800 – U+FFFF">3-byte sequence start'
-                       if $value == 0xE0;
-       }
-       elsif ($value <= 0xF4) {
-               print '<td colspan="5" class="X di-prop" style="border-right:none"',
-                         ' title="U+1·0000 – U+10·FFFF">4-byte sequence'
-                       if $value == 0xF0;
-       }
-       elsif ($value <= 0xF7) {
-               print '<td colspan="3" class="X di-b" style="border-left:none"',
-                         ' title="U+11·0000 – U+1FF·FFFF">(Overflow)'
-                       if $value == 0xF5;
-       }
-       elsif ($value <= 0xFB) {
-               print '<td colspan="4" class="X di-b"',
-                         ' title="U+200·0000 – U+3FFF·FFFF">5-byte'
-                       if $value == 0xF8;
-       }
-       elsif ($value <= 0xFD) {
-               print '<td colspan="2" class="X di-b"',
-                         ' title="U+4000·0000 – 7FFFF·FFFF">6-byte'
-                       if $value == 0xFC;
-       }
-       elsif ($value <= 0xFF) {
-               print '<td colspan="2" class="di-invalid">Invalid'
-                       if $value == 0xFE;
-       }
-       else {
-               print "\n".'<td class="X">?';
-       }
-}
-
-print "<ul>\n";
-
 my @nibble = (0..9, 'A'..'F');
-for my $tablenum (0 .. $#tables) {
-       print '<li><table class="glyphs">';
-       printf '<caption>%s</caption>', $request[$tablenum] eq 'utf-8-strict' ? 'UTF-8' : $request[$tablenum];
-       print '<col>';
+for my $row (@request) {
+       printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
+       printf '<caption>%s</caption>', $row->{set};
+       print '<col>' x 17;
        for my $section (qw{thead}) {
                print "<$section><tr><th>↱";
                print '<th>', $_ for @nibble;
                print "\n";
        }
        print '<tbody>';
-       for my $msb (0 .. $#nibble) {
-               print '<tr><th>', $nibble[$msb];
+       for my $msb (0 .. (length($row->{table}) || 256) - 1 >> 4) {
+               printf '<tr><th>%X', $msb + ($row->{offset} >> 4);
                for my $lsb (0 .. $#nibble) {
-                       if ($request[$tablenum] eq 'utf-8-strict') {
-                               printcell_utf8(($msb<<4) + $lsb);
+                       if ($row->{cell}) {
+                               print $row->{cell}->(($msb<<4) + $lsb);
                                next;
                        }
 
-                       my $glyph = substr $tables[$tablenum], ($msb<<4) + $lsb, 1;
+                       my $glyph = substr $row->{table}, ($msb<<4) + $lsb, 1;
                        if ($glyph eq $NOCHAR) {
                                print '<td>';
                                next;
                        }
+
                        my $info = [ord $glyph];
                        if (defined (my $mnem = $di{ord $glyph})) {
                                $info = $diinfo->{$mnem};
                        }
+                       else {
+                               require Unicode::UCD;
+                               my $fullinfo = Unicode::UCD::charinfo(ord $glyph);
+                               $info = [@$fullinfo{qw/code name category script string/}] if $fullinfo;
+                       }
                        my ($codepoint, $name, $prop, $script, $string) = @$info;
 
-                       $glyph = quote($string || $glyph);
+                       $glyph = EscapeHTML($string || $glyph);
                        my $desc = sprintf 'U+%04X%s', $codepoint, $name && " ($name)";
                        my @class = ('X', grep {$_} $prop, $script);
 
                        $glyph = "<span>$glyph</span>" if $prop eq 'Zs';
 
                        printf "\n".'<td class="%s" title="%s">%s',
-                               join(' ', @class), quote($desc), $glyph;
+                               join(' ', @class), EscapeHTML($desc), $glyph;
                }
                print "\n";
        }
-       print "</table>\n";
+       print "</table></div>\n";
 }
 
-print "</ul>\n";
-
 :>
 <hr>
 
+<div class="legend">
+       <table class="glyphs"><tr>
+       <td class="X Cc">control
+       <td class="X Zs"><span>whitespace</span>
+       <td class="X Mn">diacritic<table class="glyphs"><tr>
+               <td class="X Sk">letter
+               </table>
+       <td class="X Po">punctuation<table class="glyphs"><tr>
+               <td class="X Pf">quote
+               </table>
+       <td class="X So">symbol<table class="glyphs"><tr>
+               <td class="X Sm">math
+               <td class="X Sc">currency
+               </table>
+       <td class="X No">numeric
+       <td class="X Greek">greek<table class="glyphs"><tr>
+               <td class="X Latin">latin
+               <td class="X Cyrillic">cyrillic
+               </table>
+       <td class="X Aramaic">aramaic<table class="glyphs"><tr>
+               <td class="X Brahmic">brahmic
+               <td class="X Arabic">arabic
+               </table>
+       <td class="X Syllabic">syllabic<table class="glyphs"><tr>
+               <td class="X African">african
+               <td class="X Hiragana">japanese
+               <td class="X Han">cjk
+               <td class="X Bopomofo">chinese
+               </table>
+       <td class="X Alpha">alphabetic
+       </table>
+
+       <table class="glyphs"><tr>
+       <td class="X">unicode 5.0
+       <td class="X Xr">proposed
+       <td class="X Xd">deprecated
+       <td class="">unassigned
+       <td class="X Xi">invalid
+       </table>
+</div>
+
 <p class="footer">
-       <a href="http://sheet.shiar.nl/" rel="home">sheet.shiar.nl</a>/charset
-       <a href="git://git.shiar.nl/sheet" rel="vcs-git" title="Git repository"><:= "v$VERSION" :></a>
+       <a href="/" rel="home">sheet.shiar.nl</a>/charset.<a href="/source/charset.plp"
+        rel="code" title="Written in Perl">plp</a>
+       <a href="http://git.shiar.nl/sheet.git/history/HEAD:/charset.plp"
+        rel="vcs-git" title="Git repository"><:= $VERSION :></a>
        created by <a href="http://shiar.nl/" rel="author">Shiar</a> •
-       <a title="Licensed under the GNU Affero General Public License, version 3" rel="copyright"
-          href="http://www.fsf.org/licensing/licenses/agpl-3.0.html">AGPLv3</a> •
-       last update <:
-               use Time::Format qw(time_format);
-               print time_format('yyyy-mm-dd', (stat $ENV{SCRIPT_FILENAME})[9]);
-       :>
+       <a href="http://www.fsf.org/licensing/licenses/agpl-3.0.html" rel="copyright"
+        title="Licensed under the GNU Affero General Public License, version 3">AGPLv3</a>
 </p>
 
 </html>