<(common.inc.plp)><:
+my $mode = exists $get{compare};
+my @tablist = split m{/+}, $Request || 'default';
+
Html({
title => 'charset cheat sheet',
version => '1.0',
reference common overview table
'],
stylesheet => [qw'light'],
- data => [qw'charset-unicode.inc.pl charset-utf8.inc.pl'],
+ data => [qw(
+ charset-encoding.inc.pl
+ charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl
+ )],
});
+use List::Util qw( first pairmap pairfirst pairs );
+
:>
-<h1>Character encoding</h1>
+<h1>Character encodings</h1>
+
+<p>
+<:
+if ($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(
+ $selected ? '<strong>%s</strong>' : '<a href="%2$s">%s</a>',
+ EscapeHTML($title), $href
+ );
+}
+
+print join " •\n", (
+ map {
+ join " ·\n", pairmap {
+ optionlink($b || ucfirst $a, '/charset'.($a && "/$a?compare"), $a eq $Request);
+ } @{$_}
+ }
+ [
+ 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,
+ ],
+);
+:>.
+</p>
<:
use POSIX qw( ceil );
use Shiar_Sheet::FormatChar;
my $glyphs = Shiar_Sheet::FormatChar->new;
-my $cols = 16; # columns
-
-# generate character table(s)
-# (~16x faster than decoding in loop;
-# substr strings is twice as fast as splitting to an array)
-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} ? $row{offset} < 160 ? 159 : 191 : 127;
- }
- if ($row{offset}) {
- $row{setnote} = 'over cp437' if $input eq 'cp850';
- $row{setnote} = 'over iso-8859-1' if $input =~ /^iso-8859-|^cp125/;
- }
-
- if ($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 1024;
- $row{set} = 'Unicode planes';
- $row{cell} = do 'charset-ucplanes.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
- $cols *= 2;
- }
- elsif ($row{set} = Encode::resolve_alias($input)) {
- if ($row{set} eq 'Internal') {
- $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
- $row{set} = 'Unicode BMP';
- $row{cell} = do 'charset-unicode.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
- }
- elsif ($row{set} eq 'utf-8-strict') {
- $row{table} = undef;
- $row{set} = 'UTF-8';
- $row{cell} = do 'charset-utf8.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
+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} || {};
+
+ 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);
+
+ if (not defined $params) {
+ my @parents = @{ $charset->{inherit} || [] };
+
+ if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) {
+ $row{parent} = $parent;
+ $params = $part;
+ $params = 80 unless $visible->{$parent}
+ or ($input eq 'MacCroatian' and defined $visible->{MacRomanian});
+ }
+ elsif (defined $visible->{ascii}) {
+ $row{parent} = $parents[0];
+ $params = $parents[1] // 80;
+ $params = 80 if hex $params >= 0x80; # ascii offset at most
+ }
+ elsif (@parents) {
+ $row{parent} = $parents[0];
+ $params = $parents[1] if hex $parents[1] == 0; # apply ascii end
+ }
+ $visible->{$_} //= 0 for $row{parent} || ();
+ }
+
+ 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 (?<start> \p{AHex}+) (?: [-] (?<end> \p{AHex}+) )? \z }x) {
+ if (defined $row{endpoint}) {
+ # extend earlier range
+ my $skip = int(($row{endpoint} || $row{startpoint}) / $row{cols});
+ for ($skip + 1 .. (hex($+{start}) / $row{cols}) - 1) {
+ $row{skip}->{ $_ * $row{cols} - $row{startpoint} }++;
+ }
}
else {
- $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
+ $row{startpoint} = hex $+{start};
}
+ $row{endpoint} = hex($+{end} || 0);
}
else {
- Alert("Encoding <q>$input</q> unknown");
+ Alert("Unknown option <q>$param</q> for charset $input");
}
- $row{set} ? \%row : ();
}
- else {
- ();
+
+ if ($input =~ m{ \A (?:wing|web)dings \d* \z }ix) {
+ eval "require Encode::\u$input";
}
-} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
- $Request =~ /\w/ ? split(m{[/+\s]}, $Request) : '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[$cp437]->{table}, 0, 32) = pack 'U*', map {hex} qw(
- 2007 263A 263B 2665 2666 2663 2660 2022 25D8 25CB 25D9 2642 2640 266A 266B 263C
- 25BA 25C4 2195 203C 00B6 00A7 25AC 21A8 2191 2193 2192 2190 221F 2194 25B2 25BC
- );
+ if ($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', $@ || $!);
+ $row{endpoint} = 0xFF;
+ }
+ elsif ($row{set} = Encode::resolve_alias($input)) {
+ $row{offset} = delete $row{startpoint};
+ $row{endpoint} ||= 0xFF;
+ if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') {
+ # array of possibly multiple characters per code point
+ $row{table} = [
+ map { Encode::decode($row{set}, pack 'C*', $_) } $row{offset} .. $row{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} .. $row{endpoint});
+ }
+
+ if ($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->{ $row{set} } = 1;
+ }
+ else {
+ Alert("Encoding <q>$input</q> unknown");
+ return;
+ }
+ push @request, \%row;
}
+tabinput($_) for @tablist;
+
+my $NOCHAR = chr 0xFFFD;
sub range_cell {
- my ($table, $offset) = @_;
+ my ($info, $offset) = @_;
+ my $table = $info->{cell} or return;
my $def = $table->{$offset} or return;
my ($len, $class, $name, $title) = @{$def};
+ my $cols = $info->{cols};
my $colsize = $table->{colsize} || 1;
my $attr = '';
$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';
}
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;
}
$attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
$attr .= sprintf ' class="%s"', $class if $class;
$attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
- return "<td$attr>$name";
+ return "<td$attr>$name\n";
}
for my $row (@request) {
+ my $cols = $row->{cols};
my $colsize = $row->{cell} && $row->{cell}->{colsize} || 1;
my $coldigits = ceil(log($colsize * $cols) / log(16)); # uniform length of hexadecimal header
- my $rowdiv = 16 ** $coldigits; # divider of row headers
+ 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} * $colsize || 0;
printf '<div class="section"><table class="glyphs%s">', !$row->{cell} && ' charmap';
my $title = $row->{set};
- $title .= " <aside>($_)</aside>" for $row->{setnote} // ();
+ $title .= " <aside>(over $_)</aside>"
+ for $row->{parent} || ();
printf '<caption>%s</caption>', $title;
print '<col>' x ($cols + 1);
for my $section (qw{thead}) {
- print "<$section><tr><th>↱";
+ print "<$section><tr><th>", $rowdiv == 1 ? '+' : '↱';
printf '<th>%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
print "\n";
}
+
print '<tbody>';
- for my $msb (0 .. ((length($row->{table}) || 256) - 1) / $cols) {
+ while ($offset <= $row->{endpoint} * $colsize) {
+ if ($row->{skip}->{$offset}) {
+ $offset += $cols * $colsize;
+ next;
+ }
+
print '<tr><th>';
- {
- my $rowlabel = ($msb + int($row->{offset} / $cols)) * $cols * $colsize;
- if (my $rowmod = $rowlabel % $rowdiv) {
+ if (defined $row->{skip}->{$offset}) {
+ print '⋮';
+ }
+ else {
+ if (my $rowmod = $offset % $rowdiv) {
# offset in column units
printf '<small>+%X</small>', $rowmod;
}
else {
# divided row offset
- printf '%X', $rowlabel / $rowdiv;
+ printf '%X', ($offset + $row->{offset}) / $rowdiv;
}
}
- for my $lsb (0 .. $cols - 1) {
- my $val = ( ($msb * $cols) + $lsb ) * $colsize;
+ say '';
+
+ for (1 .. $cols) {
if ($row->{cell}) {
- print range_cell($row->{cell}, $val);
+ print range_cell($row, $offset);
next;
}
- my $glyph = substr $row->{table}, $val, 1;
- if ($glyph eq $NOCHAR) {
- print '<td>';
- next;
+ 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 ? () :
+ $glyphs->glyph_html($glyph);
+
+ if ($mode) {
+ state $visible = {};
+ $class = (
+ $cp == ord $glyph ? 'l4' :
+ $row->{parent} && $glyph eq
+ Encode::decode($row->{parent}, pack 'C', $cp) ? 'l3' :
+ !$class ? undef :
+ $visible->{$glyph} ? 'l2' :
+ 'l1'
+ );
+ $visible->{$glyph}++;
}
- print "\n".$glyphs->glyph_cell($glyph);
+ say sprintf $class ? '<td title="%s" class="X %s">%s' : '<td title="%s">',
+ $name, $class, $cell;
+ }
+ continue {
+ $offset += $colsize;
}
- print "\n";
}
say '</table></div>';
}
<hr>
<div class="legend">
- <table class="glyphs"><tr>
+ <table class="glyphs"><tr><: if ($mode) { :>
+ <td class="X l4">unicode
+ <td class="X l3">inherited
+ <td class="X l2">existing
+ <td class="X l1">original
+ <td class="">unassigned
+<: } else { :>
<td class="X Cc">control
<td class="X Zs"><span>whitespace</span>
<td class="X Mn">diacritic<table class="glyphs"><tr>
<td class="X Xd">deprecated
<td class="">unassigned
<td class="X Xi">invalid
- </table>
+<: } :> </table>
</div>
git.shiar.nl / sheet.git / blobdiff