X-Git-Url: http://git.shiar.net/sheet.git/blobdiff_plain/47ca3b887159be3f6e047d093ed1bafac4adc706..3efe0eca82fa7f9c0fd125ff25f42901bc7133c8:/charset.plp
diff --git a/charset.plp b/charset.plp
index 25b69b9..0dd1537 100644
--- a/charset.plp
+++ b/charset.plp
@@ -1,417 +1,400 @@
-<:
-use utf8;
-use strict;
-use warnings;
-use open IO => ':utf8';
-
-our $VERSION = '1.0';
+<(common.inc.plp)><:
-$header{content_type} = 'text/html; charset=utf-8';
+my $mode = exists $get{compare};
+my @tablist = split m{/+}, $Request || 'default';
-:>
-
+Html({
+ title => 'charset cheat sheet',
+ version => '1.1',
+ description => [
+ "Reference sheet with all glyphs in common character encoding tables,",
+ "and an overview of Unicode ranges and UTF-8 bytes.",
+ ],
+ keywords => [qw'
+ charset codepage unicode ascii utf8 latin glyph character encoding
+ reference common overview table
+ '],
+ stylesheet => [qw'light'],
+ data => [qw(
+ charset-encoding.inc.pl
+ charset-unicode.inc.pl charset-ucplanes.inc.pl charset-utf8.inc.pl
+ )],
+});
-
-charset cheat sheet
-
-
-
+use List::Util qw( first pairmap pairfirst pairs );
-
-Character encoding
+:>
+Character encodings
+
<:
-my $diinfo = do 'digraphs.inc.pl';
-my %di = map { $diinfo->{$_}->[0] => $_ } grep { ref $diinfo->{$_} }
- keys %$diinfo;
-
-use Encode qw(decode resolve_alias);
-# generate character table(s)
-# (~16x faster than decoding in loop;
-# substr strings is twice as fast as splitting to an array)
-my %ALIAS = (
- default => [qw(utf-8 iso-8859-1 cp437)],
- 0 => [qw(cp437 cp863)],
- 1 => [qw(iso-8859-1 cp1252 MacRoman cp850)],
- 2 => [qw(iso-8859-2 cp1250 cp852 MacCentralEurRoman MacCroatian MacRumanian)],
- 5 => [qw(koi8-f iso-8859-5 cp1251 MacCyrillic cp855 cp866)],
- 7 => [qw(iso-8859-7 cp1253 MacGreek cp737 cp869)],
- 8 => [qw(iso-8859-8 cp1255 MacHebrew cp862)],
+if ($tablist[0] eq 'default') {
+ say "Overview of Unicode allocation and common latin code pages.";
+ say "Compare alternate charsets:";
+}
+else {
+ say "Charset comparison:";
+}
+
+print join " â¢\n", (
+ map {
+ join " ·\n", pairmap {
+ showlink($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,
+ ],
);
-my @request = grep { defined } map {
- $_ ? (resolve_alias($_) or print("Encoding $_ unknown") && ()) : ();
-} map { defined $ALIAS{$_} ? @{ $ALIAS{$_} } : $_ }
- $ENV{PATH_INFO} =~ /\w/ ? split(m{[/+\s]}, $ENV{PATH_INFO}) : 'default';
-my @tables = map { $_ eq 'utf-8-strict' ? undef : $_ eq 'Internal' ? ' 'x512 : decode($_, 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(
- 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
- );
-}
+<:
+use POSIX qw( ceil );
+use Shiar_Sheet::FormatChar;
+my $glyphs = Shiar_Sheet::FormatChar->new;
+my @request;
-sub quote {
- local $_ = shift;
- s/"/"/g;
- s/</g;
- s/>/>/g;
- return $_;
-}
+my $charsets = do 'charset-encoding.inc.pl'
+ or Alert('Encoding metadata could not be read', $@ || $!);
-sub printcell_unicode {
- my ($value) = @_;
- if ($value > 0x1FF) {
- print "\n".'?';
- }
- elsif ($value == 0) {
- print ' | control';
- }
- elsif ($value == 2) {
- print ' | latin';
- }
- elsif ($value == 8) {
- print ' | control';
- }
- elsif ($value == 10) {
- print ' | latin supplement';
- }
- elsif ($value == 0x10) {
- print ' | latin ext-A';
- }
- elsif ($value == 0x18) {
- print ' | latin ext-B';
- }
- elsif ($value == 0x20) {
- print ' | latin ext-B';
- }
- elsif ($value == 0x25) {
- print ' | IPA';
- }
- elsif ($value == 0x2B) {
- print ' | spacing modifier';
- }
- elsif ($value == 0x30) {
- print ' | diacritics';
- }
- elsif ($value == 0x38) {
- print ' | greek';
- }
- elsif ($value == 0x40) {
- print ' | cyrillic';
- }
- elsif ($value == 0x50) {
- print ' | cyrillic+';
- }
- elsif ($value == 0x53) {
- print ' | armenian';
- }
- elsif ($value == 0x58) {
- print ' | hebrew';
- }
- elsif ($value == 0x60) {
- print ' | arabic';
- }
- elsif ($value == 0x70) {
- print ' | syriac';
- }
- elsif ($value == 0x75) {
- print ' | arabic+';
- }
- elsif ($value == 0x78) {
- print ' | thaana';
- }
- elsif ($value == 0x7C) {
- print ' | nko';
- }
- elsif ($value == 0x80) {
- print ' | samaritan';
- }
- elsif ($value == 0x84) {
- print ' | manda';
- }
- elsif ($value == 0x86) {
- print ' | reserved';
- }
- elsif ($value == 0x90) {
- print ' | devanagari';
- }
- elsif ($value == 0x98) {
- print ' | bengali';
- }
- elsif ($value == 0xA0) {
- print ' | gurmukhi';
- }
- elsif ($value == 0xA8) {
- print ' | gujarati';
- }
- elsif ($value == 0xB0) {
- print ' | oriya';
- }
- elsif ($value == 0xB8) {
- print ' | tamil';
- }
- elsif ($value == 0xC0) {
- print ' | telugu';
- }
- elsif ($value == 0xC8) {
- print ' | kannada';
- }
- elsif ($value == 0xD0) {
- print ' | malayalam';
- }
- elsif ($value == 0xD8) {
- print ' | sinhala';
- }
- elsif ($value == 0xE0) {
- print ' | thai';
- }
- elsif ($value == 0xE8) {
- print ' | lao';
- }
- elsif ($value == 0xF0) {
- print ' | tibetan';
- }
- elsif ($value == 0x100) {
- print ' | myanmar';
- }
- elsif ($value == 0x10A) {
- print ' | georgian';
- }
- elsif ($value == 0x110) {
- print ' | hangeul jamo';
- }
- elsif ($value == 0x120) {
- print ' | ethiopic';
- }
- elsif ($value == 0x130) {
- print ' | ethiopic';
- }
- elsif ($value == 0x138) {
- print ' | eth+';
- }
- elsif ($value == 0x13A) {
- print ' | cherokee';
- }
- elsif ($value == 0x140) {
- print ' | unified canadian aboriginal syllabics';
- }
- elsif ($value == 0x160) {
- print ' | unified canadian syllabics';
- }
- elsif ($value == 0x168) {
- print ' | ogham';
- }
- elsif ($value == 0x16A) {
- print ' | runic';
- }
- elsif ($value == 0x170) {
- print ' | tagalog';
- }
- elsif ($value == 0x172) {
- print ' | hanun';
- }
- elsif ($value == 0x174) {
- print ' | buhid';
- }
- elsif ($value == 0x176) {
- print ' | tagb';
- }
- elsif ($value == 0x178) {
- print ' | khmer';
- }
- elsif ($value == 0x180) {
- print ' | mongolian';
- }
- elsif ($value == 0x18B) {
- print ' | canadian+';
- }
- elsif ($value == 0x190) {
- print ' | limbu';
- }
- elsif ($value == 0x195) {
- print ' | tai le';
- }
- elsif ($value == 0x198) {
- print ' | new tai lue';
- }
- elsif ($value == 0x19E) {
- print ' | km';
- }
- elsif ($value == 0x1A0) {
- print ' | lontara';
- }
- elsif ($value == 0x1A2) {
- print ' | tai tham';
- }
- elsif ($value == 0x1AB) {
- print ' | reserved';
- }
- elsif ($value == 0x1B0) {
- print ' | balinese';
- }
- elsif ($value == 0x1B8) {
- print ' | sundanese';
- }
- elsif ($value == 0x1BC) {
- print ' | batak';
- }
- elsif ($value == 0x1C0) {
- print ' | lepcha';
- }
- elsif ($value == 0x1C5) {
- print ' | ol chiki';
- }
- elsif ($value == 0x1C8) {
- print ' | reserved';
+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 $params to multiple charsets.",
+ );
+ tabinput($_) for ref $charset ? @{$charset} : $charset;
+ return;
}
- elsif ($value == 0x1CD) {
- print ' | vedic';
+
+ 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 $row{parent} || ();
}
- elsif ($value == 0x1D0) {
- print ' | phonetic';
+
+ 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 (? \p{AHex}+) (?: [-] (? \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{startpoint} = hex $+{start};
+ }
+ $row{endpoint} = hex($+{end} || 0);
+ }
+ else {
+ Alert("Unknown option $param for charset $input");
+ }
}
- elsif ($value == 0x1D8) {
- print 'phonetic+';
+
+ if ($charset->{setup}) {
+ eval { $charset->{setup}->(\%row) }
+ or Alert("Incomplete setup of $input", $@);
}
- elsif ($value == 0x1DC) {
- print ' | combining';
+ $row{endpoint} ||= 0xFF;
+
+ if (defined $row{table} or defined $row{cell}) {
+ $row{set} //= $input;
+ }
+ elsif ($row{set} = Encode::resolve_alias($input)) {
+ $row{offset} = delete $row{startpoint};
+ if ($charset->{varchar}) {
+ # 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});
+ }
+
+ $row{endpoint} -= $row{offset};
+ $visible->{ascii}++; # assume common base
}
- elsif ($value == 0x1E0) {
- print ' | latin extended additional';
+ else {
+ Alert("Encoding $input unknown");
+ return;
}
- elsif ($value == 0x1F0) {
- print ' | greek+';
+
+ 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;
-sub printcell_utf8 {
- my ($value) = @_;
- if ($value <= 0x7F) {
- print ' | Single byte ASCII'
- if $value == 0;
- }
- elsif ($value <= 0xBF) {
- print ' | Multi-byte continuation'
- if $value == 0x80;
- }
- elsif ($value <= 0xC1) {
- print ' | (Overl.)'
- if $value == 0xC0;
- }
- elsif ($value <= 0xDF) {
- print ' | 2-byte sequence start'
- if $value == 0xC2;
- print ' | '
- if $value == 0xD0;
- }
- elsif ($value <= 0xEF) {
- print ' | 3-byte sequence start'
- if $value == 0xE0;
- }
- elsif ($value <= 0xF4) {
- print ' | 4-byte sequence'
- if $value == 0xF0;
- }
- elsif ($value <= 0xF7) {
- print ' | (Overflow)'
- if $value == 0xF5;
- }
- elsif ($value <= 0xFB) {
- print ' | 5-byte'
- if $value == 0xF8;
- }
- elsif ($value <= 0xFD) {
- print ' | 6-byte'
- if $value == 0xFC;
- }
- elsif ($value <= 0xFF) {
- print ' | Invalid'
- if $value == 0xFE;
- }
- else {
- print "\n".' | ?';
+my $NOCHAR = chr 0xFFFD;
+
+sub range_cell {
+ my ($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 - $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 = $separate && 'â¦';
+ }
+ else {
+ # minority on next row
+ push @next, $separate && '"', $title || $name;
+ }
+ $table->{$offset + $colsize*$rest} //= \@next;
+ $class .= ' joind';
+ }
+ $len = $rest;
+ }
+ elsif (my $rows = int($len / $cols)) {
+ # multiple full rows
+ my $rowsize = $colsize * $cols;
+ if ($len -= $rows * $cols) {
+ # partial row remains
+ $table->{$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 .= sprintf ' colspan=%d', $len unless $len == 1;
+ $attr .= $1 if $class and $class =~ s/( \w+="[^"]*")//;
+ $attr .= sprintf ' class="%s"', $class if $class;
+ $attr .= sprintf ' title="%s"', EscapeHTML($title) if $title;
+ return " | $name\n";
}
-print "\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; # row divide for column digits
+ $rowdiv = 1 if $rowdiv != $cols * $colsize; # divide only if all columns are matched
+ my $offset = $row->{startpoint} * $colsize || 0;
-my @nibble = (0..9, 'A'..'F');
-for my $tablenum (0 .. $#tables) {
- print '';
- printf '%s', $request[$tablenum] eq 'utf-8-strict' ? 'UTF-8' : $request[$tablenum];
- print '';
+ printf '', !$row->{cell} && ' charmap';
+ my $title = $row->{set};
+ $title .= " "
+ for $row->{parent} || ();
+ printf '%s', $title;
+ print '' x ($cols + 1);
for my $section (qw{thead}) {
- print "<$section>â±";
- print ' | ', $_ for @nibble;
+ print "<$section> |
---|
", $rowdiv == 1 ? '+' : 'â±';
+ printf ' | %0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
print "\n";
}
+
print ' | ';
- for my $msb (0 .. (length($tables[$tablenum]) || 256) - 1 >> 4) {
- printf '%X', $msb;
- for my $lsb (0 .. $#nibble) {
- if ($request[$tablenum] eq 'utf-8-strict') {
- printcell_utf8(($msb<<4) + $lsb);
- next;
+ while ($offset <= $row->{endpoint} * $colsize) {
+ if ($row->{skip}->{$offset}) {
+ $offset += $cols * $colsize;
+ next;
+ }
+
+ print ' |
---|
';
+ if (defined $row->{skip}->{$offset}) {
+ print 'â®';
+ }
+ else {
+ if (my $rowmod = $offset % $rowdiv) {
+ # offset in column units
+ printf '+%X', $rowmod;
}
- elsif ($request[$tablenum] eq 'Internal') {
- printcell_unicode(($msb<<4) + $lsb);
- next;
+ else {
+ # divided row offset
+ printf '%X', ($offset + $row->{offset}) / $rowdiv;
}
+ }
+ say '';
- my $glyph = substr $tables[$tablenum], ($msb<<4) + $lsb, 1;
- if ($glyph eq $NOCHAR) {
- print ' | ';
+ for (1 .. $cols) {
+ if ($row->{cell}) {
+ print range_cell($row, $offset);
next;
}
- my $info = [ord $glyph];
- if (defined (my $mnem = $di{ord $glyph})) {
- $info = $diinfo->{$mnem};
- }
- my ($codepoint, $name, $prop, $script, $string) = @$info;
- $glyph = quote($string || $glyph);
- my $desc = sprintf 'U+%04X%s', $codepoint, $name && " ($name)";
- my @class = ('X', grep {$_} $prop, $script);
+ my $cp = $offset + $row->{offset};
+ my $glyph = ref $row->{table} eq 'ARRAY' ? $row->{table}->[$offset] :
+ substr $row->{table}, $offset, 1;
+ my ($cell, $name, $class) = !defined $glyph || $glyph eq $NOCHAR ? () :
+ $glyphs->glyph_html($glyph);
- $glyph = "$glyph" if $prop eq 'Zs';
+ if ($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}++;
+ }
- printf "\n".' | %s',
- join(' ', @class), quote($desc), $glyph;
+ say sprintf $class ? ' | %s' : ' | ',
+ $name, $class, $cell;
+ }
+ continue {
+ $offset += $colsize;
}
- print "\n";
}
- print " |
---|
\n";
+ say ' ';
}
-print " \n";
-
:>
-
+
+ <: if ($mode) { :>
+ unicode
+ | inherited
+ | existing
+ | original
+ | unassigned
+<: } else { :>
+ | control
+ | whitespace
+ | diacritic
+ | punctuation
+ | symbol
+ | numeric
+ | greek
+ | aramaic
+ | syllabic
+ african
+ | japanese
+ | cjk
+ | chinese
+ |
+ | alphabetic
+ |
+
+
+ unicode 7.0
+ | proposed
+ | deprecated
+ | unassigned
+ | invalid
+<: } :> |
+
-
| |