\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 $input");
}
}
- if ($input =~ /^U([0-9a-fA-F]+)(?:-([0-9a-fA-F]+))?/) {
- 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} = $start % 256;
+ if ($charset->{setup}) {
+ eval { $charset->{setup}->(\%row) }
+ or Alert("Incomplete setup of $input", $@);
}
- elsif ($input eq 'U') {
- $row{set} = 'Unicode planes';
- $row{cell} = do 'charset-ucplanes.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
- $row{cols} *= 2;
- $row{endpoint} = 1023 * $row{cell}->{colsize};
+ $row{endpoint} ||= 0xFF;
+
+ if (defined $row{table} or defined $row{cell}) {
+ $row{set} //= $input;
}
elsif ($row{set} = Encode::resolve_alias($input)) {
- if ($row{set} eq 'Internal') {
- $row{set} = 'Unicode BMP';
- $row{cell} = do 'charset-unicode.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
- $row{endpoint} = ($endpoint || 8191) * $row{cell}->{colsize};
- }
- elsif ($row{set} eq 'utf-8-strict') {
- $row{set} = 'UTF-8';
- $row{cell} = do 'charset-utf8.inc.pl'
- or Alert('Table data could not be read', $@ || $!);
- $row{endpoint} = 255;
+ $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} .. $row{endpoint}
+ ];
}
else {
- if ($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} .. $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} .. $endpoint);
- }
- $row{endpoint} = $endpoint - $row{offset};
-
- if ($row{set} eq 'cp437' and !$row{offset}) {
- substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
- 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;
+ # ~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
}
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;
@@ -323,22 +216,23 @@ 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 && 'â¦';
}
else {
# minority on next row
- push @next, '"', $title || $name;
+ push @next, $separate && '"', $title || $name;
}
$table->{$offset + $colsize*$rest} //= \@next;
$class .= ' joind';
@@ -354,13 +248,15 @@ sub range_cell {
$class .= ' joind';
}
- # coalesce multiple rows
- while ($rows > 3) {
- $info->{skip}->{$offset += $rowsize}++;
- $rows--;
- }
- if ($rows > 2) {
- $info->{skip}->{$offset += $rowsize} = 0;
+ unless ($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;
@@ -371,7 +267,7 @@ sub range_cell {
$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";
+ return " | $name\n";
}
for my $row (@request) {
@@ -380,12 +276,12 @@ 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 = 0;
+ my $offset = $row->{startpoint} * $colsize || 0;
printf '', !$row->{cell} && ' charmap';
my $title = $row->{set};
$title .= " "
- for grep { $_ ne 'iso-8859-1' } $row->{parent} // ();
+ for $row->{parent} || ();
printf '%s', $title;
print '' x ($cols + 1);
for my $section (qw{thead}) {
@@ -393,8 +289,9 @@ for my $row (@request) {
printf '%0*X', $coldigits, $_ * $colsize for 0 .. $cols - 1;
print "\n";
}
+
print ' | ';
- while ($offset < $row->{endpoint}) {
+ while ($offset <= $row->{endpoint} * $colsize) {
if ($row->{skip}->{$offset}) {
$offset += $cols * $colsize;
next;
@@ -414,39 +311,39 @@ for my $row (@request) {
printf '%X', ($offset + $row->{offset}) / $rowdiv;
}
}
+ say '';
+
for (1 .. $cols) {
if ($row->{cell}) {
print range_cell($row, $offset);
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', $cp) ? 'l3' :
+ !$class ? undef :
$visible->{$glyph} ? 'l2' :
- 'l1',
- $glyphs->glyph_html($glyph);
+ 'l1'
+ );
$visible->{$glyph}++;
- next;
}
- print "\n".$glyphs->glyph_cell($glyph);
+ say sprintf $class ? ' | %s' : ' | ',
+ $name, $class, $cell;
}
continue {
$offset += $colsize;
}
- print "\n";
}
say ' | ';
}
@@ -455,7 +352,7 @@ for my $row (@request) {
- <: if (exists $get{compare}) { :>
+ <: if ($mode) { :>
unicode
| inherited
| existing
|
|