$paramsto multiple charsets.", + ); + tabinput($_) 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 = { - 'cp437' => ['cp850' => 0, 'ascii' => '0-31+128'], # ascii range overridden later - 'gsm0338' => ['ascii' => '0-127'], - 'dingbats' => ['' => '32-127+160'], - 'MacDingbats' => ['dingbats' => '128-159'], - 'AdobeZdingbat'=> ['MacDingbats' => '128-159'], # should be identical but maps to private use - 'symbol' => ['' => '32-127+160'], - 'AdobeSymbol' => ['symbol' => '32-127+160', '' => '32-127+160'], # minor differences, irrelevant except for different 'â¬' - 'wingdings' => ['' => '32'], - 'wingdings2' => ['' => '32'], - 'wingdings3' => ['' => '32'], - 'webdings' => ['' => '32'], - - 'iso-8859-2' => ['iso-8859-1' => '160'], - 'iso-8859-3' => ['iso-8859-1' => '160'], #TODO: also apply to iso-8859-9 - 'iso-8859-4' => ['iso-8859-2' => '160'], - 'iso-8859-5' => ['iso-8859-1' => '160'], - 'iso-8859-6' => ['cp1256' => '128', 'iso-8859-1' => '160'], - 'iso-8859-7' => ['iso-8859-1' => '160'], - 'iso-8859-8' => ['iso-8859-1' => '160'], - 'iso-8859-9' => ['iso-8859-1' => '208-223+240'], - 'iso-8859-10' => ['iso-8859-4' => '160'], - 'iso-8859-11' => ['iso-8859-1' => '160'], - 'iso-8859-13' => ['iso-8859-4' => '160'], - 'iso-8859-14' => ['iso-8859-1' => '160'], - 'iso-8859-15' => ['iso-8859-1' => '160-191'], - 'iso-8859-16' => ['iso-8859-2' => '160'], - 'hp-roman8' => ['iso-8859-1' => '160'], - - 'cp1252' => ['iso-8859-1' => '128-159'], - 'cp1250' => ['iso-8859-2' => '128-191', 'cp1252' => '128'], - 'cp1254' => ['iso-8859-9' => '128-159', 'cp1252' => '128-159+208'], - 'cp874' => ['iso-8859-11' => '128-159', 'cp1252' => '128'], # windows-874 actually cp1162 - 'cp1257' => ['iso-8859-13' => '128-159+255', 'cp1252' => '128'], - 'cp1251' => ['cp1252' => '128'], - 'cp1253' => ['cp1252' => '128'], - 'cp1255' => ['iso-8859-8' => '128-223', 'cp1252' => '128'], - 'cp1256' => ['cp1252' => '128'], - 'cp1258' => ['cp1252' => '128-159+192'], - - 'cp850' => ['cp437' => '144'], - 'cp860' => ['cp437' => '128-175'], - 'cp861' => ['cp865' => '128-175'], - 'cp863' => ['cp437' => '128-175'], - 'cp865' => ['cp437' => '144-175'], - 'cp852' => ['cp850' => '128', 'cp437' => '128'], - 'cp857' => ['cp850' => '128-175+208-239', 'cp437' => '128'], - 'cp775' => ['cp850' => '128'], # partial cp437 - 'cp866' => ['cp437' => '128-175+224'], - 'cp855' => ['cp437' => '128'], - 'cp1006' => ['iso-8859-6' => '160', 'cp437' => '128'], - 'cp737' => ['cp437' => '128-175+224'], - 'cp869' => ['cp437' => '128'], - 'cp862' => ['cp437' => '128-159'], - 'cp864' => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants - - 'koi8-u' => ['koi8-r' => '144-191'], - 'koi8-f' => ['koi8-u' => '144-191'], - - 'MacRomanian' => ['MacRoman' => '160-191+208-223'], - 'MacRumanian' => ['MacRomanian' => '160-191+208-223', 'MacRoman' => '160-191+208-223'], - 'MacCroatian' => ['MacRoman' => '160'], - 'MacCentralEurRoman' => ['MacRoman' => '128'], - 'MacIcelandic'=> ['MacRoman' => '160-175+208-239'], - 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned - 'MacSami' => ['MacIcelandic' => '144-191+208-223+240', 'MacRoman' => '144'], - 'MacGreek' => ['MacRoman' => '128'], - 'MacCyrillic' => ['MacRoman' => '128'], - 'MacHebrew' => ['iso-8859-8' => '128', 'MacRoman' => '128-143+160'], # partial ascii - 'MacArabic' => ['iso-8859-6' => '128', 'cp864' => '128', 'MacRoman' => '128'], #TODO: multiple parents - 'MacFarsi' => ['MacArabic' => '176-191', 'MacRoman' => '128'], - - 'cp37' => ['posix-bc' => '0'], - 'posix-bc' => ['cp1047' => '64'], - 'cp500' => ['cp37' => '64-95+176-191'], - 'cp1047' => ['cp37' => '16-47+80-95+160-191'], - 'cp1026' => ['cp37' => '64'], - 'cp875' => ['cp37' => '48'], - - }; - - my @parents = @{ $INHERIT->{$input} || [] }; + my @parents = @{ $charset->{inherit} || [] }; if (my ($parent, $part) = pairfirst { defined $visible->{$a} } @parents) { $row{parent} = $parent; $params = $part; - $params = 128 unless $visible->{$parent} + $params = 80 unless $visible->{$parent} or ($input eq 'MacCroatian' and defined $visible->{MacRomanian}); } elsif (defined $visible->{ascii}) { $row{parent} = $parents[0]; - $params = $parents[1] // 128; - $params = 128 if $params >= 128; # ascii offset at most + $params = $parents[1] // 80; + $params = 80 if hex $params >= 0x80; # ascii offset at most } elsif (@parents) { $row{parent} = $parents[0]; - $params = $parents[1] if $parents[1] == 0; # apply ascii end + $params = $parents[1] if hex $parents[1] == 0; # apply ascii end } $visible->{$_} //= 0 for $row{parent} || (); } @@ -238,66 +117,36 @@ sub tabinput { elsif ($param =~ m{ \A cols = (\d+) \z }x) { $row{cols} = $1; } - elsif ($param =~ m{ \A (?
$paramfor charset $input"); } } - if ($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 =~ m{ \A u ([0-9a-f]+) (?:-([0-9a-f]+))? \z }ix) { - 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; - } - elsif (lc $input eq 'uu') { - $row{set} = 'Unicode planes'; - $row{cell} = do 'charset-ucplanes.inc.pl' - or Alert('Table data could not be read', $@ || $!); - $row{endpoint} = 1023 * $row{cell}->{colsize}; - } - elsif (lc $input eq 'u') { - $row{cell} = do 'charset-unicode.inc.pl' - or Alert('Table data could not be read', $@ || $!); - - $row{endpoint} ||= 8191; - $row{endpoint} *= $row{cell}->{colsize}; - $row{startpoint} = $row{cell}->{colsize} * $row{offset}; - $row{offset} = 0; - $row{set} = 'Unicode ' . ( - $row{startpoint} < 0x10000 && $row{endpoint} < 0x10000 ? 'BMP' : - $row{startpoint} >= 0x10000 && $row{endpoint} < 0x20000 ? 'SMP' : - 'allocations' - ); - } - elsif ($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} = 255; + if (defined $row{table} or defined $row{cell}) { + $row{set} //= $input; } elsif ($row{set} = Encode::resolve_alias($input)) { - $row{endpoint} ||= 255; - if ($row{set} eq 'MacHebrew' or $row{set} eq 'MacThai') { + $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} @@ -308,43 +157,48 @@ sub tabinput { # 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
$inputunknown"); + return; + } - if ($row{set} eq 'cp437') { - if ($row{offset} <= 0xED and $row{endpoint} >= 0xED - $row{offset}) { - # replace phi glyph - substr($row{table}, 0xED - $row{offset}, 1) = 'Ï'; - } - if ($row{offset} < 0x20) { - # replace control characters by visible variants - my $sub = substr 'ââºâ»â¥â¦â£â â¢ââââââªâ«â¼âºâââ¼Â¶Â§â¬â¨âââââââ²â¼', $row{offset}; - substr($row{table}, 0, length $sub) = $sub; + 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 <= $row{endpoint}; + next; } - } - elsif ($row{set} eq 'symbol') { - if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60 - $row{offset}) { - # replace radical extender by closest unicode equivalent - substr($row{table}, 0x60 - $row{offset}, 1) = 'â'; + + 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; + $offset = 0; } - if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF - $row{offset}) { - substr($row{table}, 0xBD - $row{offset}, 2) = 'ââ¯'; # arrow extenders - substr($row{table}, 0xD2 - $row{offset}, 3) = '®©â¢'; # serif variants - substr($row{table}, 0xE0 - $row{offset}, 1) = 'â'; # replace lookalike, should match AdobeSymbol - substr($row{table}, 0xE2 - $row{offset}, 3) = '®©â¢'; # sans-serif variants - substr($row{table}, 0xE6 - $row{offset}, 10) = 'ââââ¡â¢â£â§â¨â©âª'; - substr($row{table}, 0xF0 - $row{offset}, 1) = 'â¬'; - substr($row{table}, 0xF4 - $row{offset}, 11) = 'â®â¡âââ â¤â¥â¦â«â¬â'; + + if ((my $excess = $row{endpoint} - $offset - $length + 1) < 0) { + $excess > -$length or next; + # trim rightmost part to prevent overflow + substr($sub, $excess) = ''; + $length += $excess; } - } - $visible->{ascii} = # assume common base - $visible->{ $row{set} } = 1; - } - else { - Alert("Encoding
$inputunknown"); - return; + substr($row{table}, $offset, $length) = $sub; + } } + push @request, \%row; + $visible->{ $row{set} } = 1 if $row{table}; } tabinput($_) for @tablist; @@ -362,7 +216,7 @@ 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 @@ -422,7 +276,7 @@ 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 = $row->{startpoint} || 0; + my $offset = $row->{startpoint} * $colsize || 0; 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; @@ -467,10 +322,10 @@ 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' : @@ -497,7 +352,7 @@ for my $row (@request) {
---|
unicode | inherited | existing @@ -535,7 +390,7 @@ for my $row (@request) { |
unicode 7.0 + | unicode 10.0 | proposed | deprecated | unassigned |