X-Git-Url: http://git.shiar.nl/sheet.git/blobdiff_plain/ccf2f210912568cc5c5188a821de5a48368021a5..ee2f769329b25592f205cc0d3b9667c49af07a2d:/charset.plp
diff --git a/charset.plp b/charset.plp
index 92b918f..e2dac1a 100644
--- a/charset.plp
+++ b/charset.plp
@@ -24,25 +24,29 @@ my @tablist = split /[^\w-]+/, $Request || 'default';
use POSIX qw( ceil );
use Shiar_Sheet::FormatChar;
my $glyphs = Shiar_Sheet::FormatChar->new;
+my @request;
sub tabinput {
# generate character table(s)
my $input = shift or return;
state $ALIAS = {
- default => [qw(unicode- utf-8 iso-8859-1 -cp1252-- --iso-8859-15-- cp437 -cp850)],
- us => [qw( cp437 -cp863 AdobeStandardEncoding gsm0338-- )],
- ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )],
- westeur => [qw( iso-8859-1 --iso-8859-15-- -cp1252-- --iso-8859-14 -cp850 -MacRoman -nextstep --hp-roman8 )],
- centeur => [qw( iso-8859-2 --iso-8859-16 -cp1250 -cp852 -MacCentralEurRoman -MacCroatian -MacRomanian )], # MacRumanian only for DB
- turkish => [qw( iso-8859-3 -iso-8859-9 -cp857 -cp1254 -MacTurkish )],
- baltic => [qw( iso-8859-4 --iso-8859-13 -cp775 -cp1257 )],
- nordic => [qw( iso-8859-10 -cp865 --cp861-- -MacIcelandic --MacSami )],
- cyrillic => [qw( koi8-f -koi8-r- --koi8-u- -iso-8859-5 -cp1251 -MacCyrillic -cp855 -cp866 )], # MacUkrainian is broken
- arabic => [qw( iso-8859-6 --cp1006 -cp864 -cp1256 MacArabic )], # MacFarsi same as MacArabic?
- greek => [qw( iso-8859-7 -cp1253 -MacGreek -cp737 -cp869 )],
- hebrew => [qw( iso-8859-8 -cp1255 -MacHebrew -cp862 )],
- thai => [qw( iso-8859-11 -cp874-- -MacThai )],
+ default => [qw( unicode+0-192 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+ us => [qw( cp437 cp863 AdobeStandardEncoding gsm0338 )],
+ ebcdic => [qw( cp37 cp500 cp875 cp1026 cp1047 posix-bc )],
+ iso => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
+ dos => [qw( cp437 cp863 cp850 cp860 cp852 cp855 cp866 cp864 cp1006 cp737 cp869 cp862 cp862 cp857 cp865 cp861 cp874 cp775 )],
+
+ westeur => [qw( iso-8859-1 iso-8859-15 cp1252 iso-8859-14 cp850 hp-roman8 nextstep MacRoman )],
+ centeur => [qw( iso-8859-2 iso-8859-16 cp1250 cp852 MacRomanian MacCroatian MacCentralEurRoman )], # MacRumanian only for DB
+ turkish => [qw( iso-8859-9 iso-8859-3 cp1254 cp857 MacTurkish )],
+ baltic => [qw( iso-8859-4 iso-8859-13 cp1257 cp775 )],
+ nordic => [qw( iso-8859-10 cp865 cp861 MacIcelandic MacSami )],
+ cyrillic => [qw( koi8-f koi8-r koi8-u iso-8859-5 cp1251 MacCyrillic cp855 cp866 )], # MacUkrainian is broken
+ arabic => [qw( iso-8859-6 cp1006 cp864 cp1256 MacArabic )], # MacFarsi same as MacArabic?
+ greek => [qw( iso-8859-7 cp1253 MacGreek cp737 cp869 )],
+ hebrew => [qw( iso-8859-8 cp1255 MacHebrew cp862 )],
+ thai => [qw( iso-8859-11 cp874 MacThai )],
vietnamese => [qw( viscii cp1258 MacVietnamese )],
symbol => [qw( symbol dingbats MacDingbats AdobeZdingbat AdobeSymbol )],
# iso-code shorthand
@@ -59,26 +63,73 @@ sub tabinput {
11 => 'thai',
};
if (my $follow = $ALIAS->{$input}) {
- return map { tabinput($_) } ref $follow ? @{$follow} : $follow;
+ tabinput($_) for ref $follow ? @{$follow} : $follow;
+ return;
}
+ state $visible = {}; # all present tables
my %row = (offset => 0, cols => 16);
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} ? $row{offset} < 160 ? 159 : 191 : 127;
- }
- elsif ($input =~ s/-$//) {
- $endpoint = 192;
+ my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
+
+ if (not defined $params) {
+ use List::Util qw( first pairfirst pairs );
+
+ state $INHERIT = {
+ 'cp437' => ['cp850' => 0], # ascii range overridden later
+ 'gsm0338' => ['ascii' => '0-127'],
+
+ 'iso-8859-2' => ['iso-8859-1' => '160'],
+ 'iso-8859-3' => ['iso-8859-1' => '160'],
+ 'iso-8859-4' => ['iso-8859-2' => '160'],
+ 'iso-8859-5' => ['iso-8859-1' => '160'],
+ 'iso-8859-6' => ['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'],
+ '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'],
+
+ 'cp1252' => ['iso-8859-1' => '128-159'],
+ 'cp1250' => ['iso-8859-2' => '128-191'],
+ 'cp1254' => ['iso-8859-9' => '128-159'],
+ 'cp1257' => ['iso-8859-13' => '128-'],
+
+ 'cp850' => ['cp437' => '144'],
+ 'cp865' => ['cp437' => '144-176'],
+ 'cp852' => ['cp850' => '128', 'cp437' => '128'],
+ 'cp857' => ['cp850' => '128-239', 'cp437' => '128'],
+ 'cp775' => ['cp850' => '128'], # partial cp437
+
+ 'MacRomanian' => ['MacRoman' => '160'],
+ 'MacCroatian' => ['MacRoman' => '160'],
+ 'MacCentralEurRoman' => ['MacRoman' => '128'],
+ 'MacTurkish' => ['MacRoman' => '208-223'], # F5 is unassigned
+ };
+
+ my @parents = @{ $INHERIT->{$input} || [] };
+
+ if (my ($parent, $part) = pairfirst { $visible->{$a} } @parents) {
+ $row{parent} = $parent;
+ $params = $part;
+ }
+ elsif ($visible->{ascii}) {
+ $row{parent} = $parents[0] || 'ascii';
+ $params = $parents[1] || 128;
+ $params = 128 if $params >= 128; # ascii offset at most
+ }
+ elsif (@parents) {
+ $row{parent} = $parents[0];
+ }
}
- if ($row{offset}) {
- $row{setnote} = 'over cp437' if $input eq 'cp850';
- $row{setnote} = 'over iso-8859-1' if $input =~ /^iso-8859-|^cp125/;
+
+ if (defined $params and $params =~ m/^ (\d+) (-\d+)? /x) {
+ $row{offset} = $1;
+ $endpoint = -$2 if $2;
}
if ($input =~ /^U([0-9a-f]+)(?:-([0-9a-f]+))?/) {
@@ -86,10 +137,11 @@ sub tabinput {
my $end = $2 ? hex($2) << 4 : $start + 240;
$row{table} = join '', map { chr } $start .. $end+15;
utf8::upgrade($row{table}); # prevent latin1 output
+ $row{len} = length $row{table};
$row{set} = sprintf 'Unicode block U+%02Xxx', $start >> 8;
}
elsif ($input eq 'U') {
- $row{table} = ' ' x 1024;
+ $row{len} = 1024;
$row{set} = 'Unicode planes';
$row{cell} = do 'charset-ucplanes.inc.pl'
or Alert('Table data could not be read', $@ || $!);
@@ -97,23 +149,32 @@ sub tabinput {
}
elsif ($row{set} = Encode::resolve_alias($input)) {
if ($row{set} eq 'Internal') {
- $row{table} = ' ' x ($endpoint < 255 ? 640 : 8192);
+ $row{len} = $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{len} = 256;
$row{set} = 'UTF-8';
$row{cell} = do 'charset-utf8.inc.pl'
or Alert('Table data could not be read', $@ || $!);
}
else {
- $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $endpoint);
- # (~16x faster than decoding in loop;
- # substr strings is twice as fast as splitting to an array)
+ if ($row{set} eq 'MacHebrew') {
+ # 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{len} = $endpoint - $row{offset};
- if ($row{set} eq 'cp437') {
+ 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
@@ -122,15 +183,18 @@ sub tabinput {
2191 2193 2192 2190 221F 2194 25B2 25BC
);
}
+
+ $visible->{ascii} = # assume common base
+ $visible->{ $row{set} } = 1;
}
}
else {
Alert("Encoding $input
unknown");
return;
}
- return \%row;
+ push @request, \%row;
}
-my @request = map { tabinput($_) } @tablist;
+tabinput($_) for @tablist;
my $NOCHAR = chr 0xFFFD;
@@ -193,11 +257,12 @@ for my $row (@request) {
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 $endpoint = $offset + (length($row->{table}) || 256) * $colsize;
+ my $endpoint = $offset + $row->{len} * $colsize;
printf '
'; next; |