charset: realsize option to keep utf-8 table unabbreviated

[sheet.git] / charset.plp

diff --git a/charset.plp b/charset.plp
index 54e3b45761017930cf0b466d7d0c3295d995e7f5..e82e3d7b25f07412a26a22805f653a702bb04664 100644 (file)
--- a/charset.plp
+++ b/charset.plp
@@ -77,8 +77,8 @@ sub tabinput {
        my $input = shift or return;
 
        state $ALIAS = {
-               default    => [qw( unicode+0-639 utf-8 iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
-               uc         => [qw( U++ unicode+0-4095 unicode+4096-6319  unicode+6320-8191 )],
+               default    => [qw( unicode+0-639 utf-8+realsize iso-8859-1 iso-8859-15 cp1252 cp437 cp850 )],
+               uc         => [qw( U+cols=32 unicode+0-4095 unicode+4096-6319  unicode+6320-8191 )],
                us         => [qw( cp437 cp863 gsm0338 AdobeStandardEncoding )],
                ebcdic     => [qw( cp37 cp500 cp1047 posix-bc cp1026 cp875 )],
                iso        => [map {"iso-8859-$_"} 1 .. 11, 13 .. 16],
@@ -126,7 +126,6 @@ sub tabinput {
 
        state $visible = {};  # all present tables
        my %row = (offset => 0, cols => 16);
-       my $endpoint = 255;
        my $params = $input =~ s/[+](.*)\z// ? $1 : undef;
 
        if (not defined $params) {
@@ -177,16 +176,16 @@ sub tabinput {
                        'cp862'       => ['cp437' => '128-159'],
                        'cp864'       => ['MacArabic' => '128', 'iso-8859-6' => '128', 'cp437' => '128'], #TODO: compare form variants
 
-                       'koi8-u'      => ['koi8-r' => 128],
-                       'koi8-f'      => ['koi8-u' => 128],
+                       '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'], #TODO: gaps at C/E
+                       'MacIcelandic'=> ['MacRoman' => '160-175+208-239'],
                        'MacTurkish'  => ['MacRoman' => '208-223'], # F5 is unassigned
-                       'MacSami'     => ['MacIcelandic' => '144', 'MacRoman' => '144'],
+                       '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
@@ -196,7 +195,7 @@ sub tabinput {
                        'cp37'        => ['posix-bc' => '0'],
                        'posix-bc'    => ['cp1047' => '64'],
                        'cp500'       => ['cp37' => '64-95+176-191'],
-                       'cp1047'      => ['cp37' => '16-95+160-191'], #TODO: gap at 3/4
+                       'cp1047'      => ['cp37' => '16-47+80-95+160-191'],
                        'cp1026'      => ['cp37' => '64'],
                        'cp875'       => ['cp37' => '48'],
 
@@ -222,28 +221,28 @@ sub tabinput {
                $visible->{$_} //= 0 for $row{parent} || ();
        }
 
-       if (length $params) {
-               # manual option to double table width
-               $row{cols} *= 2 if $params =~ s/[+]\z//;
-       }
-
-       if (length $params) {
-               $params =~ m{
-                       \A (?<offset> \d+)
-                       (?: (?: [-] (?<stop> \d+) )? (?: [+] (?<restart> \d+) ) )?
-                       (?: [-] (?<endpoint> \d+) )? \z
-               }x or Alert([
-                       "Unknown range parameters for $input",
-                       "<q>$params</q> is not in format start(-stop)(+restart(-end))",
-               ]);
-
-               $row{offset} = $+{offset};
-               $endpoint = $+{endpoint} if $+{endpoint};
-               if (my $restart = $+{restart}) {
-                       my $skip = int(($+{stop} || $row{offset}) / $row{cols});
-                       for ($skip + 1 .. ($restart / $row{cols}) - 1) {
-                               $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
+       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 (?<offset> \d+) (?: [-] (?<endpoint> \d+) )? \z }x) {
+                       if (defined $row{endpoint}) {
+                               # extend earlier range
+                               my $skip = int(($row{endpoint} || $row{offset}) / $row{cols});
+                               for ($skip + 1 .. ($+{offset} / $row{cols}) - 1) {
+                                       $row{skip}->{ $_ * $row{cols} - $row{offset} }++;
+                               }
+                       }
+                       else {
+                               $row{offset} = $+{offset};
                        }
+                       $row{endpoint} = $+{endpoint} || 0;
+               }
+               else {
+                       Alert("Unknown option <q>$param</q> for charset $input");
                }
        }
 
@@ -267,7 +266,8 @@ sub tabinput {
                        $row{cell} = do 'charset-unicode.inc.pl'
                                or Alert('Table data could not be read', $@ || $!);
 
-                       $row{endpoint  } = $row{cell}->{colsize} * ($endpoint || 8191);
+                       $row{endpoint} ||= 8191;
+                       $row{endpoint}  *= $row{cell}->{colsize};
                        $row{startpoint} = $row{cell}->{colsize} * $row{offset};
                        $row{offset} = 0;
                        $row{set} = 'Unicode ' . (
@@ -283,18 +283,19 @@ sub tabinput {
                        $row{endpoint} = 255;
                }
                else {
+                       $row{endpoint} ||= 255;
                        if ($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} .. $endpoint
+                                       map { Encode::decode($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} .. $endpoint);
+                               $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
                        }
-                       $row{endpoint} = $endpoint - $row{offset};
+                       $row{endpoint} -= $row{offset};
 
                        if ($row{set} eq 'cp437' and !$row{offset}) {
                                substr($row{table}, 237, 1) = pack 'U*', 0x3D5; # phi sign
@@ -363,13 +364,15 @@ 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;