charset: common cell formatting for utf8 schema

Only to simplify data, keeps fixed table layout.

diff --git a/charset-utf8.inc.pl b/charset-utf8.inc.pl
index 0d6a87d2eae3fea76b5bf749e52eef1c5ce3e377..a60dc30969595e85809bf741bdbe0b1c3400ad54 100644 (file)
--- a/charset-utf8.inc.pl
+++ b/charset-utf8.inc.pl
@@ -1,29 +1,15 @@
 use utf8;
-my %utf8byte = (
-       0x00, '<td rowspan="8" colspan="16" class="X l4 u-ascii"'.
-                         ' title="U+0000 – U+007F">single byte ASCII',
-       0x80, '<td rowspan="4" colspan="16" class="X l3"'.
-                         '>multi-byte continuation',
-       0xC0, '<td colspan="2" class="X l1" style="border-right:none; border-bottom:none"'.
-                         ' title="U+0000 – U+007F">(overl.)',
-       0xC2, '<td rowspan="2" colspan="14" class="X l2 u-bmp" style="border-left:none"'.
-                         ' title="U+0080 – U+03FF">2-byte sequence start',
-       0xD0, '<td rowspan="1" colspan="16" class="X l2 u-bmp" style="border-top:none"'.
-                         ' title="U+0400 – U+07FF">',
-       0xE0, '<td colspan="16" class="X l2 u-bmp"'.
-                         ' title="U+0800 – U+FFFF">3-byte sequence start',
-       0xF0, '<td colspan="5" class="X l2" style="border-right:none"'.
-                         ' title="U+1·0000 – U+10·FFFF">4-byte sequence',
-       0xF5, '<td colspan="3" class="X l1" style="border-left:none"'.
-                         ' title="U+11·0000 – U+1FF·FFFF">(overflow)',
-       0xF8, '<td colspan="4" class="X l1"'.
-                         ' title="U+200·0000 – U+3FFF·FFFF">5-byte',
-       0xFC, '<td colspan="2" class="X l1"'.
-                         ' title="U+4000·0000 – 7FFFF·FFFF">6-byte',
-       0xFE, '<td colspan="2" class="di-invalid">invalid',
-);
-
-sub {
-       return defined $utf8byte{$_[0]} ? $utf8byte{$_[0]} : ();
-}
-
++{
+       0x00 => [0x80, 'X l4 u-ascii',     'single byte ASCII',     'U+0000 – U+007F'],
+       0x80 => [0x40, 'X l3',             'multi-byte continuation'],
+       0xC0 => [0x02, 'X l1 joinr joind', '(overl.)',              'U+0000 – U+007F'],
+       0xC2 => [0x0E, 'X l2 u-bmp joinl rowspan="2"',
+                                          '2-byte sequence start', 'U+0080 – U+03FF'],
+       0xD0 => [0x10, 'X l2 u-bmp joinu', '',                      'U+0400 – U+07FF'],
+       0xE0 => [0x10, 'X l2 u-bmp',       '3-byte sequence start', 'U+0800 – U+FFFF'],
+       0xF0 => [0x05, 'X l2 joinr',       '4-byte sequence',       'U+1·0000 – U+10·FFFF'],
+       0xF5 => [0x03, 'X l1 joinl',       '(overflow)',            'U+11·0000 – U+1FF·FFFF'],
+       0xF8 => [0x04, 'X l1',             '5-byte',                'U+200·0000 – U+3FFF·FFFF'],
+       0xFC => [0x02, 'X l1',             '6-byte',                'U+4000·0000 – 7FFFF·FFFF'],
+       0xFE => [0x02, 'di-invalid',       'invalid'],
+};

diff --git a/charset.plp b/charset.plp
index 15183cfb479eae12889f06cfd41d5512b3be4c62..557c66fce9cbb448da4d8df9414e9d4560261d97 100644 (file)
--- a/charset.plp
+++ b/charset.plp
@@ -134,6 +134,7 @@ sub range_cell {
        }
 
        $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 $title;
        return "<td$attr>$name";