charset: generic glyph replacement from metadata

[sheet.git] / charset.plp

diff --git a/charset.plp b/charset.plp
index ed495fe2ae89352fa10641847cd2d19d3436378c..7f19bc183e6fb02abd1d91622041cf7f75479905 100644 (file)
--- a/charset.plp
+++ b/charset.plp
@@ -164,33 +164,6 @@ sub tabinput {
                        $row{table} = Encode::decode($row{set}, pack 'C*', $row{offset} .. $row{endpoint});
                }
 
-               if ($row{set} eq 'cp437') {
-                       if ($row{offset} <= 0xED and $row{endpoint} >= 0xED) {
-                               # replace phi glyph
-                               substr($row{table}, 0xED - $row{offset}, 1) = 'ϕ';
-                       }
-                       if ($row{offset} < 0x20) {
-                               # replace control characters by visible variants
-                               my $sub = substr ' ☺☻♥♦♣♠•◘○◙♂♀♪♫☼►◄↕‼¶§▬↨↑↓→←∟↔▲▼', $row{offset};
-                               substr($row{table}, 0, length $sub) = $sub;
-                       }
-               }
-               elsif ($row{set} eq 'symbol') {
-                       if ($row{offset} <= 0x60 and $row{endpoint} >= 0x60) {
-                               # replace radical extender by closest unicode equivalent
-                               substr($row{table}, 0x60 - $row{offset},  1) = '│';
-                       }
-                       if ($row{offset} <= 0xBD and $row{endpoint} >= 0xFF) {
-                               substr($row{table}, 0xBD - $row{offset},  2) = '⏐⎯'; # arrow extenders
-                               substr($row{table}, 0xD2 - $row{offset},  3) = '®©™'; # serif variants
-                               substr($row{table}, 0xE0 - $row{offset},  1) = '◊'; # replace lookalike, should match AdobeSymbol
-                               substr($row{table}, 0xE2 - $row{offset},  3) = '®©™'; # sans-serif variants
-                               substr($row{table}, 0xE6 - $row{offset}, 10) = '⎛⎜⎝⎡⎢⎣⎧⎨⎩⎪';
-                               substr($row{table}, 0xF0 - $row{offset},  1) = '€';
-                               substr($row{table}, 0xF4 - $row{offset}, 11) = '⎮⌡⎞⎟⎠⎤⎥⎦⎫⎬⎭';
-                       }
-               }
-
                $row{endpoint} -= $row{offset};
 
                $visible->{ascii} =  # assume common base
@@ -200,6 +173,32 @@ sub tabinput {
                Alert("Encoding <q>$input</q> unknown");
                return;
        }
+
+       if (my $replace = $charset->{replace}) {
+               while (my ($offset, $sub) = each %{$replace}) {
+                       $offset -= $row{offset};
+
+                       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) = $sub;
+               }
+       }
+
        push @request, \%row;
 }
 tabinput($_) for @tablist;