Ignore:
Timestamp:
2003-10-22T13:28:35+13:00 (21 years ago)
Author:
jrm21
Message:

convert non-standard entities as used by Microsoft code pages to the
correct unicode values.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/gsdl/perllib/ghtml.pm

    r3181 r5709  
    151151    # a numeric entity
    152152    if ($entity =~ /^\#0*(\d+)/) {
    153     return &unicode::unicode2utf8([$1]);
     153    my $code=$1;
     154    # non-standard Microsoft breakage, as usual
     155    if ($code < 0x9f) { # code page 1252 uses reserved bytes
     156        if ($code == 0x91) {$code=0x2018} # 145 = single left quote
     157        elsif ($code == 0x92) {$code=0x2019} # 146 = single right quote
     158        elsif ($code == 0x93) {$code=0x201c} # 147 = double left quote
     159        elsif ($code == 0x94) {$code=0x201d} # 148 = double right quote
     160        # ...
     161    }
     162    return &unicode::unicode2utf8([$code]);
    154163    }
    155164   
Note: See TracChangeset for help on using the changeset viewer.