Wide character in %s

(W utf8) Perl met a wide character (>255) when it wasn't expecting one. This warning is by default on for I/O (like print). The easiest way to quiet this warning is simply to add the :utf8 layer to the output, e.g. binmode STDOUT, ':utf8'. Another way to turn off the warning is to add no warnings 'utf8'; but that is often closer to cheating. In general, you are supposed to explicitly mark the filehandle with an encoding, see the open manpage and binmode in the perlfunc manpage.

But after all, it is just a warning, and Perl is most likely doing "The Right Thing", despite complaining about it. This sort of warning about wide characters is Perl's way of saying "if you weren't already aware of this condition, you probably should be, just in case it might alter some assumptions you may have been making about your data..."

Interestingly the script works fine with ActiveStates's 5.8.1 build (no warnings whatsoever), so AFAIK the problem is limited to the Linux build on the production server. Note that the latter does not merely throw the typical "wide character" warnings -- the script terminates with error (see below) even if the "no warnings" pramata is issued.

Since I am hardpressed to come up with a working minimal code, below is some pseudocode.

use LWP::UserAgent;
use HTML::Parser;
use Encode;

# Grab a HTML document from the Net.
# Extract links from HTML document (legacy encoding, in this case Big5
+).
# Decode linked strings
# In my original post I used from_to($linktext, big5, utf-8)'s 
# in-place conversion.  Same error, though.

$linktext = decode( 'big5-eten', $linktext, Encode::FB_HTMLCREF );

# Test linked texts against a list of strings

#Loop of linked texts

foreach ( @keywords ) {
    local $_ = decode('utf-8', $_); # Force utf-8 flag
    if ( $linktext =~ /$_/ ) {
        # Do something
    }
}

# Typically the initial iterations appear OK but the loop never comple
+tes
# before the Wide Character error shows up.

# End of loop
[download]

I'm at a loss as to why the two 5.8.1 builds would behave differently even as 5.8.0 and 5.8.6 work. I also have no idea whether a workaround is possible.