> The file is already in UTF-8, otherwise it wouldn't display properly > in Firefox or IE. The problem is either your display or perl doesn't > know that the file is in UTF-8. > > The first step is make sure Perl knows it is working with UTF-8. Add > > export PERL_UNICODE=SDL > > to your .profile, .bashrc, or whatever you use for your profile. > Logout and log back in. This tells perl to use UTF-8 for STDIN, > STDOUT, and STDERR (the S), input and output streams (the D), and all > of it dependent on locale (the L). The next thing to check is the > value in your LANG environment variable. It should be something like > en_US.UTF-8. If you are still having problems check to see if your > terminal is expecting something other than UTF-8 (this is highly > dependent on the terminal, so you will need to tell us what terminal > you are using).
Thank you, this was very helpful. I added those lines to my .bashrc and set my terminal to UTF-8, and my text editor to UTF-8, and I see those characters display correctly now. I even used my editor to save the CSV encoded as Windows-1250 and got it to open and display correctly in Microsoft Excel on a Windows box. Incidentally, I still get a warning with -w: Parsing of undecoded UTF-8 will give garbage when decoding entities at /usr/share/perl5/LWP/Protocol.pm line 114. I am using Perl 5.8.8 on linux. I might be doing something wrong, or perhaps upgrading to Perl 5.10.0, which I plan to do in the next week anyway, will fix this. I will look into it further. Thanks again, John
