Niko Tyni <[EMAIL PROTECTED]> writes: > I see. The Perl internal encoding is UTF-8, but there are ways to get > invalid UTF-8 in there, for example by using :utf8 on binary input. > This invalid UTF-8 will then be output as-is with if :utf8 is set on > output. > > I can't really think of a case where setting :encoding(utf-8) on output > does the right thing but :utf8 doesn't. It does turn the output into valid > UTF-8, but do you have an example where the content is not gibberish?
I can't easily duplicate what I was seeing now, but I was getting output that was not UTF-8 while using that output encoding in combination with Pod::Simple. I'm not quite sure what was going on. It's possible that I had made some mistake in the middle of my testing, though. > How about providing your own parse_from_file() wrapper in Pod::Man that > knows about the utf8 option, does the open() and then sets the binmode? I guess I could do that, but I think I disagree with this: > I don't think there's any need to touch the filehandles of people using > parse_file(). I would prefer not to touch the filehandle, but I don't think it's acceptable to say that if you're using the utf8 option, you still have to set up output encodings yourself. Maybe I'm overreacting to how difficult I found this area of Perl to understand, but I'd really rather that Pod::Man and Pod::Text do the right thing without requiring people understand Perl's very strange Unicode handling. Pod::Text also has to do something more complex in order to preserve its traditional encoding agnosticism if utf8 is not given. -- Russ Allbery ([EMAIL PROTECTED]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
