[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
Neil Williams wrote: > > I still suspect an incomplete conversion of the environment to UTF8. I suspect that it is down to "vim" ;) I'm fairly sure the issue relates to the mode "vim" chooses automatically when editing the file. It tries to do the right thing, which meant when I had a malformed "utf-8" character in the file, it didn't trigger it's utf-8 mode. GIGO - but it is quite hard to tell when the garbage looks correct in about 50% of your tools, and another tool is silently switching modes based on that garbage. By the time we ran hexdump, I'd fixed it all up. Basically the program works fine if you don't start with a corrupt £ sign in it, and yes all your characters display fine in Thunderbird. Several issues this end. 1) Dojo doesn't pick up the encoding from a form, so doing asynchronous form submissions with the Dojo toolkit means you have to explicitly set the encoding in the Javascript. http://www.get-the-answer.info/pdf/1.pdf 2) The Perl Template Toolkit filters (well html_entity) appears to need some explicit hinting as to the decoding to use. But I'm still working on understanding exactly what is going on there. The nice folk on #catalyst seem to understand it - well some of them ;) Basically TT doesn't assume UTF-8 as far as I can tell. I assume backward compatibility. Our own filters are fine -- hohum - tomorrow... -- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/linux_adm/list-faq.html