Last modified: 2012-04-16 09:15:36 UTC
When requesting a page such as <http://bits.wikimedia.org/fr.wikipedia.org/load.php?debug=true&lang=fr&modules=site&only=scripts> currently, the Content-Type is "text/javascript". It should be "text/javascript; charset=utf-8".
does this actually cause anyone problems?
Daniel Friesen helpfully pointed out that the URL itself demonstrates the problem.
Created attachment 8359 [details] Sets character set to UTF-8 within the HTTP header's Content-Type field. Here's a patch that will do the trick. I would have applied it, but I wanted to get some input on one concern. Should we really be assuming that the JavaScript and CSS content is UTF-8? Should we be detecting the encoding and using that? What happens when we are combining multiple encodings into a single response? Should we convert everything to UTF-8? I'm looking for some opinions, especially those based in experience with handling text encoding within MediaWiki.
Everything in MediaWiki is UTF-8 nowdays (including the source files). I'd say go for it.
Patched trunk in r85152.