Izak Burger added the comment: Hi again,
I noted that the detector that is deployed on the main tracker, for ensuring that text/html content types are not allowed, was not on the meta tracker. I've copied it there. I don't think this is enough though. Consider for a second this command, just to illustrate: (wget -q -S -O- http://psf.upfronthosting.co.za/roundup/meta/file291/doc7.html 3>&1 1>&2- 2>&3- ) 2> /dev/null | grep Content-Type It looks terrible but I just swapped stdout and stderr so I can grep through the headers. It appears that once roundup finds the document it is looking for, the rest of the traversal string is ignored, except that it causes the Content-Type header to be set to "html". This is not a valid content type, and will cause some versions of IE to default to html, and some other browsers to look at the extension (html), so this will be rendered as html in almost all browsers. The fix might be as simple as using "text/plain" instead of "html", wherever this default happens to be. regards, Izak ---------- status: unread -> chatting _______________________________________________________ PSF Meta Tracker <metatrac...@psf.upfronthosting.co.za> <http://psf.upfronthosting.co.za/roundup/meta/issue505> _______________________________________________________ _______________________________________________ Tracker-discuss mailing list Tracker-discuss@python.org http://mail.python.org/mailman/listinfo/tracker-discuss
