Hi, I've been pointed at you by Brad Marshall as the people maintaining and setting the direction for APT.
I currently look after a couple of archives in australia hosting debian and debian related archives and have run into a bit of an issue with apt. Basically apt doesn't have too much functionality in it (and probably for good reason) for some of the HTTP protocol implementations, but we'd really like to see it acquire some. The biggest single thing for us is that apt has no ability to understand a 302 Redirect. We are currently using rewrite rules in apache to point people towards a lightweight web server for "large object" downloads, i.e not web pages - iso images, zips, rpms, gz and bz2 files and we were hoping debs. Unfortunately apt breaks if we implement this - in fact i just realised it broken on gz files so we've had to turn that off too - because of its inability to follow the redirect. Is there any chance of getting apt to offer this functionality ? The second query we have is getting apt to have some authentication mechanism. There seems to be no way to get apt to authenticate itself directly against a web server (unlike wget or lynx). I realise that it should be possible to specific a username/password as part of the http string, but it would really be nicer if the user could simply configure a site, login and password in a config file somewhere and apt could look this up when it was attempting to make a connection and fetch a file. I'd appreciate feedback on this - hopefully i haven't made a request for something too outrageous. If you have other ideas on how we could deal with these issues i'd love to talk to you.. regards, -jason -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
