I've got some good stuff I've brewed up for this meeting. The download
library is now fully awesome. I've got it working just like how APT works
which is as follows:
1) Create an HTTP connection request for the file
2) Read the datums from it to determine the full content length of the file
and the last time it was modified
3) Check our local file to see if it exists
- If it does we make sure the file sizes match, and make sure that the
modification date is not older than the latest
- We resume the file if it is not finished, redownload the file if it is
old
4) Download the file if we need to
This creates (potentially) twice as many HTTP connections, but the first
ones request a very small amount of bytes (just the headers) so it doesn't
increase bandwidth usage much. It works just like how APT does, so that you
get the "Hit" messages when the file doesn't need to be redownloaded so it's
smart about it. Everything seems to be working properly, but I haven't been
able to test much of the "outdated" files yet. That might have bugs, but
what code do I write that doesn't have bugs anyways? :P
See you at the meeting!
On Sun, Jul 18, 2010 at 9:30 PM, mac9416 <mac9...@keryxproject.org> wrote:
> I've been on the road for several days and have done almost no coding.
> So, unless Chris has something going on, there shouldn't be much going
> on at the meeting tomorrow night. But, of course, everyone's welcome
> to hang out anyway.
>
> The place is #keryx (irc.freenode.net) and the time is 0000 UTC
> (8:00PM Eastern, 7:00 Central).
>
> -mac
>
> _______________________________________________
> Mailing list: https://launchpad.net/~keryx
> Post to : keryx@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~keryx
> More help : https://help.launchpad.net/ListHelp
>
--
Chris Oliver
http://excid3.com
_______________________________________________
Mailing list: https://launchpad.net/~keryx
Post to : keryx@lists.launchpad.net
Unsubscribe : https://launchpad.net/~keryx
More help : https://help.launchpad.net/ListHelp
