Page 1 of 1
Grab RSS files with a delay
Posted: July 30th, 2008, 10:19 am
by splususer
Hi all,
I was be able to setup RSS to grab nzb from binsearch.info but there is one little problem. If sanbnzbdplus grab the nzb while the uploader still uploading, I have a incomplete nzb and sabnzbdplus will skip it because it think that file has been downloaded. Is there a way to delay, for example grab only files with a day old? Or any way to get around ?
nzbindex.nl RSS feed has an option to feed only complete release but sabnzbdplus can't grab nzb from its feed. Is there anything in development ?
Thanks,
sabnzbdplus user
Re: Grab RSS files with a delay
Posted: August 5th, 2008, 2:16 am
by shypike
You don't get incorrect NZB files, but incomplete ones.
I don't see how SABnzbd would be able to see that an NZB file is incomplete.
Unless binsearch has some minimum age parameter, I don't see a solution.
Re: Grab RSS files with a delay
Posted: August 14th, 2008, 7:53 am
by MageMinds
When SABnzbd add a nzb to the queue, you could have a setting in the rss feed that say that only download when age reach X hours or X minutes. That way when a nzb is downloaded with less than this minimum age it will wait, giving the news server times to get the posts. The user will have to adjust that setting itself according to it's own experience, because you're right there's no way SABnzbd can tell if posts are complete.
An other solution could be that SABnzbd try to download the post, but get "article doesn't exists", if the age of the post is less than a day then SABnzbd keep everything in the queue and retry every 15 minutes or every hour, if the post age reach a day and still not downloaded then SABnzbd give up and clear the queue.
MageMinds
Re: Grab RSS files with a delay
Posted: August 14th, 2008, 10:19 am
by shypike
Delay: maybe.
Auto-retry: forget it: too complex. Given that many posts miss a few articles, refires would occur often for no reason.
Also, often a post is already partially on the server. Do you suggest SABnzbd stops halfway, waits and retries?
Effectively this would mean some pause every time an article is missing.
There is already a ticket to examine an option to verify the completeness of a job before the actual download.
The feasibility of this needs to be examined first, because it's not obvious that it would actually be useful.