Backlog - RSS Feed - Cache Model?

[sbuser]

Is there a plan to move to a model like sickbeard uses to prevent hammering provider sites?

SB runs an initial search for something if it is added - a limited backlog search - and then just checks the RSS feeds for the categories at the given providers periodically to stay on top of new nzb files. It compares each feed to the listed of wanted and pulls them down. Weekly backlogs fill in any gaps in case the RSS feeds moved too quickly or for whatever reason something didn't get caught.

I ask because once you start getting a large database of "Wanteds" headphones really starts to hammer the providers. I've taken to turning off headphones if I goto sleep because of the unnecessary hammering (I could also just up the timer on the searcher thread but you get my point).

[rembo10]

I just started noticing this because im maxing out my api hits on one site. Yeah it should really be cached

On Aug 11, 2011, at 9:28 AM, sbuserreply@reply.github.com wrote:

Is there a plan to move to a model like sickbeard uses to prevent hammering provider sites?

SB runs an initial search for something if it is added - a limited backlog search - and then just checks the RSS feeds for the categories at the given providers periodically to stay on top of new nzb files. It compares each feed to the listed of wanted and pulls them down. Weekly backlogs fill in any gaps in case the RSS feeds moved too quickly or for whatever reason something didn't get caught.

I ask because once you start getting a large database of "Wanteds" headphones really starts to hammer the providers. I've taken to turning off headphones if I goto sleep because of the unnecessary hammering (I could also just up the timer on the searcher thread but you get my point).

Reply to this email directly or view it on GitHub:
#236