Author Topic: Post Processing  (Read 3251 times)

Offline jonnyboy

  • Epic Indexer
  • *****
  • Posts: 1046
  • Helpful: +93/-1
  • Lazzy Trucker
    • nZEDb
Post Processing
« on: 2013-06-12, 08:38:34 AM »
When we talk about post processing(pp), we mean things that happen after a release is created(stage 4) and the nzb is written to the disk(stage 5). The release is not waiting for the post processing to be completed before it is available.

pp additional - this downloads parts of the release looking for content, this is where previews and samples come from
pp nfo - this opens the nzb and looks for an nfo
pp movies - this queries google, bing and trakt looking for an imdb number, you are limited to 100 advanced searches per 24 hour period from google
pp tv - this queries your local db and tvrage trying to match it to a tv show
pp console, music books - this queries amazon for information - you are limited by amazon to 1 query per second, bursting is allowed, but it is undocumented

Offline jonnyboy

  • Epic Indexer
  • *****
  • Posts: 1046
  • Helpful: +93/-1
  • Lazzy Trucker
    • nZEDb
Re: Post Processing
« Reply #1 on: 2013-06-22, 01:51:49 PM »
The threaded postprocess additional and nfos has been changed. Now, python makes one query, gets the results and sends one out per thread. This may be easier on mysql, but watch your load. This processes faster per release.

I will eventually rewrite all of the postprocessing threaded to function like this. But, it takes a little time to write and test.