SABnzbd Smart Downloader

[imthenachoman]

You will need to install python (if you're using anything other than Windows this should already be installed). Then you will need to download and configure the script. Once configured you run it with the following:

Code: Select all

python tv_rss.py

Hope this helps,

Markus

How exactly do I download and configure it? I am on OS X.

[markus101]

You will need to use an SVN client, or manually download all the files in tag/0.9.3 from the SVN link in the first post of this thread. You configure it by editing SSD.cfg and assigning values appropriate to your setup, Log path, RSS feed, download directory etc.

If you're having issues with the setup of this, it might be beneficial to start out with SAB's built in RSS feeds and then go from there as this acts similar to SAB's RSS feed (but checks your local disk for files first).

-Markus

[minimad]

@pinion,

What's in the log just before the INFO: FAILED PROCESSING, can you post the log to pastebin.com?


@imthenachoman

OS X, something different for the script.  Not been tested on OSX before so you might get some problems (I hope not though)

@All,

Fixed the bug Markus found, it's in 'trunk' and if that works out OK, it'll become 0.9.4.

My only outstanding ticket is limiting the number of log files it produces. If I get that sorted I'll release 1.0.0 (well, when we have some documentation for new users)

Oh,

And by the way, the script names have changed in 'trunk', tv_rss.py is now SSD-TV.py and the config & aliases files have changed names now SSD-Config.cfg & SSD-Aliases.cfg.

just thought I'll keep everyone on their toes :-)

[pinion]

minimad:  Sure thing, here it is: http://pastebin.com/m4803eac0

I'm going to try to go back to earlier versions and see if I can't get something going but I'd love to be able to use that config file.

[pinion]

Ok, so I configured whatever version is in the trunk right now and ran it and it was able to add some stuff.  .93 and .92 both did not work for me.  Now my problem is the newzbin feed that I did a search on a particular show downloaded like 10 different versions of the same episode. Here is that lenghty log: http://pastebin.com/m3bc827e4

Also I didn't have to run the script as root (sudo) to get it going with the one in the trunk.

[markus101]

Trunk version is working great for me so far. No issues to report one this one.

-Markus

[pinion]

Trunk version is working great for me so far. No issues to report one this one.

-Markus

I made my newzbin feed more selective getting just xvid files in english so now I'm not getting a bunch of the same episode but I'm still getting duplicates.  I think it may be because I have a few feeds from other sites.  Could that be the problem if it thinks it's a different nzb (or if it is a different nzb)?

[markus101]

It shouldn't be an issue as long as the file is already downloaded, but if it's in the middle of downloading I can see it having problems. I think the smart queue check is meant to get around this issue, but I'm not sure.

I'm just using a single Newzbin feed at the moment, but was lookign to add a nzbs.org feed (sometimes they get nzbs on there faster).

-Markus

[minimad]

Hi All,

I think the Smart Queue Check may not be working in 'trunk'.

Pinion et all,

When you notice that SSD is trying to download several of the same episode, do you see the 'CHECK QUEUE' or 'CHECK REFORMAT QUEUE' entries in the log showing what's in your queue (each queue entry should be there twice), if not then it's broken :-(

Stay Tuned.

[minimad]

Well, Smart Queue is working.

But it maybe that the script is too quick for itself.

If the queue has these type of entries  'fetching msgid xxxxxxx from www.newzbin.com' then the queue currently sees that as NOT in queue and tries to download it. 

I'm looking at an additional check that should prevent these duplicate downloads on those messages.

[pinion]

Ok, so I did a search on my log for 'fetching' and I get several in a row.  It started with one and then theres two in a row and then three etc...  All the way up to like 10 or so

Code: Select all

2009-03-02 20:54:10,019 INFO CHECK QUEUE: fetching msgid xxxxxxxx from www.newzbin.com

[minimad]

Hi pinion,

yep, thats the ones.  script is too fast for itself.

I'll need to make the inQueueSmart check 'Smarter', I believe that there are other temporary 'titles' for adding url's, I just need to find the exact strings.

Might have something tonight for the one I know about.

[naf305]

Hi all,

First off, I'm very impressed by your work and how you (all) coordinate testing and bug fixes.

Second, I have some issue using the TV script. Basically  SDD managed to find what I have in my tv show directory but can get to download the reports for those episodes that I'm missing.

The error i get all the time (and I've tried different shows) is: INFO FAILED PROCESSING: ShowName - SeasonxEpisode - EpisodeName

I've uploaded a copy of my log to pastebin as it's seems the way to share such info: http://pastebin.com/m7aa46c17

I really hope you guys can help me out.

PS: I'm running SAB on XP using a feed from Newzbin.

[minimad]

Hi naf305,

Which version are you using? 'trunk', 0.9.3 etc.

If it's 'trunk' then please bear in mind that it doesn't always work, thats the 'extreme bleeding edge', mostly I'm testing & trying different things out or just making spelling mistakes in 'trunk'.

Best stick to the 'tag' releases.  0.9.3 is the current.

From the log it looks like the isInQueueSmart is just failing, I think that was trunk a few days ago. try 0.9.3

[minimad]

Hi All,

'trunk' now contains the additional NewzBin check, however I have noticed another bug while testing against one show (you know the one with special powers :-) ).

It's not a big BUG but possibly will only show up if uses searches, e.g. Season 3, Episode 16 has about 6 entries on the RSS feed if using the query function.  Now SSD did spot when run twice that the entry was in the queue at 'fetching' BUT as all 6 where at that stage SSD ended up putting ALL 6 into my queue as they each had a different msgid.

Now I don't know if I can fix this, but I have an idea that might work, just need to get my logic sorted first.