Lidarr support

[waredrive]

Hi, Thanks for the great script. I'm wondering though if it would be possible to add support for Lidarr as well? I'm usually having a lot of stalled downloads or downloads with missing data when it comes to Lidarr.

[ManiMatter]

hi @waredrive , thanks for reaching out.

I had a quick look at the lidarr API doc, and it looks pretty similar to Sonarr/Radarr, so I guess expanding it should not be that difficult. Since I do not use Lidarr myself, what I can offer is that I duplicate the logic from Sonarr for Lidarr, but I'd need your help with testing it and letting me know any issues.

Would you that work for you? If yes, I can put a first attempt with lidarr together, publish it on the dev package, and you can test. How does that sound?

[waredrive]

Sounds great! I'll gladly help out with the testing

[ManiMatter]

Done, it's in the DEV branch (and package). As said: I've added the code completely blind by implementing the same structure as for Sonarr; 99% there will be something that won't work. Please try it out, and let me know :)

[waredrive]

Will do :+1:

[waredrive]

I'm glad to report that everything is working the same way it works in Sonarr and Radarr. Downloads are being removed, added to blocklist and re-fetched. I saw one warning in the log the next run after all the files have been cleaned up, but it seems it didn't affect anything?

[VERBOSE]: Cleaning queue on Radarr:
[VERBOSE]: >>> Queue is clean.
[VERBOSE]: Cleaning queue on Sonarr:
[VERBOSE]: >>> Queue is clean.
[VERBOSE]: Cleaning queue on Lidarr:
[WARNING]: >>> Queue cleaning failed on Lidarr. (File: queue_cleaner.py / Line: 218 / Error Message: 'downloadId' / Error Type: <class 'KeyError'>)
[VERBOSE]: 
[VERBOSE]: Queue clean-up complete!

[ManiMatter]

great news, that means we're almost there. That error we see there we need to iron out though.

Can you please do the following:

• I've added a new version :dev with more logs, can you please pull it
• In the settings, under features, please leave only REMOVE_ORPHANS on true, set all other features to false
• put the LOG_LEVEL to DEBUG
• post the logs to pastebin and add the link here

[waredrive]

Ok, will do

[waredrive]

Just FYI, the error seems to have vanished after restarting the containers. I will run Lidarr for a while, downloading new files, and we'll see if I get it again.

[ManiMatter]

I suspect you can force this bug by 1) add a new artist 2) grab album (so it‘s in the queue) 3) delete the artist (item will still be in queue but „orphaned“) 4) run the script

[waredrive]

Ok, just did what you wrote and the album was deleted directly (on container start) without any error messages..

[ManiMatter]

Ok and you got the corresponding info in the log that it would be removed, right?

so let‘s let it run for a few days and when you are happy with it we merge it to the latest package

[waredrive]

Sure, will report back in a few days.

[waredrive]

Ok, I have been using the dev version now for a couple of days, and the Lidarr part is working as expected. I think it's ok to merge it to latest.