Avoid accessing the database directly if possible

[ukdtom]

Since supportfiles for SQLite has to be added to every Linux distro, and since Plex discourage direct access to the database, I humbly suggest, that you perhaps looks into getting the info out of the database like this:

http://127.0.0.1:32400/library/sections/"Number of section"/all?X-Plex-Token="myToken"&addedAt>=1434158288

Best Regards

dane22

[ukdtom]

And adding here....To see howto grap the other MetaData, feel free to steal from my Plex2CSV bundle

https://github.com/ukdtom/plex2csv.bundle

[aptalca]

At linuxserver.io we just deprecated the docker container for plexemail after confirming that it corrupts the plex database.

I would like to echo the above request by @ukdtom so the damage can be prevented as I would like to use it myself and perhaps consider republishing the docker container.

Apart from the database corruption, the script itself works really well.

Thanks

[jakewaldron]

I've made some changes around the DB usage in the dev branch. It now properly closes connections and before opening the db files, it first makes a copy (and deletes the copy after closing the db connection). This should prevent corruption issues in the future. I plan on pushing it to the master sometime this weekend.

As for switching to the API, I have a new child and no longer have lots of free time, so I have no plans as of now to make the switch. If someone reads this and wants to take it on, please feel free to make a fork (or make the changes and submit a pull request).

[jakewaldron]

Dev branch has been merged into master. As said in my previous comment, I don't have the time currently to do a major update to the script to use the API vs DB. Will leave this open though in case someone else wants to take on the task.