Feature Request -- add entries to excludeList only on success

[AnotherMatt]

If a scan fails on a particular file, it should not be added to the excludeList.

If you're running strict matching, you can often find that amc fails matching a particular file. However, even when it fails, it is added to the excludeList as this seems to happen before the script runs. IMO, the excludelist is to ensure you don't process a file that has already handled. However if a file fails, it still needs processing by the app.

[rednoah]

If a particular file fails, then it won't be moved, and thus remains in the input folder, thus gets processed again, which fails again, and again, and again, indefinitely. The --def excludeList option exists precisely to prevent this kind of accidental infinite loop.


If a particular file fails, then you need to process it in a different way. The --def excludeList option of the amc script notably does not prevent you from manually processing the file in a different way, e.g. via the FileBot Desktop application or filebot -rename calls.


How would you prefer to solve the problem? How would not adding failed files to the exclude list help you process files successfully? Perhaps you have forgotten to mention key details of your process / setup? Maybe you mean to run different / multiple amc script calls with different options depending on the files at hand?


Please read How to Request Help.

[AnotherMatt]

Thanks for the speedy reply!

I have a single directory where I store all media files. I then run filebot against this directory on a regular basis to pickup new files, process them, and hardlink to my Plex media folders. I therefore use excludeList so that I don't process files already scanned. The process is handled automatically so I don't monitor the output to check it was successful.

My plan is to run a manual scan on a less regular basis (eg once per week), where I pick up any files that have not been processed owing to scan failure and then manually fix the match.

Can you suggest an alternate solution?

[rednoah]

I can think of many solutions.


If you hardlink files, then the hardlink count will tell you if a file has been processed already or not. See How do I process only specific kinds of files? › Option 4: Start FileBot with --file-filter from Terminal for ideas.


The existence of xattr metadata or internal history might also allow you to track and find unprocessed files. See How do I process only specific kinds of files? › Option 3: Use Presets with Includes for ideas. You could also use Custom Post-Processing Scripts to somehow mark (e.g. add xattr, create file, add lines to a custom log file, etc) files that have been processed.


The amc script itself also has the --def unsorted option:

--def unsorted=y Process unidentifiable files (as opposed to skipping them) and move them to the Unsorted folder

[AnotherMatt]

Thanks for your suggestions. I'll experiment with them.