Skip through unmatched movies using strict hash matching

I'm trying to make a bash script to pass through my movie collection twice, once using the strict method and moving all the correctly matched items to a particular directory (e.g. /media/movies/strict). Then the second time using --non-strict and advanced matching and moving all those to another directory (e.g. /media/movies/double-check)

That way I can go through the logs for the advanced matched files (seems to be only 5% of the movies) to check for errors.

Is there a switch or other simple way for filebot to just skip through all the movies that it can't match and keep on processing the ones that it can match?

Since so far I have 95% matches using the strict method, I would like to get those done first and then deal with the others using --non-strict and a visual/manual check for mistakes.

Thanks!

[rednoah]

Does strict mode mess up your files in anyway? Basically it should work first process in strict mode and again with -non-strict. Now strict mode will not only use hash (found it unreliable for many movies) but also a matching movie+year sequence. The movie+year combination should make the match very reliable. if its not possible to detect the movie then the files should be left untouched.

filebot -script fn:renall --db tmdb --format "/strict/nonstrict/{movie}/{movie}" [-non-strict]

Running that twice with argument for strict/non-strict should do most of the work. Let me know what's still missing from that solution.

Thanks for your reply.

I didn't have a chance to try out your

filebot -script fn:renall --db tmdb --format "/strict/nonstrict/{movie}/{movie}" [-non-strict]

solution as I had already finished processing all the files available to me by the time I got the message.

The original question was how to stop the -rename command from exiting every time it reaches a file that it cannot match. As you will see from the below process I had to do a lot of manual manipulation to get through my test.

Anyway, here are the results of my test with 123 movies:

I first ran:

Code: Select all

filebot -rename /media/* --format "/media/strict/{n}_({y})_by_{director}/{n}_({y})_by_{director}_IMDB({imdb})_Part{pi}_of_{pn}"

96 files were processed successfully in the strict mode. 27 movies made the rename command stop and exit. Every time the command failed I moved the offending movie into another directory called /media/failed and then re-ran the -rename.

After all the files that could be done using strict were matched I then ran the following to try and match the 'failed' movies using the non-strict method.

Code: Select all

filebot -rename /media/failed/* --format "/media/non-strict/{n}_({y})_by_{director}/{n}_({y})_by_{director}_IMDB({imdb})_Part{pi}_of_{pn}_{fn}" -non-strict

The number of incorrect matches in strict mode was 3 out of 96 (though one was just for a srt file, the movie passed ok)
The number of incorrect matches in non-strict mode was 4 out of 27

Not too bad, but the 3 incorrect matches in strict mode was a little disappointing which means I will have to manually check all the incorrect movies too. Originally my thought was to just check the non-strict matches manually.

Here are a list of the incorrect matches:

strict mode:
Face Off.avi -> A Face of War_(1968)
Family Guy - Its A Trap.avi -> Family Guy_(1999)
fright.night.2011.bdrip.xvid-amiable.avi -> Fright Night Part 2_(1988)

non-strict mode:
Fast & Furious[Eng][Subs].srt -> Foxpro Fast & Furious_(2010)
Hard.Boiled.1992.ReRip.Xvid.AC3.CD1-ALLZINE.avi -> Hard Boiled_(1919)
Family Guy Presents Blue Harvest.avi -> L.A.A.P Presents Family Values_(2007)
Dune - Extended Edition (1984).avi -> La fin d'une tradition_(1991)

I tried out

filebot -script fn:renall --db tmdb --format "/strict/nonstrict/{movie}/{movie}" [-non-strict]

and it works well for two scans switching the strict/non-strict options.

Thanks a lot!

[rednoah]

Yeah, that script is really just processing things folder by folder. You could do the same as cmd/bash script but that'd be 麻烦极了。

I'll send you a sample on how to read torrent files and stuff after my trip in 江西。