[PP-Script] VideoSort - better video sorting

[DoGGY2DoPE]

...

[douche]

Per the last post, I'm also having the first letter of a show being dropped.

It's fairly inconsistent, usually seems to happen with shows that start with "The" or another short first word.

[Identifier]

Thanks for VideoSort - it's great, especially when reading the "My Cart" RSS feed from NZBGeek.info since it doesn't contain metadata like imdbtitle or imdbyear, so there's no better way to get the clean name/year other than using your script. In fact, NZBGeek.info's Cart RSS feed doesn't contain the newznab imdb ID either, so NZBGet's built-in dupe handling isn't able to work since it doesn't know which NZBs are duplicates.

In this case, when NZBGet isn't able to find a dupekey on its own, would it be possible to have VideoSort simply set the dupekey to the eventual file name? That would cover 99% of cases, and let us avoid ending up with Movie (2018).mkv, Movie (2018) (2).mkv, Movie (2018) (3).mkv etc. This would probably require that VideoSort run as a Scan Script though (instead of a PP script) so that it happens before the duplicates are downloaded... is that possible?

[hugbug]

VideoSort uses python library GuessIt to analyze filenames. In a similar approach a scan-script can be written to generate dupe keys from nzb-filenames. There is no reason to make that as part of VideoSort. A separate script would be easier to write and maintain.

If it's supposed to be used for RSS then it's better to write the script as feed-script to avoid fetching of duplicate nzb-files from indexer. That's not important if you need this for cart feed only but good for general feeds.

I'm afraid I won't be able to write such script any soon because there are many other things to do for the project and I have to prioritize. Maybe you could try to write such script yourself. I could help with tips.

However you should first clarify why your feed doesn't include imdbid. Have you tried adding "&extended=1" to RSS feed URL? That parameter usually makes the feed be filled with additional fields. If that doesn't help ask NzbGeek support. Does this site have links to IMDb for nzbs? If that's the case then they should already have necessary information and it's just a matter of extending feed generator with extra fields.

[Identifier]

Thanks for the reply hugbug, that's understandable. Yes, I tried adding &extended=1 to the end of the feed URL but it didn't change anything. I know they have the IMDB data because if you get the results for a specific movie then it will include all of the IMDB information. It's just not showing up in the feed returned for the Cart. I've started a thread with them here: https://nzbgeek.info/showthread.php?tid=12454

I'm currently trying to get as close as possible by writing RSS filter scripts that set the dupekey. Here's what I have so far. (By the way, I would have used \d instead of [0-9] everywhere but \d doesn't seem to work. At least \w and \W work though ) If you can think of a more concise way to strip punctuation in POSIX regex, please let me know. Also this might still not be foolproof if different titles have different lowercased/uppercased words. I don't know how to normalize the case using this script, or if NZBGet even considers case in dupekeys or not. [Edit: I just tested NZBs with different casing, and looks like uppercase/lowercase is ignored when comparing the dupekeys - so this works great!)

Code: Select all

# For standard movie and episode titles, grab everything up to the year or season/episode, including any dots, spaces, punctuation, etc.
O(dupekey:${1}): $(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))

# For better duplicate matching, attempt to separate out the words to ignore differences in punctuation (dots vs spaces, parenthesis around years, etc).
O(dupekey:${1} ${2}): $(\w+)\W+(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3}): $(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3} ${4}): $(\w+)\W+(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3} ${4} ${5}): $(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3} ${4} ${5} ${6}): $(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3} ${4} ${5} ${6} ${7}): $(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3} ${4} ${5} ${6} ${7} ${8}): $(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))
O(dupekey:${1} ${2} ${3} ${4} ${5} ${6} ${7} ${8} ${9}): $(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W+(\w+)\W*(.*((19|20)[0-9][0-9]|S[0-9][0-9](E[0-9][0-9])?))

# Adjust the duplicate score based on personal preferences.
O(dupescore+:720): 720 | 720p
O(dupescore+:1080): 1080 | 1080p
O(dupescore+:2160): 2160 | 2160p
O(dupescore+:-1000): size:<5GB
O(dupescore+:-1000): size:>15GB
O(dupescore+:-1000): size:>40GB
O(dupescore+:-1000): category:foreign

# Copy over the category (needed for NZBGeek feeds since NZBGet doesn't get the category from them correctly).
O(category:${1}): category:$(.*)

# Finally, accept all items along with whatever attributes we added above (if any).
A: *

[Rrrr]

Per the last post, I'm also having the first letter of a show being dropped.

It's fairly inconsistent, usually seems to happen with shows that start with "The" or another short first word.

Same here. I do not understand if hugbug's comments are related to this issue. They seem more related to the other issue in the OP.

Any tips?

[hugbug]

I'm also having the first letter of a show being dropped.

Can someone provide a sample filename (nzb-name and video file name) with VideoSort settings?

[Rrrr]

Code: Select all

The.Big.Bang.Theory.S10E06.The.Fetal.Kick.Catalyst.1080p.WEB-DL.DD5.1.H264-BTN

VideoSort:

Code: Select all

PP-Script Version: 8.0.
SeriesDir: /video/2 - TV Shows
SeriesFormat: %sn/Season %s/%sn - S%0sE%0e - %en
MultipleEpisodes: Range
EpisodeSeparator: E
SeriesYear: Yes
DatedFormat: %sn/%sn - %y-%0m-%0d
OtherTvFormat: %t
LowerWords: the,of,and,at,vs,a,an,but,nor,for,on,so,yet
UpperWords: III,II,IV
DNZBHeaders: Yes
PreferNZBName: No
Overwrite: No
CleanUp: Yes
Preview: No
Verbose: No

Result:

Code: Select all

2 - TV Shows\He Big Bang Theory\Season 10\He Big Bang Theory - S10E06 - The Fetal Kick Catalyst

[hugbug]

Rrrr, Thanks.

Weird indeed. I'll look into it.

[Rrrr]

Just to note that the behaviour is not consistent:
1) All with BTN in the name go wrong;
2) Some others go wrong as well.

Wrongly named and placed:
S10E02: The Big Bang Theory S10E02 The Military Miniaturization 1080p WEB-DL DD5 1 H 264-BTN
S10E06: The Big Bang Theory S10E06 The Fetal Kick Catalyst 1080p WEB-DL DD5 1 H264-BTN
S10E09: The Big Bang Theory S10E09 The Geology Elevation 1080p WEB-DL DD5 1 H264-R2D2
s10E20: The Big Bang Theory S10E20 1080p WEB-DL DD5 1 H264-RARBG

Correctly named and placed:
S10E08: The Big Bang Theory S10E08 The Brain Bowl Incubation 1080p WEB-DL DD5 1 H264-R2D2
S10E15: The Big Bang Theory S10E15 1080p WEB-DL
s10E16: The Big Bang Theory S10E16 1080p WEB-DL DD5 1 H264-RARBG