Inspired by the existing data18 agent (which works on a movie based library structure), i've been developing an adult agent which targets series based adult libraries, following the structure:
- Some Multi-Series Release
- Season 01 <-- first series of a movie
-- Some episode title - s01e01.mp4 <-- first scene of the movie
-- Some other episode title - s01e02.mp4 <-- second scene of the movie
- Season 02 <--- second series of a movie
-- Some episode title - s02e01.mp4 <-- first scene of the movie
-- ...
I've release the first beta version of it on gitlab:
many thanks in the first place. I have installed the agent and it seems to be working partially. It did detect the movies and for some it put episodes, not for all. That might be now something with the naming convention mine is Title and then P01.mp4 for example.
Then it did not catch the posters of all movies, even if they are from the same label.
Hi, great to see somebody working on the current adult situation! The agent did work for me (just some basic tests). I am not so sure if I like the idea to split my database into "tv series" aka movie scenes (with your agent) and "scenes" aka site updates like bz,na,rk (with Data18-content), e.g. because I cant select "anything" with actor abc from both databases then. What I would like to see is a possibility to see the scene thumbnails when I am at the episode section (its kinda boring to see 7x the movie poster then). I will write back after more testing.
@spanishu: Regarding the naming convention: It's basically the same as the convention for tv shows, so if a filename does not contain 'sXXeXX', the series scanner (which is the one that runs before the adult agent and determines if it's a movie or a tv show) will not recognize it as an episode, and therefore the adult agent will not be able to do anything.
Regarding the missing thumbs: Unfortunately data18.com does not have a thumbnail gallery for all scenes. The ones that don't have thumb galleries only have 3-4 very small images on the scene page, which i thought i'll skip because the resolution is simply too small for the plex UI, it will be pretty pixelated if i add them. But if you want i can add them too and make this configurable.
@fritz276: I agree with your problem splitting up the library into "movie scenes" and "site updates", but i haven't found any proper solution yet for this. The problem is the following:
If the series based agent (the one provide in this thread) gets called by plex to scrape for metadata, it gets the info on what the "show" title is, which is determined by the plex scanner based on the parent folder. So for example, if you have a folder named "Kink" containing scenes from kink.com, the agent will try to search data18 for "Kink", which will also return movie results for movies named "Kink".
So for a fully generic solution, the agent needs to somehow figure out if the thing that is requested by plex is a "movie" or a "site update".
Actually, here's a possible solution for the "site" vs "movie scenes" problem: We could simply add a naming convention for the parent folders:
1. If it contains "site" content, the the folder should have the full name of the site, e.g. "brazzers.com"
2. If it contains a movie series, the folder should have the name of the movie
This way the agent can check if the movie name can be reaches as a website, and if it does, it will scrape the single episode names instead
of scenes from a movie.
@fritz276: Also, the episodes/scenes should have separate thumbs already. If they don't than it's probably a bug. Sending me the name of the folder and the filename in a PM can help me finding the issue.
@globochem thanks for the feedback my thought was actually not so much on the thumbs of the scenes, where scenes = episodes but to get the actor information from there. Not sure if that is doable but would be actually a cool feature.
I will think about your idea, sounds good at the moment. I just have to think / sleep it over, I have more than 7000 files and I dont want to screw it up. With the seperate thumbs: I will send you a PM on the weekend, also with some more ideas / requests :-)
Honest question, because on more than one occasion the state of Adult media Metadata has driven my inner OCD to produce some code. However its not a 'set and forget' kind of thing. So... if people really want this, I could polish up my code and put something online, but I would kinda want it to earn at least enough money to cover costs. This may seem odd, 'surely its easy to maintain a library, just fix it when it breaks', yeah thats the easy part. I actually build my own metadata database to normalize the hundreds of sites and publishers so to do that ... i pretty much had to build my own version of data18.com and that involves a little more maintenance.
