@halkon - thanks for the great work that is done with the content agent over the past month. I have one question, could you maybe change that it uses a 400x600 picture (“portrait”) by default for “Cover” and 640x480 picture (“landscape”) by default for “Background”? The size does not actually matters to me, but the format. I know some sites, BB e.g., don’t offer portrait to data18, but most (brazzers, etc) also offer portrait and nowadays, it seems kinda random, what is choosen? Or at least the possibility to download “all” pictures offered by data18 at once? Now, I sit there, and have to manually refresh as long as it downloads a possible correct one and pick that manually.
@halkon - Any idea why the scraper fails when my directory structure is set up as:
Studio -> Collection -> VideoName.mp4 ? Is there a specific structure that the scraper expects?
@alex.holliz Structure doesn’t matter, the main thing that matters is the VideoName. So long as it matches the title in data18.com it should match, if you want DM me some of your failed videos and I can see what’s up with them.
@fritz276 I’ll look into it, it already pics anything marked as @poster for the main poster image and collects a series of images, sadly I was not the original writer, and that’s something I’ve been wanting to improve.
But you’re right, at least the main video picture should be portrait, while the rest can be landscape.
I’ll look into it, thanks!
@halkon - thanks for looking into it. At least as most important is that the background picture is landscape only (for Samsung TV) 
Ah yes I totally understand that, I have one of those too, though recently switched to just using the Roku. Nonetheless, you’re totally right. Even on mobile it looks bad
Can someone give me a quick basic idea of how the agent works in matching things up. Is it real basic and just go by file name or does it do some “Shazam” magic in like comparing the video to some database i.e. this scene was in x so it must be x
It’s really simple and just goes by the Folder/Filename combination, though mainly the filename.
Both the Data18 and Data18-Content plugin do searches against data18.com, much like:
http://www.data18.com/search/?k=Test
The rest is just matching the filename to the title of the scenes/movies. Similar to the Movies/TV Shows matches in Plex
hi,
data18 works great, but i have my problems with data18-content.
1157399.mp4 works (but i’d never find anything again in my filesystem if i rename all scenes like that)
Blonde Gets BBC at Interview.mp4 works often, but there are many identical names
so i made a small change in the code, its not pretty, but it works for me.
arround line 263:
def search(self, results, media, lang): Log('Data18 Version : ' + VERSION_NO) Log('**************SEARCH****************') title = media.name[:7] content_id = False if title.isdigit(): Log('Media.name is numeric') content_id = True contentURL = EXC_MOVIE_INFO % title html = HTML.ElementFromURL(contentURL) curdate = parse_document_date(html)
now i can rename my files to 1157399-what ever i want.mp4
maybe someone else finds it useful.
P.S. sorry for my bad english
Does this plug in pull “Description” details as well? I’ve installed and run it and it ‘appears’ to be working but not sure if its just pulling a image from the video itself as a cover art and no description are showing up.
@halkon I found that it has something to do with how plex imports videos into the library. If there is only one video in the directory, then it will import that video with the directory name. I have figured out how to work around it for now. Thanks for your help!
@ropeontherocks it should pull in the description as well, if you have any specific titles that it is not pulling description for please let me know so we can track it down, you can DM me the title if you want.
@ferengi that’s actually interesting, for me it’s been working with the title as well as a date, so long as the year and the title match that of data18, it works. However I am quite interested in your fix because that means people could just name the files however they want, do you have github, perhaps you can create a PR against the code and I can further look at it. Thanks!
@alex.holliz that’s good to know, I haven’t messed with it much before this time, so it’s good that you found that out. Thanks!
@halkon i have github, but never used it to push something. I tried it, but i get permission denied error. If you contact me at USERNAMEfromPLEX @gmx.de i will send you the file.
@ferengi given the permissions on the repo you have to fork it and create a pull request from your origin. No worries.
I’ll email you later this afternoon to fetch the file
Hello when the agent looks for data it ends up with an empty line do we know what is the problem?
I think I might have asked that question already once but… Would it be possible to extend the data18 and data18-content agent to do the following
- pull the scenes (where they are) with the information (typically is the actor name)
- if we could pull the scenes maybe even the thumbnails which are on the web ?
Would be really helpful in for larger movie collections to find what I am looking for. Thanks !!!
Curious question…does data18 only have straight movies/scenes in their database?
Sorry guys, new year started and got hella busy.
@bhuisman can you send me a screenshot via DM and we can work through it? I thought we had fixed something like that before, but want to verify what the actual issue is.
@spanishu that’s interesting, you want to essentially have scene information embedded within plex, so that when you open up a movie you can choose the scene that you want to play? I’m not sure how much Plex supports that, but if it does, it might be doable? not sure at all though.
@LKeels they do not, if you go through data18.com you will see movie/content for everything
Hey there; 1st of all thanks for this plugin it works great and I’d like to add that @ferengi I think that is a brilliant idea as I was having the same problem. I wanted to keep the name of the scene so I could easily find it BUT matching them up to the ID number works so much better.
One thing I’ve come across that I’m looking for a solution. I’ve now got a pretty large library (couple thousand titles). Many times the scraper will not pickup all of the names of the actresses in every movie. So let’s say I’m in the mood for a Jenna Haze flick. If I do an actor search I will find a good bit but certainty not ALL the titles I have with her in the film. So I switched to using the “collection” field. The problem is, that collection field is often filled in with something. Is there anyway I can like mass wipe out or change the script to NOT touch that collection field?
I’ve even tried meta data editors but that field doesn’t show up in the ones I’ve tried thus far.
Thanks again!
