I'v posted before with little success about how to get sport to index properly, I was trying to do it with nfo files and all sorts - I haven't managed to get anything to work. Well - I've had enough and I would like to write a scanner and metadata agent to do this.
The only two sports I am really looking at right now are NHL and European football (mainly EPL and associated competitions). There aren't any decent APIs that I can find that would allow me to retrieve all that I want - but having looked at the source/URL composition of the ESPN NHL pages and the skysports.com Football pages, I really think that I can just scrape the pages to get the information.
The biggest issue we then face is how do we scan these things?
- We need an exact date so that we can look up the relevant information
- We need to be able to deal with multipart downloads and single file downloads with games being downloaded in both formats
- We need to be able to work out what sport it is that we are looking at (I'm OK with the idea of creating an NHL only scanner/agent for starters to alleviate this issue)
I know perl but not python so this is a bit of a learning experience for me.
So - I am really posting for two reasons. The first is me wondering if anyone else has an interest in an NHL scanner being produced? The second is looking for a bit of help as to how the scanner/metadata scanners are called and what is actually passed between them. There was some documentation floating about which I can't find now that the site has been redesigned, but if anyone has experience with writing scanners and advice I would very much like to hear it :)
Thank you in advance for any help anyone can give me!
What IS possible though is scraping sites (I have one in mind that is relatively easy to do). I know that it is less than ideal what with sites changing more often and potentially some pesky terms of service, but from what I can tell it is pretty trivial.
Due to the nature of things i don't think it's viable to make a site for Sports Events Metadata considering what it would be used for in the long run as they don't make DVD/BR of Sport Events
I get your point, but then I think that the metadata you actually require is very limited and very easily maintained. All you really want is the competition a match is played in, the data it is played, the teams involved, the location.....and that's pretty much it! Maybe a poster involving the two teams at play.
I'm beginning to think that it would be more of a community service starting up a metadata site/API with this simple setup and letting everyone else use it though....I may just host a local one and see how it works really.
Hmmm....I wonder how well I could use thetvdb.com source?
For anyone that finds this - there is an up-and-coming metadata source at www.thesportsdb.com, take a look, sign up, contribute and I’ve written a scanner and Metadata Agent that uses it.
The website is improving/growing as it is missing things currently such as support for cup competitions, and the scanner is only a few weeks old and is my first python project so there are certainly things to improve.
Thanks for this! I had this working in Kodi, and getting it in Plex is one of my hurdles for switching. I’m running PMS 1.0.0.2261 on Ubuntu. I downloaded the files, and put them into the required directories (I think). When I created the new sports library in Plex, I chose TV shows. I can select ‘SportsScanner’ as the Scanner, but the Agent doesn’t have an option for it; just Personal Media Shows, TheTVDB, and The Movie Database.
I’m really looking forward to getting this working. Can you help ascertain what I’ve done wrong?
Sure I can try and help - do you mind opening an issue over on my github page for the project - the troubleshooting we do might be a good bit of information for other people to look at.
