Hi,
I have a pretty good Media Companion generated metadata for all my content and would like XBMC to use only that when scraping. I could not find that option - is that even possible?
Ideally, I would set up XBMC to skip the content for which .nfo file does not exist, only to be added to the library on the next manual or automatic scan after the .nfo file has been added.
BTW, I am in the process of setting up the centralised MySQL database for all my XBMC installs, and will probably set up an automated nightly library scan on of them so all can be in sync.
Thanks,
-LM
If you have a properly formatted (for XBMC) .nfo file then XBMC will use the data contained on that instead of scraping from the internet (imdb, tvdb, etc.).
By default, XBMC really only updates your library when you start the program. So if you launch XBMC it will scan your library for new content and attempt to scrape. You could name the (video) file in such a way that it is not scrapable. Then change it when you get around to creating your own .nfo files.
(2012-09-26, 05:46)outatouch0 Wrote: [ -> ]You could name the (video) file in such a way that it is not scrapable.
You mean, adding an underscore or any such prefix that would confuse the scraper? Any known method or should I just revert to trial-and-error?
I was thinking trial and error but i guess you don't want to do so good a job you confuse yourself! Just come up with soemthing that works for you.
Perphaps something like 'Fast.Times.at.Ridgemont.High.1982' could be FTARH'82? The scraper wont recognize it but you should be able to find it when you want it or recognize it if you run across it.
Another option may be to move all your video files that do not yet have NFO files to a separate directory. You can add this new directory as a source, but don't specify a scraper for it (directory type = "None") and, for good measure, select the option "Exclude selected folder from scans". This way, you'll have access to these files in Video/Files view, but no attempt will be made to scrape them into the library using TMDB info. When you get around to creating an NFO for a given movie, move the movie file and the NFO file to the main movie directory and update the library.
Yes, I guess I'll just have to be more disciplined and collect all unscraped films under a separate folder and when scraped move to the main library location. I think this will work... but this "nfo only" scraping option would actually be even better.
(2012-09-28, 01:14)LabudMoj Wrote: [ -> ]I think this will work... but this "nfo only" scraping option would actually be even better.
Depends how long you're willing to wait... by then, you may have created nfo files for all your movies.
Placing an <id>-1</id> tag inside an NFO disables attempts to scrape extra details off the net regardless of which scraper is selected for the directory, by the way. You'd need to add that into every NFO file directly, though, but that's easy to automate.
(2012-09-28, 16:53)Mihara Wrote: [ -> ]Placing an <id>-1</id> tag inside an NFO disables attempts to scrape extra details...
Yes, but you are losing a possibly valuable reference to the original scraping source. However, not a bad idea if that "feature" is not required.
Anyway, I just discovered that XBMC will not refresh the database if the .nfo changes, which happens quite often for me (when I discover incorrectly scraped details, or just the missing ones). As far as I know, there is no way to refresh but to remove and add the source again. How on earth am I going to automate that for the nightly refresh of the database? Any ideas?
I think maybe you could also do a database export/import cycle (assuming whatever you use to edit nfos doesn't throw away data), but I'm thinking you might need to have the "clear=true" attribute set. This is something I am also trying to get my head around. I guess the theory is that scraping is a major advantage of XBMC so there wasn't really a concept of offline database updates.
scott s.
.
(2012-10-03, 23:25)LabudMoj Wrote: [ -> ]Yes, but you are losing a possibly valuable reference to the original scraping source. However, not a bad idea if that "feature" is not required.
Well, I only found out about that when figuring out how to add something that isn't in a database and shouldn't be. Something more explicit like a "<authoritative/>" tag which would prevent attempts to rescrape but keep the IDs would be better, but there's no such thing at the moment.
(2012-10-03, 23:25)LabudMoj Wrote: [ -> ]Anyway, I just discovered that XBMC will not refresh the database if the .nfo changes, which happens quite often for me (when I discover incorrectly scraped details, or just the missing ones). As far as I know, there is no way to refresh but to remove and add the source again. How on earth am I going to automate that for the nightly refresh of the database? Any ideas?
There's probably no practical way for XBMC to even know that an NFO is different from what it was without re-parsing it every time,
(Keep track of file dates? Keep track of file checksums? Both quite annoying to code.) so I doubt it can be done in a reasonable fashion. Show info on the item -> Refresh is a bit smoother, since it works without add-remove or change content type. It might be possible to automate that with scripting, but I haven't gone that deep in yet. For singular items which I have just edited the NFO on it typically sufficed for me.
(2012-10-04, 10:14)Mihara Wrote: [ -> ]There's probably no practical way for XBMC to even know that an NFO is different from what it was without re-parsing it every time
In my opinion, it would not be too painful to reparse it each time if it happens only at 5am. It takes up to 10mins for me, movies only, I guess would be 30mins tops for everything.
(2012-10-04, 10:14)Mihara Wrote: [ -> ](Keep track of file dates? Keep track of file checksums? Both quite annoying to code.) so I doubt it can be done in a reasonable fashion.
Using checksums would actually be much more elegant. I have dabbled with something like this in the past, but don't know enough of XBMC plumbings to estimate how long it would take. I prefer a solution that refreshes the whole collection, so will look into this at some point (Xmas holidays perhaps
).
(2012-09-26, 05:46)outatouch0 Wrote: [ -> ]If you have a properly formatted (for XBMC) .nfo file then XBMC will use the data contained on that instead of scraping from the internet (imdb, tvdb, etc.).
i don't think that is true, i've got .nfo files for movies not on imdb etc; and they're just ignored. if it can't be found on the internet, it doesn't get added to the library it would seem, which is pretty stupid.
(2012-10-24, 00:29)sej7278 Wrote: [ -> ] (2012-09-26, 05:46)outatouch0 Wrote: [ -> ]If you have a properly formatted (for XBMC) .nfo file then XBMC will use the data contained on that instead of scraping from the internet (imdb, tvdb, etc.).
i don't think that is true, i've got .nfo files for movies not on imdb etc; and they're just ignored. if it can't be found on the internet, it doesn't get added to the library it would seem, which is pretty stupid.
Sure they'll work if they are
properly formatted. Compare your NFO files to what's shown on the wiki and assure they are named correctly.