[ludespeedny]

I am wanting to get the actor thumbnails scanned to my library, but not sure how to get them. I went into settings and checked the box, and then changed my content type for movies to none, cleaned the library, and changed it back to movies, and it re-scanned my library, but I didn't get actor thumbs. Is there a way to force it to download them?

[Milhouse]

Depends on what movie scraper you are using. Are you scraping local NFOs or depending entirely on internet content? I know that Ember will populate NFOs with urls for cast thumbs, and these will be scraped by the Universal Movie Scraper. As long as your movies can be found on IMDB then I'd expect the Universal Movie Scraper to find cast members and their artwork.

To check whether you have URLs assigned to your cast members, look in your media library (artists table) or use the Texture Cache utility to query your movies using extended json fields (add cast to extrajson.movies then use the Jd option). You can use the same utility to pre-cache all artwork content.

[ludespeedny]

I am just using the default (although I use media center master for all my meta prior to importing).
I downloaded Universal Movie Scraper, and didn't see options for actor thumbs. Am I missing it, or is it linked to another setting?

Where are the actor thumbs saved once downloaded?

[Milhouse]

I downloaded Universal Movie Scraper, and didn't see options for actor thumbs. Am I missing it, or is it linked to another setting?

The only "actor" related setting is in Settings -> Video -> Library -> "Download actor thumbnails when adding to library".

Where are the actor thumbs saved once downloaded?

In your texture cache. You should also see urls linked to cast members in each movie/tvshow. You can see all this information with the utility I linked you. I'd be using that to query the true state of your media library/texture cache, then start working from there - if there is no cast member artwork listed in your media library then that's your problem, maybe something to do with incorrect formatting in your "Media Center Master" NFOs.

[ludespeedny]

Ok, I'll check it out. Thanks!

[ludespeedny]

Looks like the problem is with my NFO's generated. If I delete the NFO's and rescrape it appears ok, but the problem is on the Pi it will take a long time. Will the utility above be able to just cache and clean the actor thumbs?
I tried to download the utility via SSH on my Pi and couldn't do it. Gave me an error any suggestions?

I can give you a NFO that was generated by MediaCenterMaster, but couldn't find a way to attach to this post.

[Milhouse]

What was the error? Try curl if wget doesn't work, or download using your browser from github.

[ludespeedny]

certificate is not trusted, and certificate doesn't have a known issuer.

Here is a screenshot:
https://picasaweb.google.com/lh/photo/ee...directlink

[Milhouse]

Try curl, or add "--no-check-certificate" to your wget command line.

[ludespeedny]

Ok that worked. now my question is will this script be able to fix bad references in the NFO, or does this just fix xbmc's reference to the thumbs?
Also, I am still pretty new at linux, would you be able to provide a sample code to run to check a particular movie for the thumbnail references? The movie I have been working with to test is 27 Dresses.

Thanks for your help so far!

---

here is the query I just ran, but doesn't show actor thumb reference.
https://picasaweb.google.com/lh/photo/xK...directlink

[Milhouse]

Ok that worked. now my question is will this script be able to fix bad references in the NFO, or does this just fix xbmc's reference to the thumbs?

No, it won't fix the references in the NFO, they're the root cause. The utility will only force whatever references you have scraped in to be cached (or to be viewed). It should allow you to reload your cache with correct cast artwork once you have scraped in suitable urls.

Also, I am still pretty new at linux, would you be able to provide a sample code to run to check a particular movie for the thumbnail references? The movie I have been working with to test is 27 Dresses.

Thanks for your help so far!

First of all, make sure you have the webserver enabled on port 8080 (Settings -> Services -> Webserver) - this is essential if you proceed to re-cache items.

To check a specific movie, you need to create a properties file (texturecache.cfg) with the following settings:

Code:

extrajson.movies  = cast
extrajson.tvshows.tvshow = cast
extrajson.tvshows.episode = cast
cache.castthumb = yes

then for a specific movie (eg. Avatar), run the following to see what has been scraped in to your media library:

Code:

./texturecache.py Jd movies avatar

You will get a JSON response which you should be able to decipher.

To re-cache a movie, eg. Avatar:

Code:

./texturecache.py c movies avatar

and it should download all artwork (posters, fanart, cast artwork) that is not currently cached.

If you want to re-cache only your cast arwork, your best option would be to find a way to identify these in your texture cache and then delete them ("s" option to search, "d" to delete). For instance, if they all include a common url (eg. http://ia.media-imdb.com/images) then

Code:

./texturecache.py s "http://ia.media-imdb.com/images"
./texturecache.py d <paste in matching row ids>

*then*

Code:

./texturecache.py c movies

to re-cache all the missing artwork (which would be your cast artwork).

[ludespeedny]

ok, I am getting there I think. I got the texturecache properties created by running sudo nano texturecache.cfg and entering the values and saving and exiting.
I ran the query, and I get a lot returned, but only can see some of it, I can't scroll up to see the rest (I am using putty) is there a way to see all the result? I can see some actors and then the title, art, movieid, and label

---

also I have my webserver enabled, but on port 80 is that a problem?

[Milhouse]

Scroll up/down using the wheel on you mouse or shift page-up/shift page-down.

Also check your settings: increase the scrollback buffer in your default Putty profile (right click in toolbar, "New Session...", load your "Default Settings", click on Window, set "Lines of scrollback" to something like 10000, tick "Display scrollbar", untick "Display scrollbar in full screen mode" - go back to Session, and click on Save (for your "Default Settings").

[ludespeedny]

Ok, the scrollback was what was needed. Thanks! Still new at putty.

ok looks like some thumbnails are referencing a location on my C: drive. So will I be able to use this to clear it out or will I just need to delete the nfo's and let it scrape them all?

[Milhouse]

ok looks like some thumbnails are referencing a location on my C: drive. So will I be able to use this to clear it out or will I just need to delete the nfo's and let it scrape them all?

If you are referencing artwork on you C:\ drive then there will be nothing to clear out from the texture cache as it wouldn't have been able to access the artwork in order to cache it.

You can try re-scraping the NFOs, which should fix the artwork references in your media library, and the act of scraping should then cache the cast artwork locally on your Pi (however if this isn't the case, then you can pre-load the texture cache using the c option, as described above).

To see what artwork needs to be cached on your Pi, run the "nc movies" command.