[JohnWPB]

Ok, I am at a loss here, and have never quite figured this out, in the years that I have been using XBMC. In past builds having a movie.nfo could over ride a <movie_Title>.nfo, so I worked around that after running scrapers by globally either renaming them with a script I wrote, or deleting them if the <movie-Title>.nfo existed. Similar things with <MovieTitle>-fanart.jpg and other files have popped up....learning along the way........

I have read on the wiki and in many posts, that XBMC will use local data if it is available

So, now with the latest XBMC Eden (released build) installed here, I am wondering why XBMC takes so long to scrape each movie, as it downloads information from the net, when everything is already local.

I have used Media Info Plus in the past (That probably dates me a bit ) but now use Ember Media Manager. When it scrapes the info, it places the fanart, cover, .nfo, .tbn's ect all in the folder. When I then import a test folder of 10 movies, it takes around 2 minutes to complete...

Thinking some info was missing with EMM, I exported "separate files" from the XBMC library. I confirmed the .nfo was overwritten, however covers and fanarts are not exported, as it would be nice if they were. So now with all the images, and .nfo in XBMC's "Native format", I clear the library, and re-import the folder with the 10 movies, and it still takes the 3 minutes, as it still is downloading information from the net.

Now and then, when I removed a drive from the library, and re-added it, it would download a fanart or cover to use that is all wrong, when the correct ones are right in the movie folder.

Is there something I am missing?

I am getting ready to do a complete rebuild of my folder structures and combining older .5 Tb drives onto new 1 & Tb drives. Once this is done, I will import the 3,500 movies back into the library, and would like it to use the LOCAL DATA, and not download anything. That and the fact that it would take hours to re-download all the information all over again, and unnecessarily slamming the servers of many that are kind enough to supply for XBMC's use, TheTVDB, Fanart.TV ect ect

For Reference, My folder structure is DVD's as root, with individual folders for each movie:

DVDs\
--Goons (2011)\.actors
--Goons (2011)\Folder.jpg
--Goons (2011)\Goon.mkv
--Goons (2011)\Goon.nfo
--Goons (2011)\Goon.tbn
--Goons (2011)\Goon-fanart.jpg
--Goons (2011)\Goon-Trailer.flv

Any help or advice is greatly appreciated!

See Signature below for MediaPC Specifications.

[JohnWPB]

Ok, 90+ views of this thread and no ideas? The database has been building on the Media PC all friggin day, and is only about half way done. I just don't get it.

It is currently in the TV\NCIS\ folder and taking about 5-8 seconds on EACH episode. That may not seem like long, but with 24 episodes in each of the 9 seasons, just NCIS is gonna take over a half hour to scan! A half hour for one TV series, what gives?

I made sure ALL banners, folder art, fanart, tvshow.nfo, all folder and show naming conventions are correct, everything... And just to make sure, as mentioned above, I did a "Export as separate files", "Overwrite All" before I started over with the Media PC. I verified after I did that, that the file dates on the .nfo files changed, so XBMC did export the data as expected. It just seems that XBMC can not read it's own data that it exported......

[gabbott]

Maybe a debug log while scanning will shed some light

[PatK]

If all the material is in the folder, and you do a manual local scrape...with refresh, does it bring up the 'locally stored information found, ignore and refresh from the internet' requester? If not then you're missing something in the folder (although you example doesn't seem to lack). I've scanned in a 2TB drive with local metadata present in each folder and Eden 11 took under an hour with a 10 year old single core machine. I'm just wondering what build you have, (do a debug log to get a devs attention) and perhaps XBMC is getting picky as regards to the material inside the .nfo like sets.

+1 on a proper debug log

[jmarshall]

It's practically 100% likely that you didn't set the "Movies in separate folders named after the movie" option.

Regardless, without looking at a Debug Log there is no way to diagnose correctly.

[JohnWPB]

It's practically 100% likely that you didn't set the "Movies in separate folders named after the movie" option.

That is quite possible..... I may have overlooked doing that on a couple of the drives....

On a side note....

Shouldn't XBMC have internal checking, even if the "Movies in separate folders" is not selected?

Something like: (This is VERY rough coding in .net code just to kinda get the idea across)

NewItem=MoviePath & DatabaseScanCurrent(TotalMovies)
if exists MoviePath & DatabaseScanCurrent(TotalMovies) & "-fanart" then FanartExists="true"
If FanartExists = "true" then use it!

The same for posters, banners, checking for he .nfo files ect ect


This simple little check, would save hours of time for anyone that forgot to check that item while scanning (Like I did, even as a seasoned user), and certainly save the movie scraper sites THOUSANDS OF GIGS of unnecessary bandwidth!

[JohnWPB]

Ok, I just decided to get to the bottom of this. After XBMC had scanned and scraped all movies again, I did a library export, to individual files, exported actors, and told to overwrite all files.

I went into a few movie folders, and verified that ALL required files, as mention above, as well as all the actors thumbs were saved in the movie individual folders.

Once this was all done, I removed all sources from the library, and then did a "clean library" just for good measure.

I then re-added the movie directories MAKING SURE to have the "movies are in individual folders" checked. It appears from the log, that it is indeed downloading the actors, instead of using the ones in the local folder:

PHP Code:

18:08:38 T:3112    INFO: Caching image from: http://ia.media-imdb.com/images/M/MV5BMTMwMDc3OTA5OF5BMl5BanBnXkFtZTYwMjQ4OTAz._V1._SY275_SX400_.jpg to special://masterprofile/Thumbnails/Video/5/5c0f2279.tbn with width 512 and height 512 


This could take up a LOT of time, as there are MANY actors in each movie.

It did see in the log that it did copy the fanart, and movie cover from the folder to it's database cache, without downloading them again.

While I am looking at the log, I have a couple more questions, and what is going on, that maybe someone can answer......

First, I see lines after every movie scrape about music video's. I have no music videos on the MediaPC, and none defined for any of the sources (I checked each source to verify)

PHP Code:

18:08:41 T:2912   DEBUG: CVideoDatabase::GetMusicVideosByWhere query = select * from musicvideoview order by idMVideo desc limit 10
18:08:41 T:2912   DEBUG: CVideoDatabase::GetMusicVideosByWhere time for actual SQL query = 2 


Lastly, it looks like it is excluding movie trailers, even though they appear to be named in the proper naming convention.

PHP Code:

18:08:34 T:3112   DEBUG: CUtil::ExcludeFileOrFolder: File 'H:\DVDs\The Order\The Order-trailer.flv' excluded. (Matches exclude rule RegExp:'-trailer') 


It says it has to be named -trailer, and that is how it is named.... not sure what is going on under the hood there.....