[loady]

I got a great deal of advice from some members here and read a good few threads on other people having troubles with scraping so i thought i would like to give something back in the hope it will help others.

The following instructions pertain to DHARMA 10.1 on WIN7 for me, it may work in other circumstances where applied (maybe someone can add to) but i can offer no help otherwise.

Let me start by saying that this, as the title suggests, is very much a nuclear option, why ?, because my movies/tv directory on my NAS was previously used with a popcorn hour c-200, slight differences occur in the way it and XBMC deal with fanart and thumbs etc and this is what was causing a lot of trouble for me. One golden rule i have learnt since using XBMC, name ALL of your movies folders 'filmname (year)', accept this fact now or stop reading, it will hold you in good stealth, you are also about to delete every jpg, tbn, fanart, nfo etc you have in your archive.

Here we go....

1) unmount/remove source in XBMC then shutdown XBMC so you are in windows only

2) go into windows my computer and navigate to the root of your movies folder, then in the search box at the top type in .jpg, let it find them all within that folder then delete them, do this for any other files other than the film itself, ie, .fanart or .tbn and also do the same for .nfo files

3) to do the next step you need to make sure that you have 'show hidden files' option enabled, you can do this in folder options, when you have done this navigate to...

Code:

c:/users/*YOURS HERE*/appdata/roaming/xbmc/userdata/database

(where *YOURS HERE* is your user name, whatever your logged in windows with)

once in there delete the following files..

Code:

textures.db
myvideos.db

next delete the video folder at...

Code:

c:/users/*YOURS HERE*/appdata/roaming/xbmc/userdata/database/thumbnails/video

when your there do a search for .tbn files and delete them all, also do a search for .jpg for good measure but it shouldnt find any .jpg

Ok, now fire up XBMC and go add your movie source, at the settings option make sure that auto update is checked and also scan folders recursivley is checked and also setup your preferences for your chosen scraper, i found moviesdk to work best for me. Now let XBMC do the scrape, mine took about 15 mins on 250 movies.

When the scrape is finished you should have a good scrape, there will be films that have scapred wrong here and there but not to many (depending on your collection), different scrapers give different results especially for weird anime stuff etc, anything that is not right you can correct manually by right clicking and selecting 'reload' or even 'edit title' in some case where you forgot to name a folder correctly. When you are completely happy you want to export the library, In the library, press C or right click and select settings, select the Library Menu Item, then choose Export Video Libary.
select 'Seperate files per entry', Choose yes to export thumbs and fanart, choose yes to overwrite old files. and it will dump the .tbn files for thumbs and fanart in each respective film folder and also write the nfo so if you set up another HTPC you can just import the library and XBMC will already have everything you need there without having to do a scrape again.


Please feel free to correct or add anything i may have missed here.

[Martijn]

when your there do a search for .tbn files and delete them all, also do a search for .jpg for good measure but it shouldnt find any .jpg

Just delete the video folder instead of searching for the files. It will be automaticly re-created.

[isamudaison]

I've found that for movies, "<movie title> (year)" works very well for scraping (with the imdb plugin). However, if you want to use the default tv scraper, having tv series in directories named "<tv show name> (year)" does not work at ALL. It's very bizzarre. I've had to omit the "(year)" from all tv shows EXCEPT Battlestar galactica (which is doubly-weird; it's as if it only understands the year in the name when there is ambiguity).

[clubwerks]

I've found that for movies, "<movie title> (year)" works very well for scraping (with the imdb plugin). However, if you want to use the default tv scraper, having tv series in directories named "<tv show name> (year)" does not work at ALL. It's very bizzarre. I've had to omit the "(year)" from all tv shows EXCEPT Battlestar galactica (which is doubly-weird; it's as if it only understands the year in the name when there is ambiguity).

XBMC 101 right there.

[loady]

Just delete the video folder instead of searching for the files. It will be automaticly re-created.

Great, i changed that bit now..i didnt want to presume it would recreate the folder

I've found that for movies, "<movie title> (year)" works very well for scraping (with the imdb plugin). However, if you want to use the default tv scraper, having tv series in directories named "<tv show name> (year)" does not work at ALL. It's very bizzarre. I've had to omit the "(year)" from all tv shows EXCEPT Battlestar galactica (which is doubly-weird; it's as if it only understands the year in the name when there is ambiguity).

Ok, certainly worth noting, i have had no major issues with TV as i have sickbeard set up on my server and doing post processing all in house, my TV collection is quite small, currently about 10 series, however i had some problems with a few here and there, again, where the popcorn hour box had scraped, so what i did was to move the series out of thier folders all in to one folder and a seperate folder for each series and then got sickbeard to manually process the files, it asked where they were and i pointed it to the folder and it scraped all the banners and other meta data wonderfully and moved them into thier respective folders..of course i had made sure that i had previously added the shows prior to initiating the post process.