[SofaKng]

I've finished writing the first version of my .NET library for accessing XBMC-style metadata scrapers. I'm calling the library XScraperLib. Basically this library will allow you to use the XBMC XML metadata scrapers without requiring any modification.

I've tried to make this library as easy to use as possible and I've documented every class, property, and method as well as included a readme.txt with several examples on how to use this library.

Also included is a demonstration application that can also be used a scraper editor/debugger. It includes several scrapers that I recently downloaded from XBMC and all of them have been tested and work fine. The editor/debugger allows you to call any scraper XML function (with any parameters) and then receive the result. You can also pass in URLs or specify local files and have it's content retrieved before calling the function. Finally, an XML tree view is included to view the results of the scraper function in an easy to read tree.

The project is currently hosted on Google Code: http://code.google.com/p/xscraperlib/

...and the latest downloads can be found here: http://code.google.com/p/xscraperlib/downloads/list
...and the readme.txt can be found here: http://xscraperlib.googlecode.com/svn/tr...readme.txt

Finally, here are some screenshots of the demo application and class diagram:



Huge thanks to spiff for all of this help!

[spiff]

niiice

does it handle the fluff as well (addon.xml etc)?

[SofaKng]

I'm not sure what you mean by fluff but it reads the addon.xml files to locate scrapers. (it uses the same folder structure as the XBMC addons so you can just copy the entire folder such as metadata.imdb.com)

It also builds a list of required addon imports and generates an XML tree containing the functions from all imports so you end up with a complete XML tree with all functions needed.

Other stuff like the extension point is also read to determine if the scraper is for movies, tv, or music. (Right now it one reads the first extension point so scrapers that parse several media types may not work correctly but I don't know if they even exist?)

[spiff]

cool. can you generate the addon.xml's from the ui as well?

there are no scraper implementing several extension points. if we want to add this, we need to change semantics, we cannot use GetDetails CreateSearchUrl etc, but rather we would need to have dedicated functions per content type. not worth it imo.

[SofaKng]

Not really... the application is just a demonstration program for the library but it can also execute/debug scraper functions. However, it would fairly trivial to create an addon.xml editor.

Also, doesn't each extension point refer to a different library XML file? That's why it seemed possible to have addons with multiple extension points, etc, but if they can't then it's even better (easer).

Anyways, my end goal is to create yet another metadata manager because I'm really not happy with anything else out there. Ideally I'd like something that just works automatically and monitors the file system, etc, and then allows you to go back and make changes if needed. It also seems silly that so many different scrapers exist for the same function (eg. tvdb.com) so I'm hoping this library could be implemented into other media managers and they would then instantly have access to 50+ (or whatever) scrapers that exist for XBMC.

[spiff]

AH yes, you are correct that one add-on can consist of multiple scrapers. i thought you meant several scrapers in one xml for some reason. my bad