[small_frenchy]

Okay, I've done it.... I have a setup with a working DLL and the originals scrapers from svn included in the setup. This is just a small app in .Net that use XBMCDLL which is generated from the original code. Let me know if anyone is interested and how to transfert the setup anywhere (the setup is about 10Mb, I know, the XBMC dll take a large place for now because there too much things in it )

[spiff]

i'm impressed

[small_frenchy]

It's not a real good work for now. There is a lot of thing to cleaned from the dll. But it seems to work fine. How can I send the setup to you (if want to test it, for sure, there will be bugs I'm not a C++ dev so...)

[spiff]

mail it. spiff at xbmc dot org

[small_frenchy]

I have just send it. I think all dependencies are ok, let me know if it doesn't work. For now, it work only for video scrappers.
[UPDATE] First dependency bug : It miss MSVCR71.dll
[UPDATE] Encoding bugs also : there is some encoding bugs, with allocine.xml scraper for exemple

[small_frenchy]

For testing :
http://www.4shared.com/file/118290170/32...ester.html

Dependencies problems are resolved

[small_frenchy]

A little enhanced version. Some little (too litlle) cleanup in the main dll. Double click on fanart / thumb to see images. Better support of callbacks from the main dll to the .Net one.
I have test a lot of scraper, and it works fine in my opinion but only for movie scraper.
download the setup here :

http://www.4shared.com/file/118524777/4d...ester.html

I will publish sources (in fact a modified version of the XBMC project and the .Net wrapping) when the code will be more clean.

[Nicezia]

you obviously know what's going on in the scraper code, and i've got an almost complete framework done supporting all scraper types (minus the stuff i don't understand from the C++ code).


I'd like to ask for your help, cause i can barely follow the flow of the c++ code and mine is intended to be natively .Net, without the need for wrapping code, and i'm sure spiff is tired of me bothering him about what's going on in his code.

[spiff]

nah i have no issues with explaining stuff. as i have said before it's what i get for refusing to document stuff that being said i totally agree that joining efforts is your best choice of action

[Nicezia]

and i'm just the opposite, i've commented almost every line of my code.

[Nicezia]

My parser is solid by the way, fortunately, i could read and understand everything in the parser code, settings & scraper info, and the other things used by the parser work just like xbmc's parser. (though the functions are differently implemented due to using native .Net features rather than writing a completely new Regular Expression Engine (which i tried at the start and decently implemented but a few things i weren't happy with - then a svn error wiped the whole thing away) and as you said, XBMC can handle little discrepencies like there not being a space before the function attribute in that one scraper

so if i could just get some help figuring out exactley what goes on before running the parser (during and after the parser i've already got handled) then i can finish this completely within a day's time.

instead of trying to untangle Scraper code from XBMC why not just help me understand the little bit of code i need to understand to wrapp this up. And we can all be happy.

the only part of my code that needs any work is setting buffers before running the scraper functions.

[small_frenchy]

I'm not sure that I will be able to help you as you want. Anyway, I can do my best but I will not do as good as spiff.

In fact, I haven't understand a lot a thing. I've just study the call stack with the visual studio debugger at scraper call from xbmc. Then I have just fake necessary input parameter (to handle loading / unloading dll xbmc system, and also xbmc file system handling) and export functions to scrap and set callback to .Net.

Anyway, the big advantage I see in the manner is the fact that it follow the xbmc code as it grow or change. I can update from svn the scraping objects and that's all (except if the definitions of objects change).

But be sure that if I can help you in anyway I will do. Let me know.

For me, I'll continue the dll way, I have to handle ScraperSettings then continue it for tv show and music. I think the hard part is done, the rest is only more exports from the dll.

(PS : sorry for my poor english, I do my best...):o

Another version with settings handling : http://www.4shared.com/file/118733463/bc...tings.html

[Nicezia]

I'm not sure that I will be able to help you as you want. Anyway, I can do my best but I will not do as good as spiff.

In fact, I haven't understand a lot a thing. I've just study the call stack with the visual studio debugger at scraper call from xbmc. Then I have just fake necessary input parameter (to handle loading / unloading dll xbmc system, and also xbmc file system handling) and export functions to scrap and set callback to .Net.

Anyway, the big advantage I see in the manner is the fact that it follow the xbmc code as it grow or change. I can update from svn the scraping objects and that's all (except if the definitions of objects change).

But be sure that if I can help you in anyway I will do. Let me know.

For me, I'll continue the dll way, I have to handle ScraperSettings then continue it for tv show and music. I think the hard part is done, the rest is only more exports from the dll.

(PS : sorry for my poor english, I do my best...):o

Another version with settings handling : http://www.4shared.com/file/118733463/bc...tings.html

actually i think i'm just about done now, once spiff explained to me how multiple urls are handled (and i learned that its a good idea to use the Visual studio help to look up things i don't understand in c++) I still may need a little help in understanding the flow of the scraper code if anything changes... but for now i think i'm up to speed.

[small_frenchy]

Okay,let me know if I can help you.