Scraper request - www.csfd.cz [CZ] (1 Viewer)

[no.diggity]

Still dont know, where the problem is. Just made a new installation of MP RC2 and latest version of the moving pictures plugin, also used last version of your CSFD scraper and it just keeps getting info from imdb.com. I tried to disable all other scrapers, that does not help either. I use Stream MP skin, which includes moving pictures plugin (latest version). Wondering, if it has something to do with my problems...

 

[Trottel]

I have last StreamedMP skin with included Moving Pictures plugin too (version StreamedMP-1.0.3.1223). Do you use NFO files with movie information in movie folder? If you have in Importer Settings enabled NFO Scanner and "Auto-Approve on NFO File IMDB Match" it could have larger priority than scraper script.

 

[no.diggity]

I don't use those, tried all these settings disabled in import page, but no change...

 

[ltfearme]

Perhaps you can attach your logs no.diggity with scraper debug enabled. The skin plays no part in this process.

 

[no.diggity]

Here is log file, thanks for checking.

 

[RoChess]

Here is log file, thanks for checking.

I scanned over log file real quick, and I see a lot of failures that are caused by bad filenames that fail to be cleaned up properly by the noise filter you use (advanced settings, you probably rely on default one). I can help you create a better noise filter, but it might be easier for you to just rename your files.

For example "WHITE HEAT Xvid DVDRip METALOONA.avi" is seen as movie title "White Heat Metaloona", which then fails.

I also see lot of mistakes in this CSFD.CZ scraper, but you are importing too many movies at the same time, and the log file shows all movies being scraped asynchrone. Meaning that instead of AAABBBCCC order for movies A, B, C, I'm looking at ABCAABCCB type of order, which is extremly confusing.

What I need is a proper filename (as to avoid noise filter problems which is something different), such as "Avatar [2009].avi" that fails to get scraped correctly by the CSFD.CZ scraper, then upload the scraper-debug enabled log file, like you already did (should be a 1-2MB file, so please zip it).

Then it shouldn't be hard for me to fix all the problems (unless Trottel beats me to it).

On a side note, there are major problems with IMPAwards website right now, they are being fixed, but requires a new version of MovingPictures. So please be patient on that, so disable the IMPAwards scraper, or put TheMovieDb cover/fanart scrapers at a higher priority in the mean time.

 

[Trottel]

Then it shouldn't be hard for me to fix all the problems (unless Trottel beats me to it).

If there are some mistakes in my scraper script you are more than welcome to fix them.

 

[no.diggity]

Ok. I tried to import one movie only, named as "A Bronx Tale.mp4", debug file enclosed. Thanks again for checking.

 

[no.diggity]

I got your reply with

----
Ok, all I can see is that you have XBMC (Local) at top priority, which skips the file due to lack of NFO file, and then you have IMDb.com v1.4.7 scraper (v1.5.3 is out btw) which finds the movie and auto-approves it, thereby skipping any other scraper.

So it doesn't allow me to see at all what csfd.cz scraper might do.

Please put csfd.cz scraper at top position and try again.
---

Problem is, I got CSFD scraper on top position...

I tried again with Data sources set to "automatically retreive movie data in the - czech - čeština", before it was set to "Manually manage movie data sources" and (I checked it again) the CSFD scraper was at top position. Maybe this will be more helpful.

 

[RoChess]

Ok. I tried to import one movie only, named as "A Bronx Tale.mp4", debug file enclosed. Thanks again for checking.

Ok, you are getting no correct response back from the time you start to query csfd.cz website, so I've added some extra safeguards. I just can't test it myself quick, so this might end up being a long trial&error sequence with me making small modifications and you testing them. If you have an IRC client, then join us at Freenode on #moving-pictures, so that we can do this in real-time (assuming I'm around).

For now, try the attached v0.1.4