[Mackan2]

Hi.
I updated XBMC by using the ubuntu repositories about a week ago and today. My latest update was around may, and that one worked nicely.

The scraper has since my first update been very bad at finding the correct movies, a title like "Gladiator.2000.720p.HDTV.DTS-ES.x264-ESiR.mkv" will result in it finding ""Gladiators 2000"". If i try to reload it finds some outer movies but not the correct one, Gladiator.

What is new with the scraper?? How can I fix this??

[althekiller]

"The scraper" there are only ~10 in that list...

Please see the sticky about posting problems in a useful manner.

[Mackan2]

Sorry, I'm talking about the IMDB-scraper.

[althekiller]

You still need to read that sticky!

[Mackan2]

OK, I think I'll get it right this time... but my main question is if there has been any change made to the IMDB-scraper over the summer?

Running XBMC pre-9.10 r22290 installed from the SVN repositories on ubuntu 9.04 minimal (2.6.28-15-generic) (i686).
Ran "aptitude update" earlier today...

Things I did to produce the log:
*started the machine
*navigated to my movies
*brought up info for robocop (already scanned to wrong IMDB info)
*hit the refresh button for robocop

This is where I stopped. I got a list with 3 movies to choose from, none was robocop... This would have worked with my earlier SVN install from the beginning of this summer.


http://pastebin.com/f5d3a3700

I guess this is the important part:

Code:

00:20:33 T:2926513040 M:1853755392   DEBUG: InternalFindMovie: Searching for 'robocop 1988' using IMDb scraper (file: 'imdb.xml', content: 'movies', language: 'en', date: '2009-08-10', framework: '1.1')
00:20:33 T:2926513040 M:1853669376   DEBUG: FileCurl::Open(0xbf87377c) http://akas.imdb.com/find?s=tt;q=robocop%201988
00:20:33 T:2926513040 M:1853669376    INFO: easy_aquire - Created session to http://akas.imdb.com
00:20:34 T:2926513040 M:1853407232   DEBUG: FileCurl::Close(0xbf87377c) http://akas.imdb.com/find?s=tt;q=robocop%201988
00:20:34 T:2926513040 M:1853407232   DEBUG: scraper: GetSearchResults returned <?xml version="1.0" encoding="iso-8859-1" standalone="yes"?><results><entity><title>Capital Combat</title><year>1990</year><url>http://akas.imdb.com/title/tt0275994/</url><id>tt0275994</id></entity><entity><title>&#x22;RoboCop: Prime Directives&#x22;</title><year>2000</year><url>http://akas.imdb.com/title/tt0220008/</url><id>tt0220008</id></entity><entity><title>Flames 88-89</title><year>1989</year><url>http://akas.imdb.com/title/tt0190396/</url><id>tt0190396</id></entity></results>

especially this url: http://akas.imdb.com/find?s=tt;q=robocop%201988
Has the scraper always used both title and year for the search? I think that is whats messing it all up... Since I'm positive I have managed to scrape gomorra to my library before, and now it scrapes like this: http://akas.imdb.com/find?s=tt;q=gomorra%202008 which is bad, http://akas.imdb.com/find?s=tt;q=gomorra on the other hand results in correct answers.

[badmoviesnbeer]

nm

[Mackan2]

The file names follow some sort of standard scene convention and it worked before in earlier XBMC versions... There must be a better solution.

[Mackan2]

Any ideas how to solve this, am I the only one with this problem?? I guess the parsing has changed in later versions, how do i change it back?

[DanTheHam]

I know your problem, i have it too i understand where the problem is but don't know how to fix it
It's scraping from akas.imdb.com instead of imdb.com.
akas is for titles in their original language or alternernate title names.

It should be changeable from the settings menu, imdb addon settings. there should be the url which you can remove "akas." from but for me there isn't there is only language choice which doen't alter the results i'm getting