[RELEASE] Movieplayer.it scraper for Movies new version (Italian) - Printable Version
+- XBMC Community Forum (http://forum.xbmc.org)
+-- Forum: Help and Support (/forumdisplay.php?fid=33)
+--- Forum: Add-ons Help and Support (/forumdisplay.php?fid=27)
+---- Forum: Metadata scrapers (/forumdisplay.php?fid=147)
+---- Thread: [RELEASE] Movieplayer.it scraper for Movies new version (Italian) (/showthread.php?tid=60585)
- gulp - 2011-07-27 19:01
Also for me, collecting all infos from Imdb, and retrieve only plot, and possibly posters from italian site seems the best way (personally i prefer MoviePlayer and FilmUp over ComingSoon and MyMovies because they have shorter plot description, i dont like to read for half an hour for a plot ).
In this way we also use function from official scraper, that is always updated.
I've just visited film.it, nice, i didin't know it.
Maybe we can switch to film.it (that has a better layout and the original title) and use imdb.com for all details.
Btw, now MP scraper seems good enough, but another option is always good!!
- KoTiX - 2011-07-27 21:48
Using MP as secondary search will probably cause false results because we will be forced to retreive the "first" result instead of the "best" result and we could have a plot that doesn't match the movie.
Personally I like more the idea of scraping all the infos just from one main site and leave the other sites's (IMDb, TMDB, etc..) things as an option, so right now i'm really thinking to stop developing the movieplayer scraper (for movies at least) and leave it for another site that have all the stuff needed by xbmc. I just have to find out which one is the best.
- henrava - 2011-07-28 01:11
KoTiX Wrote:Using MP as secondary search will probably cause false results because we will be forced to retreive the "first" result instead of the "best" result and we could have a plot that doesn't match the movie.Kotix, can you explain a little more ? I'm totally noob in the scraper's world and I'd love to understand how work a scraper.
- KoTiX - 2011-07-28 09:27
A scraper is made exentially of these sections:
1) - Create Search url
2) - Get Search results
3) - Get Details
4) - Special functions
1) is the url used for searching the main movie page (Google, Bing or MP in our case)
2) get the main movie page from the results of 1)
The priority of the results can be left the same or adjusted by xbmc to perfectly match the movie (sometimes is better to leave to Google this) or the user can choose manually the result in xbmc.
3) From the main movie page retreive all the detailed infos
4) For all the infos that aren't in the main page, exist some special "url functions" that are called inside the 3)
For these functions (IMDb, TMDB stuff etc.) to work properly we need the "original title", with it we will find the IMDb tt...... code of the movie that point to the right movie using this page http://akas.imdb.com/find?s=tt;q=moviename.
Once we have the right tt.... code we can have fanarts from TMDB, all the imdb stuff etc.
Now, if we want IMDB as 1) and MP as 4) we will need to find the right movie in MP using the "url functions", this mean that whatever search engine we use (Bing or MP) we will use always the first result of this search and we cannot even choose it manually.
- henrava - 2011-07-28 16:55
I see now a little more.
Is not possible to have a first search from imdb (among other things there is a imdb.it with info in italian!) and a second search from a italian site ?
1 search -> fill all info avalaible
2 search -> fill info not retrieved by first one
In this way we don't need the point 4) - special functions, that seems less flexible.
Is possible also my words are a flood of bullshit.
- KoTiX - 2011-07-28 17:21
No, just one search and all the special functions you want.
- gulp - 2011-07-29 11:22
with film.it probably we can colect all information from imdb, and use film.it as secondary source with a very high accuracy (impossible with movieplayer).
Film.it in-site search, uses original title + italian titile + year, all information we can easily find on imdb and use to search on film.it. And has also a good tv-series section.
- KoTiX - 2011-07-29 16:18
I've already started to make some tests modding the imdb.de scraper to make an imdb.it one and retreive the plot from comingsoon.it
The results aren't very well for now, for example looking for the movie "afterlife" i got the right stuff from imdb but comingsoon find me "Resident evil: afterlife" as plot.
I'll try with film.it too, but IMO is better to keep imdb as secondary search.
I'll keep you update.
- KoTiX - 2011-07-31 01:11
Ok, i hope the german guys in the other thread don't kill me, but here is a quick and dirty conversion of the imdb.de scraper to imdb.it
Most of things work good, the plot is taken from comingsoon.it using googlecache but it is still a little bit buggy because for now it works with movies with just one word in the title as "Avatar" and not with something like "lo chiamavano trinità".
I repeat this is just a sample and buggy scraper that i don't even know if i'll keep working on it.
A lot of thanks to Eisbahn and the other developers of the IMDb.de scrpaper.
- henrava - 2011-07-31 16:15
For a first try, it's good.
I tried with a folder with 21 Movies from '70, 20 are correctly recognized, for 1 i have to enter title manually.
Original title missing
votes wrong ( it seems to me are rounded down, so 6.8 become 6)
votes number ok
Plot missing (i tried all the combos imdb.it/comingsoon... nada)
actors ok (possible to add in settings a limit , eg 8 ?)
Poster and fanart ok
So, keep up this very good work !
Sorry if i can't help you more, all i can is debug.