A small addition to CheckMovieHash please?

[stavros_sk]

Hello again, during these few days of extensively testing the database functionality , i have encountered a little 'incovenience' that makes the CheckMovieHash function return a wrong IMDb ID.

Actually, it's not the function's fault, it's just that a very small percentage of users or applications upload the wrong information about a film. (wrong IMDb ID basically).

So.. a believe this would be an efficient enough workaround: If you could add an optional parameter for the language these database entries belong to, this would greatly lower the chance for incorrectly identified film (because of said movihash - IMDbID foreign keys mismatch).

So here's an example of the addition i'm talking about: CheckMovieHash($token, $moviehash, $sublanguageid).


Thanks for everything os, and looking forward to hearing from you soon

[oss]

I dont understand exactly, what is $sublanguageid - it is language of what and why you need it ?

I know there are some errors, and as I wrote before - after addmoviehash there will be needed more functions like reportwrongmoviehash ...

but thats ok. I will work on that function soon I guess, I just need to finish ranking system properly.

[stavros_sk]

Sorry, it was my mistake calling it "$sublanguageid". It's just the language of the subtitle that a user or application uploaded.

So when I call CheckMovieHash( $token, $moviehash) it returns the first found hash (right?) of all the hashes which are found for *all* languages (if i am not mistaken).

For example: if you call this: CheckMovieHash( token, "a94c9aeacd0288d2") you will see that while the actual movie in this video file is "The Dark Knight", the movie this function returns is "Batman: Gotham Knight" which obviously is the wrong film.


Now, if it was possible to call something like this instead: CheckMovieHash($token, $moviehash, $language) and we called for an example CheckMovieHash( token, "a94c9aeacd0288d2", "eng" ), our function would only return the mathces that are contained *only* in English subtitle uploads, thus greatly reducing the probability of returning a wrong movie.

Hope it's more clear now.
Thanks again, and no worries about when the AddMovieHash function will be completed.

Looking forward for TV series support too!

edit: A better example to illustrate how the CheckMovieHash function can fail in enough cases to provide the correct IMDb Id:

For the "Heroes" tv series for the episode "One of Us, One of Them" , if you have the hash of the .avi file which is "53fc6fe84ad5ee31" and you search by this hash using the CheckMovieHash method, it will return the episode "Chapter Two 'The Butterfly Effect' " which is the wrong episode. But if you search with this method in the browser:

http://www.opensubtitles.org/en/search2 ... eid-[b]eng[/b]/moviehash-53fc6fe84ad5ee31

... the search results will include the correct episode!

In many other cases for TV series, if one doesn't search by english or other language that have regural releases updates, the TV Series name will be correctly assumed, but the episode name will not be provided because the IMDb ID which this hash refers to is old, before IMDb provided a seperate page for each individual epsiode. Example:
http://www.imdb.com/title/tt0813715/

Now, i have discovered that SubDownloader actually gets the correct episode information but i'm wondering how it does this without the CheckMovieHash function supporting language preference?

[oss]

nice post. Ok, checkmoviehash() can return more than one IMDBID. SubLanguageID - I can add it, but it has no reason (in different words - checkmoviehash with all english uploaded subtitles)...

Anyway, adding moviehash for no subtitles is problem, because database is not prepared for that, and I need change some tables, which takes more time ((

But I am doing progress slowly on this issue, I will let you know, when it will be ready.

[stavros_sk]

Hi os,

No worries about when this function will be ready and thanks for considering my proposition

Best regards,
Stavros.