If movie name match (only if you use english language for your primary agent) then add a bonus OS_TITLE_MATCH_BONUS to the score (currently 10).If IMDB id associated to the sub match the IMDB id associated to the video file add a bonus OS_MOVIE_IMDB_MATCH_BONUS to the score.If sub movieKind attribute is not set to "movie" then reduce the score with OS_WRONG_MOVIE_KIND_PENALTY penalty (currently 1000).If subs rating is between 0.1 and 4.1 then reduce the score with OS_SUBRATING_BAD_PENALTY penalty (currently 100).If subs rating is 0.0 or 10.0 then add a bonus OS_SUBRATING_GOOD_BONUS to the score (currently 20).If subs have the flag SubBad then reduce the score with OS_BAD_SUBTITLE_PENALTY penalty (currently 1000).Currently order in the response is not taken into account due to preconisation on their forums.Default score for each subs is set to 50.Search database by using video file hash ans size.Depending of the data and the result of the comparaison a bonus or a penalty is applied to the score of the subtitle. The main idea is to compare metadata associated to the subtitle file in the database and the metadata associated to the video file in your local Plex database. Open issues on Github with usefull information (like video name, hash, size, sub language and logs).Share your comments and results on the Plex Forums on this thread.Try to used it to check if results are improved.Don't forget to rename it with the name "OpenSubtitles.bundle" (remove the tag reference in the bundle name). Replace the old metadata agent (OpenSubtitles.bundle) in Plugin folder by the new one in the zip archive.Download the zip package from this last release: on the Github Release page.So this fork is a mock-up to try to have better result by making some consistency check on the results returned by. Results in Plex: it's difficult to have the good subs for our movie using it. This algorithm isn't efficient enough since the database is fed by many sources which are not rigorous enough and can mixup some data. If no results try with the IMDB id instead of the hash/size (only for movies).Search subs by using hash and size of the video file.The origin version used a simple algorithm: This fork is a prototype for an improvement of the Plex metadata agent used by default to fetch subtitles on. metadata agent for Plex improvment prototype
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |