Closed swapagarwal closed 7 years ago
I can work on this. There is no IMDB API available, so I want to do this: get the IMDB id from TMDB API, then use the IMDB id to construct a full IMDB website link, and scrape the Rating. Is that okay? And yes, can I work on this module while the WIT AI response problem still persists? I think the AI is taught in the older version style, right?
Let's explore more options (avoid scraping). The Wit response structure update doesn't affect jarvis, for old as well as new modules. The effect is seen when you create a new app for testing purposes.
Well, do you have any other option in mind? Please let me know. And if wit response doesn't affect new modules, how do you propose to use WIT if I want to continue working on the Soccer module? I need to create new app so that I can train it about user inputs, right? And Soccer module is new of course. Frankly speaking, I am highly confused at this stage: At one time, you're saying that new modules are not affected, again, you're saying that effect is seen only when creating new apps. So, how contributors can create new modules without creating new apps?
New modules (soccer) in old app (jarvis) are not affected. If you create new app (say jarvis-test), it will have new response structure. While submitting a Pull Request, you only submit the module code. I'll train the new module (soccer) in old app (jarvis), so it'll have the old response structure. Hope that makes sense! 😅
Oh! That's what I was asking for days. When I asked for help, some people in Gitter told me that I need to create own app in WIT to make module. I never knew that you were going to train that for me. It was the only barrier between me and soccer module, I will start on that right away!
So you can create your own app for testing purposes, till your module is integrated. Sorry for the misunderstanding! 😞
Okay, about the IMDB rating: you already know that OMDB provides api that can fetch IMDB rating, what if we use OMDB just for fetching the Rating?
OMDb API is not free anymore, hence we moved to TMDb.
What about this: http://imdbpy.sourceforge.net/
IMDbPY looks promising. If you were to use that to get an IMDB rating, I think it would make sense to just rely on it for all of the information that the module fetches.
Then I am starting to work on it.
Sent from my Xiaomi MI 5 using FastHub
Done. See #268
happyness movie
returns Rating = 7.7 This rating is from TMDb: https://www.themoviedb.org/movie/1402-the-pursuit-of-happyness IMDb rating is 8.0 (http://www.imdb.com/title/tt0454921/)Options: