Netflix Sample

Setup

This dataset requires access to Kaggle.

From the samples/netflix folder.

cd input

sh download_data.sh
python download_cleanup.py

cd ..

# Applies to DGraph
sh publish.sh

This will apply data to your local DGraph instance. Then you can try an example queries below.

Example Queries

Query 1: Find all Actors related to Horror Movies in 2018

{
  q(func: type(genre)) @filter(eq(identifier, "Horror Movies"))
    {
        name: identifier
        titles: ~genre @filter(type(title) AND eq(release_year, 2018)) {
        name: title
        description
        cast {
          name: identifier
        }
      }
    }
}

Query 2: Find all the Movies that Robert Downey Jr is in and the associated Rating

{
  q(func: type(cast)) @filter(eq(identifier, "Robert Downey Jr."))
    {
      name: identifier
      titles: ~cast @filter(type(title)) @normalize {
        name: title
        rating {
          rating: identifier
        }
      }
    }
}

Query 3: Find all the Titles that both Mike Myers and Seth Green are appear in

{
   mike as var(func: type(cast)) @filter(eq(identifier, "Mike Myers")) { uid }
   seth as var(func: type(cast)) @filter(eq(identifier, "Seth Green")) { uid }

    title(func: type(title), orderdesc: release_year)
     @filter(uid_in(cast, uid(mike)) AND uid_in(cast, uid(seth)))
   {
        name: title
        description
        release_year
   }
}

Query 4: Find all Titles where the term 'Autobots' appears in the description

{
  q(func: type(title)) @filter(anyofterms(description, "Autobots"))
    {
      identifier
        name: title
        description
    }
}

Query 5: Find the total counts of titles per Genre

{
  var(func: type(title)) @groupby(genre)
  {
    c as count(uid)
  }

  genre_counts(func: uid(c), orderdesc: val(c))
  {
    name: identifier # from the genre node (grouped)
    count: val(c)
  }
}