movies_metric.yaml

steps:
- dataFrameName: moviesWithRatings
  sql:
    SELECT userid,
           movies.movieid,
           rating,
           timestamp,
           title,
           genres
    FROM ratings
    JOIN movies ON ratings.movieid = movies.movieid
- dataFrameName: fantasyMoviesWithRatings
  sql:
    SELECT movieId,
           cast(rating AS float) AS rating,
           timestamp,
           title,
           genres
    FROM moviesWithRatings
    WHERE genres LIKE '%Fantasy%'
- dataFrameName: topFantasyMovies
  sql:
    SELECT movieId,
           title,
           avg(rating) AS averageRating
    FROM fantasyMoviesWithRatings
    GROUP BY movieId,
             title
    ORDER BY averageRating DESC
    LIMIT 100
- dataFrameName: myFavoriteMovieRated
  sql:
    SELECT *
    FROM topFantasyMovies
    WHERE title = ${myFavoriteMovie}
output:
- dataFrameName: topFantasyMovies
  outputType: Parquet
  outputOptions:
    saveMode: Overwrite
    path: topFantasyMovies.parquet
- dataFrameName: topFantasyMovies
  outputType: CSV
  outputOptions:
    saveMode: Overwrite
    path: topFantasyMovies.csv
    coalesce: true
    extraOptions:
      multiline: "true"
- dataFrameName: topFantasyMovies
  outputType: JSON
  outputOptions:
    saveMode: Overwrite
    path: topFantasyMovies.jsonl
    repartition: 10