Changing values of a JSON with RDD

Viewed 32

How do you set a value in RDD once you transform it? I am modifying a JSON file with pyspark and I have this case:

categories: [ {alias, title}, {alias, title}, {alias, title} ]

I have made a transformation that creates a list of titles for each row: [title, title, title].

But how do I set the result back to the key categories?

At the end I want to get: categories: [title, title, title]

This is the transformation that I am doing:

restaurantRDD.map(lambda x: x.data).flatMap(lambda x: x).map(lambda x: [row.title for row in x.categories])

There are also multiple transformations from restaurantRDD similar to this one which are modifying some parts of the JSON. How can I apply them both at the same time and then write them in a new JSON file?

Should I use something else instead of RDD?

0 Answers
Related