jsonSpark 0.0.2

This is a wrapper package for pyspark to process json files. It pythonifies the json pyspark object.

JsonSpark

This package is meant to give a python simplicity and feel to pyspark while handling json files.

It is very simple to use and doesn't need extra information if you are using python.

Installation

pip install jsonSpark

Sample Usage:

• Import the package
  import jsonSpark

• Pass the pyspark json file object
  df = sql.read.json("filename", multiLine=True) # or get from S3 bucket

• Create a JsonSpark object.
  df = jsonSpark(df)

• See the schema if you wish.
  df.printSchema()

• Display the Data
  df.show()

• Use it as python dictionary
  df["key1"]["key2"]["key3"]["key4"].show()

• You can use the pyspark functions by converting the object back to pyspark object.
  pysparkObject = df._toDF()

I will update the documentation and include a working example soon ....