从 JSON 中提取少量字段并在 Pyspark Dataframe 中以地图的形式返回其余部分

Question

I am reading a streaming data in pyspark dataframe, the data contains few fields which are present in every data/request. I want to exact those fields and create a dataframe column for it and want to store the rest of fields as map in another dataframe column. I am not able to achieve it

If someone can help with it?

Example:

Sample Values :

{"event1":"Value","event2":"Value","event3":"Value","event4":"Value","event5":"Value","event6":"Value"}
{"event1":"Value","event2":"Value","event3":"Value","data1":"Value","data2":"Value","data3":"Value"}

Now suppose event1,event2,event3 is present in every row, so I want to extract it and make it as a separate dataframe column and rest of the fields as map of key values pairs which will be another dataframe.

Answer 1

You need to create a schema for your dataframe and use from_json to convert it to StructType in spark. Then you are able to select your specific event's and create another dataframe for other events.