在 pyspark 数据帧中插入数据时出错

Question

I have a sample pyspark code where I am trying to generate a json structure. Below is the code

def func(row):
    temp=row.asDict()
    headDict = {}
    headDict['type'] = "record"
    headDict['name'] = "source"
    headDict['namespace'] = "com.streaming.event"
    headDict['doc'] = "SCD signals from  source"
    fieldslist = []
    headDict['fields'] = fieldslist
    for i in temp:
        fieldslist.append({i:temp[i]})
    return (json.dumps(headDict))
if __name__ == "__main__":
    spark = SparkSession.builder.master("local[*]").appName("PythonWordCount").getOrCreate()
    payload=udf(func,StringType())
    data = spark.createDataFrame(
        [
            (1, "a", 'foo1'),  # create your data here, be consistent in the types.
            (2, "b", 'bar'),
            (3, "c", 'mnc')
        ],
        ['id', 'nm', 'txt']  # add your columns label here
    )
    df=data.withColumn("payload1",payload(struct([data[x] for x in data.columns])))
    df.show(3,False)

I am getting an error while inserting data into dataframe

  raise ValueError("Unexpected tuple %r with StructType" % obj)
ValueError: Unexpected tuple '{"namespace": "com.streaming.event", "type": "record", "name": "source", "fields": [{"txt": "mnc"}, {"id": 3}, {"nm": "c"}], "doc": "SCD signals from  source"}' with StructType

If I am trying to print the json payload I am getting correct output

{"namespace": "com.streaming.event", "type": "record", "name": "source", "fields": [{"txt": "mnc"}, {"id": 3}, {"nm": "c"}], "doc": "SCD signals from  source"}

I have also verified this is a valid json.

I am not sure what I am missing here.

Could this be a python version issue?I am using python 2.7

Update-I tried to run the exact same code using python 3.7 and it is running fine

Answer 1

it works for me in spark 3.x with python 2.7.x.,

Welcome to
      ____              __
     / __/__  ___ _____/ /__
    _\ \/ _ \/ _ `/ __/  '_/
   /__ / .__/\_,_/_/ /_/\_\   version 3.0.0
      /_/

Using Python version 2.7.17 (default, Jul 20 2020 15:37:01)
SparkSession available as 'spark'.

results from pyspark shell

import json
from pyspark.sql.functions import * 
from pyspark.sql.types import *

def func(row):
    temp=row.asDict()
    headDict = {}
    headDict['type'] = "record"
    headDict['name'] = "source"
    headDict['namespace'] = "com.streaming.event"
    headDict['doc'] = "SCD signals from  source"
    fieldslist = []
    headDict['fields'] = fieldslist
    for i in temp:
        fieldslist.append({i:temp[i]})
    return (json.dumps(headDict))

spark = SparkSession.builder.master("local[*]").appName("PythonWordCount").getOrCreate()
payload=udf(func,StringType())
data = spark.createDataFrame([(1, "a", 'foo1'),     (2, "b", 'bar'),    (3, "c", 'mnc')],['id', 'nm', 'txt'])
data.show()
'''
+---+---+----+                                                                  
| id| nm| txt|
+---+---+----+
|  1|  a|foo1|
|  2|  b| bar|
|  3|  c| mnc|
+---+---+----+
'''


df=data.withColumn("payload1",payload(struct([data[x] for x in data.columns])))
df.show(3,False)
'''
+---+---+----+----------------------------------------------------------------------------------------------------------------------------------------------------------------+
|id |nm |txt |payload1                                                                                                                                                        |
+---+---+----+----------------------------------------------------------------------------------------------------------------------------------------------------------------+
|1  |a  |foo1|{"namespace": "com.streaming.event", "type": "record", "name": "source", "fields": [{"txt": "foo1"}, {"id": 1}, {"nm": "a"}], "doc": "SCD signals from  source"}|
|2  |b  |bar |{"namespace": "com.streaming.event", "type": "record", "name": "source", "fields": [{"txt": "bar"}, {"id": 2}, {"nm": "b"}], "doc": "SCD signals from  source"} |
|3  |c  |mnc |{"namespace": "com.streaming.event", "type": "record", "name": "source", "fields": [{"txt": "mnc"}, {"id": 3}, {"nm": "c"}], "doc": "SCD signals from  source"} |
+---+---+----+----------------------------------------------------------------------------------------------------------------------------------------------------------------+
'''