Pyspark's Dataframe study (1)

From pyspark.sql import sparksession
spark= sparksession\
                . Builder \.
                appName ("DataFrame") \
                . Getorcreate ()
#1生成JSON数据
Stringjsonrdd = spark.sparkContext.parallelize ((' ' ' {'
    id ': ' 123 ',
    ' name ' : "Katie",
    "age": +,
    "Eyecolor": "Brown"} "", ""
{"
    id": "234",
    "name": "Michael",
    "Age": "
    eyecolor": "Green"} "", ""
{"
    ID": "345",
    "name": "Simone",
    "age" :
    "Eyecolor": "Blue"} ")
#2创建DataFrame
Swimmersjson = Spark.read.json (Stringjsonrdd)
#3创建一个临时表
Swimmersjson.createorreplaceglobaltempview ("Swimmersjson")
#4查看前几行
Print Swimmersjson.show () #默认是10行

#5可以通过编写sql语句查询
Print Spark.sql ("SELECT * from Swimmersjson"). Collect ()

#6打印模式
Print Swimmersjson.printschema ()