试试这个：

df2=spark.sql("select id,label,lead(id)over(order by id) as id_1,lead(label)over(order by id) as label_2 from df  ")
df2.createOrReplaceTempView("df2")
df3=spark.sql("select concat(CONCAT(id, ' ',  label)  ,  ' ',concat(id_1 ,' ', label_2)) as one_col from df2 where id_1 is not null")
df3.show()

+-----------+
|    one_col|
+-----------+
|1 foo 2 bar|
+-----------+

是，尝试使用**concat_ws()和collect_list() + array_join()**函数。

Example:

from pyspark.sql.functions import *
df = spark.createDataFrame([(1, "foo"),(2, "bar"),],["id", "label"])

df.withColumn("temp", concat_ws(" ", *df.columns)).groupBy(lit(1)).agg(array_join(collect_list(col("temp"))," ").alias("new_column")).\
  drop("1").\
  show(10,False)
#+-----------+
#|new_column |
#+-----------+
#|1 foo 2 bar|
#+-----------+