所以你要在这里集中注意力 txt 以及 uppertx 值应以“|”分隔。你可以试试下面的代码。

# Load required libraries

from pyspark.sql.functions import *

# Create DataFrame

df = spark.createDataFrame([(1,"a","foo","qwe"), (2,"b","bar","poi"), (3,"c","mnc","qwe")], ["id", "nm", "txt", "uppertxt"])

# Concat column txt and uppertxt delimited by "|"

# Approach - 1 : using concat function.

df1 = df.withColumn("concat_val", concat(df["txt"] , lit("|"), df["uppertxt"]))

# Approach - 2 : Using concat_ws function

df1 = df.withColumn("concat_val", concat_ws("|", df["txt"] , df["uppertxt"]))

# Display Output

df1.show()

输出

+---+---+---+--------+----------+
| id| nm|txt|uppertxt|concat_val|
+---+---+---+--------+----------+
|  1|  a|foo|     qwe|   foo|qwe|
|  2|  b|bar|     poi|   bar|poi|
|  3|  c|mnc|     qwe|   mnc|qwe|
+---+---+---+--------+----------+

您可以在spark文档中找到有关concat和concat的更多信息。
我希望这有帮助。