error during delta table merge with CHAR data type

I think it would be better to use STRING data types instead of CHAR

Hi Winnie2024 

Thanks to SachinNandanwar 's suggestion. Based on my test, it is very helpful! 

PySpark doesn’t have a direct CHAR or VARCHAR type, you can use StringType() or "STRING" to represent the VARCHAR(15) and CHAR(3) data types.

Try

from delta.tables import DeltaTable
from pyspark.sql.types import StringType

# Create a Delta table with two columns
DeltaTable.createIfNotExists(spark) \
    .tableName("test") \
    .addColumn("id", StringType()) \
    .addColumn("code", StringType()) \
    .execute()

or

from delta.tables import *
from pyspark.sql.types import *

DeltaTable.createIfNotExists(spark) \
    .tableName("test") \
    .addColumn("id", "STRING") \
    .addColumn("code", "STRING") \
    .execute()

Based on my test, both of the above work for your later merge test code without any error.

Best Regards,
Jing
If this post helps, please Accept it as Solution to help other members find it. Appreciate your Kudos!