I can do the following which performs a merge on the entire target table. Such as;

#target
target = DeltaTable.forPath(spark, target_path)  

#source
src = updates

#merge
target.alias("target")
    .merge(src.alias("source"), "source.id = target.id and source.cust_name = target.cust_name")
    .whenMatchedUpdateAll()
    .whenNotMatchedInsertAll()  
    .whenNotMatchedBySourceDelete()
    .execute()
   

But how can I perform a merge on a subset of the target such as; merge only for target.id=4 and other existing records in the table should not be impacted by this subset merge at all. Is there a way to achieve this?

from pyspark.sql.functions import col

#pseudocode - trying to filter a subset of target but the following will not work
target = DeltaTable.forPath(spark, target_path).filter(col('tbl_id').isin(4)) 

#source
src = updates

#merge
target.alias("target")
    .merge(src.alias("source"), "source.id = target.id and source.cust_name = target.cust_name")
    .whenMatchedUpdateAll()
    .whenNotMatchedInsertAll()  
    .whenNotMatchedBySourceDelete()
    .execute()