# Hive python self-defined udf function error

``````add file /home/sdev/yanan/udf.py;
select TRANSFORM (news_entry_id)
USING 'python udf.py'
AS (comb)
from tmp.yanan_gbdt
where p_date='20180708'
limit 10
;
``````

I have defined a function to process the field of `news_entry_id`. The `tmp.yanan_gbdt` table take the `p_date` as partition. The codes works well.

But if I get rid of this condition `where p_date='20180708'` that I do not specify the partition, there comes the error: `FAILED: NullPointerException null` So what's wrong with this ?