Yijun Xie

> Given this is a PR with more than 1k lines of changes and touches many different places, could you provide a guidance on how to review the code or...

@mbkupfer We're adding an API and context manager to manually drop the cached table. PR https://github.com/snowflakedb/snowpark-python/pull/549 The automatic cleanup will need to explore the underlying query plan data structures because...

That's correct. You need to either use `drop_table` or the `with` statement for each `dff`. This use case is a good candidate for automatic clean up. But we'll need to...

@frsann this already works today ```python df1 = session.sql("select 1 as a, 2 as b") df2 = session.sql("select 1 as a, 2 as b") df3 = df1.join(df2, df1["a"] == df2["a"])...

@sfc-gh-sfan Does the SP server check the arguments of the handler function? We can remove this validation.

Hi @mbkupfer thank you for putting out the PR. Agree that it's important to clean up cached tables. We'll discuss the solution in the next week. Will keep you updated....

My understanding is usually when we `group by` a column the result has the group by column and the aggregate column. Your workaround is a good workaround. Another way to...

@sfc-gh-aalam I'm very confused why this test failed from this change. It looks irrelevant to StringType. `FAILED tests/integ/test_udf.py::test_udf_timestamp_type_hint`

@sfc-gh-hchaturvedi Do you happen to know anything about this?

@brian-kalinowski-sonos In Snowflake, int and long are the same as `number(38, 0)`. float is the same as double. Refer to https://docs.snowflake.com/en/sql-reference/data-types-numeric