Github user holdenk commented on the issue:
https://github.com/apache/spark/pull/13737
@rxin ah sorry, I though you wanted to remove the dependency around the
file loading functionality - that makes a bit more sense. I can take a look at
that possibility in more detail tomorrow if we don't want the simple fix.
At first glance the scala `TestHiveContext` is used inside of
`_createForTesting` in PySpark which is used by `HiveContextSQLTests`). From
what I remember of working in Spark Testing Base there is some useful bits done
inside of the hive testing context base to allow them to work with a test
metastore (but that could be a bit out of date).
I'm not convinced removing the Python dependency on the `TestHiveContext`
is the best way to fix this problem which seems to have been introduced just to
allow the Hive tests to be used in more places but I'll investigate that more
tomorrow.
The problem seems to be that the files aren't always packaged as resources
when you build locally (and in Jenkins it was somehow a race condition some of
them time). It seemed better to try and get a small working fix in quickly and
do the deep dive later.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
