Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/1977#discussion_r28003603
--- Diff: python/pyspark/shuffle.py ---
@@ -367,32 +372,13 @@ def iteritems(self):
def _external_items(self):
""" Return all partitioned items as iterator """
- assert not self.data
--- End diff --
I noticed that you moved a few of these assertions. I guess the old
assumption was that once we've spilled, we'll stop using `data` and only
aggregate into `pdata`, given that we clear `data` in the first branch of
`_spill`. Why has this assumption changed here? It looks like we do end up
writing to `data` again inside of this `_external_items` method, but then we
end up clearing `data` at the end after the iterator has been consumed.
Was this change necessary in order to support iterating multiple times over
the merged result? Just want to double-check my understanding here.
---
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]
