Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/5368#issuecomment-89770686
I think this makes more sense, although I think it might even be a smaller
change and more widely applicable if it were an `InputFormat` decorator as
@sryza mentioned. I don't know if for some reason that becomes hard to make
work. A whole new RDD and a 'reliable' method (is the current method
unreliable?) still seems like a lot of weight, to me, to deal with user-space
problems, and ones that seem straightforward to solve directly (i.e. taste and
delete bad files if you expect bad zip files). Transient failures can be helped
by the internal retry logic but if the upstream data is just unreadable, how
far do you go to make sense out of it?
---
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]
