Github user andrewor14 commented on the pull request:
https://github.com/apache/spark/pull/1222#issuecomment-66820412
Hey @vanzin I just took a complete pass over the whole thing. I'm actually
somewhat confused by what the new format the event log file takes. Based on my
understanding, it is now readable and contains key-value pairs that looks like
`k=v`. But according to the comments we also log `[len][bytes]`? What exactly
does it look like as of the latest commit? Is it not like the following?
```
SPARK_VERSION=1.2.0-SNAPSHOT
COMPRESSION_CODEC=org.apache.spark...
=== HEADER END MARKER ===
// json
```
Another thing is I noticed that we no longer test the code path that reads
legacy event logs. I think the way this PR is structured makes this hard to
test. If we move the logic that reads these logs back into
`EventLoggerListener` then we can just keep the old test code for free.
---
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]
