[
https://issues.apache.org/jira/browse/MAPREDUCE-2965?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13104306#comment-13104306
]
Vinod Kumar Vavilapalli commented on MAPREDUCE-2965:
----------------------------------------------------
The patch looks fine overall. Some comments:
- We can reuse the NumberFormat definitions for various IDs, no need to repeat
them in each class.
- Several needed null-checks are missing in {{toString()}} and
{{comparesTo()}} methods in all the IDs.
- Looks like after your adding the licenses to the two tests, the allowed
rat-warnings can be set to zero in dev-support/test-patch.properties.
It just occurred to me that the way we are doing the {{hashCode()}},
{{equals()}} and {{comparesTo()}}, we are not synchronizing them as we like to
avoid issues like MAPREDUCE-2954. But at the same, I did find that each of
those IDs can have some fields set to zero when a thread calls {{getProto}} on
them and the builder is still in progress. This seems to indicate that we will
have very unreliable {{hashCode()}}, {{equals()}} etc.
> Streamline hashCode(), equals(), compareTo() and toString() for all IDs
> -----------------------------------------------------------------------
>
> Key: MAPREDUCE-2965
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-2965
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mrv2
> Affects Versions: 0.23.0, 0.24.0
> Reporter: Vinod Kumar Vavilapalli
> Assignee: Siddharth Seth
> Fix For: 0.23.0, 0.24.0
>
> Attachments: MR2965_v1.patch, MR2965_v2.patch
>
>
> MAPREDUCE-2954 moved these methods to the record interfaces from the PB impls
> for ContainerId, ApplicationId and ApplicationAttemptId. This is good as they
> don't need to be tied to the implementation.
> We should do the same for all IDs. In fact some of these are missing for IDs
> like MR AM JobId, TaskId etc.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
