[
https://issues.apache.org/jira/browse/AVRO-513?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12878791#action_12878791
]
Tom White commented on AVRO-513:
--------------------------------
> So all objects in the queue must be unique.
You're right - because we are buffering in the queue, we need to make sure that
all instances in the queue are distinct copies. So the pooling idea you
mentioned above might be a good optimization for later.
> The risk that user code swallows the InterruptableException can be fixed by
> setting 'done=true' before calling interrupt().
I think this is correct.
We need to do something about the close() method - at least document that if
overridden in subclasses they must call super.close() as the first line in the
overridden method.
> java mapreduce api should pass iterator of matching objects to reduce
> ---------------------------------------------------------------------
>
> Key: AVRO-513
> URL: https://issues.apache.org/jira/browse/AVRO-513
> Project: Avro
> Issue Type: Improvement
> Components: java
> Reporter: Doug Cutting
> Assignee: Doug Cutting
> Fix For: 1.4.0
>
> Attachments: AVRO-513.patch, AVRO-513.patch, AVRO-513.patch
>
>
> The Java mapreduce API added in AVRO-493 requires reducers implementations to
> explicitly detect sequences of matching data.
> Rather the reduce method might better look something like:
> void reduce(Iterator<IN>, Collector<OUT>);
> Where all equal values are passed in a single call.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
