[
https://issues.apache.org/jira/browse/PIG-290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12614547#action_12614547
]
Alan Gates commented on PIG-290:
--------------------------------
I don't think it's quite fair to characterize this is as relying a on side
effect. Santhosh's changes guarantee that side effect and then rely on it. I
agree that the resulting API is a little confusing (using disconnect/connect
will have a different affect that using replaceNode). That needs to be very
clearly documented. But I'm sure it would be any better if operators had
inputs and anyone rearranging the tree had to remember to reconnect the inputs.
In that case the programmer would have to remember to reset the inputs, which
is just as hard as remembering to use the correct API function.
> LOCross output schema is not right
> ----------------------------------
>
> Key: PIG-290
> URL: https://issues.apache.org/jira/browse/PIG-290
> Project: Pig
> Issue Type: Bug
> Components: impl
> Affects Versions: types_branch
> Reporter: Pi Song
> Assignee: Santhosh Srinivasan
> Fix For: types_branch
>
> Attachments: insert_between.patch
>
>
> From the schema generation code:-
> {noformat}
> List<LogicalOperator> inputs = mPlan.getPredecessors(this);
> for (LogicalOperator op : inputs) {
> // Create schema here
> }
> {noformat}
> The output schema is generated based on inputs determined in the logical
> plan. However, mPlan.getPredecessors() doesn't always preserve the right
> order (A x B and B x A result in different schemas). I suggest maintaining
> mInputs variable in LOCross (as it used to be) to resolve this issue.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
