hadoop-pig-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alan Gates (JIRA)" <j...@apache.org>
Subject [jira] Commented: (PIG-276) Allow UDFs to have different implementations based on input types
Date Mon, 30 Jun 2008 16:39:45 GMT

    [ https://issues.apache.org/jira/browse/PIG-276?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12609287#action_12609287
] 

Alan Gates commented on PIG-276:
--------------------------------

Patch looks very good.  A couple of comments:

# In min and max functions, initial values of, for example, Integer.MAX_VALUE and Integer.MIN_VALUE
are being used.  In the case where all nulls or an empty bag are passed to the function, this
will result in those values being returned.  We have not defined the semantics of MIN and
MAX when they are passed all nulls.  SQL returns null in this case, which is probably the
right answer.

# In many of the initial functions, a call is made to getTupleFactory.  We should look into
making this static to avoid the cost of getting the tuple factory each time (a similar change
in the trunk brought a significant speed up).

In both cases, I don't think these are changes you introduced, you just extended what was
already there.  But while we're in there reworking it, we might as well improve it.

> Allow UDFs to have different implementations based on input types
> -----------------------------------------------------------------
>
>                 Key: PIG-276
>                 URL: https://issues.apache.org/jira/browse/PIG-276
>             Project: Pig
>          Issue Type: Sub-task
>            Reporter: Alan Gates
>            Assignee: Pradeep Kamath
>         Attachments: EvalFunc.patch, EvalFunc_unittestcases.patch
>
>


-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message