[
https://issues.apache.org/jira/browse/PIG-2886?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13444694#comment-13444694
]
Dmitriy V. Ryaboy commented on PIG-2886:
----------------------------------------
Hi Ted,
Great to see clouderians contributing to Pig again! :)
Couple of notes:
minTimeRange, maxTimeRange -- maybe better names would be minTimestamp and
maxTimestamp ?
That's the signature for HBase's scanTimeRange.
Also, please fix up documentation -- minTimestamp in scan.setTimeRange is
*inclusive* (so, not strictly greater then). maxTimestamp is, indeed, exclusive
-- the range is [min, max)
space between } and "else" around maxTimeRange handling.
HBase scan also provides setTimestamp(). Might as well throw that in?
Does your client care about # of returned versions? That's a much tricker
change..
> Add Scan TimeRange to HBaseStorage
> -----------------------------------
>
> Key: PIG-2886
> URL: https://issues.apache.org/jira/browse/PIG-2886
> Project: Pig
> Issue Type: Bug
> Reporter: Ted Malaska
> Priority: Minor
> Attachments: PIG-2886-0.patch, PIG-2886-1.patch
>
>
> I have a client that wants to use pig. They are using MR now. They can't
> use PIG right now because they only want to fetch the last day's worth of
> data in HBase. A filter with time range would require reading all the HStore
> files. If we hold major compaction until after the fetch and use Scan Time
> Range we only need to read very little in compression.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
