Good work, so can you give us a clue on the comparison contract violation?
--
Jody Garnett
On Thu, May 15, 2014 at 8:27 PM, Robert Justice <[email protected]>wrote:
> Hey guys..we figured it out. There is definitely a comparison contract
> violation with the JTS libraries using JDK 1.7 due to the
> move to default TimSort. See my previous email which I think is awaiting
> moderation due to its length. However, if you get
> into a situation running this out of a UDF in Hadoop Hive using JDK 1.7,
> the mistake we made was to
>
> set mapred.child.java.opts='-Djava.util.Arrays.useLegacyMergeSort=true
> -Xmx8g ..and so on';
>
> If you put single quotes around the options, the useLegacyMergeSort=true
> does not take effect. Simply remove the single
> quotes and JTS will no longer use TimSort in JDK 1.7 and should have the
> same memory footprint and performance as it did in
> JDK 1.6. Also the comparison contract violation will not occur as
> MergeSort is not as strict.
>
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Jts-topo-suite-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jts-topo-suite-user
