> On Sept. 8, 2014, 10:31 p.m., Matthew Hayes wrote: > > datafu-pig/src/main/java/datafu/pig/sampling/SampleByKey.java, line 117 > > <https://reviews.apache.org/r/25428/diff/1/?file=682466#file682466line117> > > > > We should throw an exception if the seed is null instead. Otherwise > > this UDF will give unexpected behavior.
Thank you for the review Matt! I've completely missed your response from some reason :-S I've used constant here so that I'll always give expected outcome. I'm fine with throwing exception as well though, so I'll update the patch later today. - Jarek ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/25428/#review52644 ----------------------------------------------------------- On Sept. 8, 2014, 8:57 a.m., Jarek Cecho wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/25428/ > ----------------------------------------------------------- > > (Updated Sept. 8, 2014, 8:57 a.m.) > > > Review request for DataFu. > > > Bugs: DATAFU-68 > https://issues.apache.org/jira/browse/DATAFU-68 > > > Repository: datafu > > > Description > ------- > > Attaching patch that: > > * Fixes trailing whitespaces in the file > * Adds validation that signature is not null before calling hashCode > * Added check to intToRandomDouble to ensure that we won't use the null > > > Diffs > ----- > > datafu-pig/src/main/java/datafu/pig/sampling/SampleByKey.java d94a038 > > Diff: https://reviews.apache.org/r/25428/diff/ > > > Testing > ------- > > I didn't add new test case as I wasn't able to easily force Pig to call the > method setUDFContextSignature with null parameter on the unit test level. > Existing unit tests are passing. > > > Thanks, > > Jarek Cecho > >
