> On Dec. 27, 2017, 3:19 a.m., Bill Farner wrote: > > I have mixed feelings about this, but am leaning towards -1 on declaring > > this a fix. Skimming the scheduler parameters, i can't say for sure > > whether all `List` parameters are invalid when empty. To me, this suggests > > a code fix is the only answer. > > > > We don't need to wait for my jcommander PR to fix this up; we can use a > > custom value for `@Parameter(splitter=CustomSplitter.class)` for `List` > > parameters. In this case, `CustomSplitter.class` can cargo cult > > `CommaParameterSplitter.java` from my jcommander PR.
My main motivation for this patch was that we don't have to go through another 0.19.1 release cycle to get the packages out of the door. The bug only applies if one actually adds an argument via the command line, but not if the default is used. That is also why most people that build their own packages did not run into the problem, yet. I can follow your reasoning though. Closing. - Stephan ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/64824/#review194522 ----------------------------------------------------------- On Dec. 23, 2017, 3:45 p.m., Stephan Erb wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/64824/ > ----------------------------------------------------------- > > (Updated Dec. 23, 2017, 3:45 p.m.) > > > Review request for Aurora and Bill Farner. > > > Bugs: AURORA-1962 > https://issues.apache.org/jira/browse/AURORA-1962 > > > Repository: aurora-packaging > > > Description > ------- > > Ensure we can package 0.19.0 without having to wait for > https://github.com/cbeust/jcommander/pull/422 > > > Diffs > ----- > > specs/debian/aurora-scheduler.startup.sh > 3f97a73249cd6601465be8a0031067715290cb9c > specs/debian/aurora-scheduler.upstart > 3c3206887070f72061095b342bfbf8b535c82cff > > > Diff: https://reviews.apache.org/r/64824/diff/1/ > > > Testing > ------- > > > Thanks, > > Stephan Erb > >
