> On Oct. 2, 2012, 10:31 p.m., John Sirois wrote:
> > src/zookeeper/group.cpp, line 410
> > <https://reviews.apache.org/r/7391/diff/1/?file=172701#file172701line410>
> >
> >     Since we don't know if join will ever be called I'd be good with not 
> > writing the ephemeral test - in cases of watch only this is an overly 
> > restrictive test and in the case of join we just find out a little later 
> > that the join fails.  I realize now that when I added the write test I 
> > assumed mesos usages.  Although rooted in mesos, the src is otherwise not 
> > logically attached.  Basically you are getting burned by the same mistake I 
> > made mixing 2 responsibilities in 1 class.

Done.


- Benjamin


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/7391/#review12118
-----------------------------------------------------------


On Oct. 2, 2012, 6:45 p.m., Benjamin Hindman wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/7391/
> -----------------------------------------------------------
> 
> (Updated Oct. 2, 2012, 6:45 p.m.)
> 
> 
> Review request for mesos and John Sirois.
> 
> 
> Description
> -------
> 
> Refactored recursive ZooKeeper path create code from Group into 
> ZooKeeper::create (which now takes a boolean 'recursive' flag that determines 
> whether or not it should try and create intermediate znodes). Also refactored 
> ZooKeeperMasterDetector to use this new version of create.
> 
> 
> Diffs
> -----
> 
>   src/detector/detector.cpp 0246846 
>   src/tests/zookeeper_tests.cpp 2e04069 
>   src/zookeeper/authentication.hpp eb60467 
>   src/zookeeper/authentication.cpp 45a41ac 
>   src/zookeeper/group.cpp d7660c5 
>   src/zookeeper/zookeeper.hpp da43e81 
>   src/zookeeper/zookeeper.cpp 98193ee 
> 
> Diff: https://reviews.apache.org/r/7391/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Benjamin Hindman
> 
>

Reply via email to