> On Oct. 23, 2014, 12:46 p.m., Vinod Kone wrote:
> > Why are some definitions in .hpp and some in .cpp? Why not all in .cpp?
> > 
> > Also, it's not clear to me how this split would help in unit testing? 
> > AFAICT, all these visitors take Master or Framework or Slave which needs 
> > bringing up a cluster.

They could be all in cpp, but some are so simple that being inlineable seemed 
beneficial.

This is a first step. The Master/Slave/Frameworks passed in could be mock/stub 
versions that would support lightweight testing. The change also has a benefit 
in reducing the amount of code in the master source file, which helps with 
compile time and readability.


- Dominic


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/20423/#review58071
-----------------------------------------------------------


On Oct. 23, 2014, 10:51 a.m., Dominic Hamon wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/20423/
> -----------------------------------------------------------
> 
> (Updated Oct. 23, 2014, 10:51 a.m.)
> 
> 
> Review request for mesos and Benjamin Hindman.
> 
> 
> Bugs: MESOS-1064
>     https://issues.apache.org/jira/browse/MESOS-1064
> 
> 
> Repository: mesos-git
> 
> 
> Description
> -------
> 
> This is the first step toward being able to write independent unit tests for 
> the validation visitors.
> 
> It also uses Owned to make visitor cleanup simpler (non-existent).
> 
> 
> Diffs
> -----
> 
>   src/Makefile.am 2617f77b757cb7414889520c88b1bc203dedef09 
>   src/master/master.cpp 95589b8f25a3e496eabc0cf84319c883c1ed7aec 
>   src/master/offervisitor.hpp PRE-CREATION 
>   src/master/offervisitor.cpp PRE-CREATION 
>   src/master/taskinfovisitor.hpp PRE-CREATION 
>   src/master/taskinfovisitor.cpp PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/20423/diff/
> 
> 
> Testing
> -------
> 
> make check.
> ran Java test framework.
> 
> 
> Thanks,
> 
> Dominic Hamon
> 
>

Reply via email to