> On April 8, 2014, 6:36 p.m., Ian Downes wrote:
> > 3rdparty/libprocess/3rdparty/stout/include/stout/os.hpp, line 555
> > <https://reviews.apache.org/r/20127/diff/1/?file=553118#file553118line555>
> >
> >     Can we return Try<int> for descriptive (hard coded) error messages and 
> > remain async signal safe? If not, perhaps document this so someone else 
> > doesn't add it later :-)

Try involves "new" -> involves "malloc" -> async signal unsafe.

Will document this.


> On April 8, 2014, 6:36 p.m., Ian Downes wrote:
> > 3rdparty/libprocess/3rdparty/stout/include/stout/os.hpp, line 574
> > <https://reviews.apache.org/r/20127/diff/1/?file=553118#file553118line574>
> >
> >     standard system() just returns the exit status of the shell, whatever 
> > that may be.

Yeah, but the users of os::system needs to check WIFEXITED(status) and 
WEXITSTATUS(status) then. I prefer not having users worry about that:) What do 
you think?


> On April 8, 2014, 6:36 p.m., Ian Downes wrote:
> > 3rdparty/libprocess/3rdparty/stout/tests/os_tests.cpp, line 136
> > <https://reviews.apache.org/r/20127/diff/1/?file=553119#file553119line136>
> >
> >     what about testing the failure path in the child?
> >     
> >     EXPECT_EQ(-1, os::system("invalid.command.that.should.not.be.found"));

Done.


- Jie


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/20127/#review39808
-----------------------------------------------------------


On April 8, 2014, 6:16 p.m., Jie Yu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/20127/
> -----------------------------------------------------------
> 
> (Updated April 8, 2014, 6:16 p.m.)
> 
> 
> Review request for mesos, Ben Mahler and Ian Downes.
> 
> 
> Repository: mesos-git
> 
> 
> Description
> -------
> 
> See summary.
> 
> 
> Diffs
> -----
> 
>   3rdparty/libprocess/3rdparty/stout/include/stout/os.hpp db5ae69 
>   3rdparty/libprocess/3rdparty/stout/tests/os_tests.cpp d61db30 
> 
> Diff: https://reviews.apache.org/r/20127/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Jie Yu
> 
>

Reply via email to