-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/66961/#review202589
-----------------------------------------------------------




3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 82 (patched)
<https://reviews.apache.org/r/66961/#comment284462>

    Close server?



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 86 (patched)
<https://reviews.apache.org/r/66961/#comment284463>

    Close server?



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 92 (patched)
<https://reviews.apache.org/r/66961/#comment284464>

    Close server?



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 97 (patched)
<https://reviews.apache.org/r/66961/#comment284465>

    Close server?



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 106-109 (patched)
<https://reviews.apache.org/r/66961/#comment284467>

    Just make a `WindowsSocketError`, check its `.code`, and return it.



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 117-120 (patched)
<https://reviews.apache.org/r/66961/#comment284466>

    Ditto.



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 123 (patched)
<https://reviews.apache.org/r/66961/#comment284468>

    Should this just get wrapped into something that will `os::close` when it 
leaves scope? There's a lot of manually handling its destruction in this 
function.



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 188-189 (patched)
<https://reviews.apache.org/r/66961/#comment284469>

    You can just use `sandbox` instead of `os::getcwd()`.



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 201 (patched)
<https://reviews.apache.org/r/66961/#comment284470>

    `s/We sending/We are sending`



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 207 (patched)
<https://reviews.apache.org/r/66961/#comment284471>

    // NOTE: A pending operation is represented by `None()`.



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 215-217 (patched)
<https://reviews.apache.org/r/66961/#comment284473>

    I'm starting to change my mind on the "be consistent with Windows 
`TRUE/FALSE` policy"... This would be so much cleaner as `EXPECT_TRUE(...)`.



3rdparty/stout/tests/os/sendfile_tests.cpp
Lines 219-221 (patched)
<https://reviews.apache.org/r/66961/#comment284472>

    These (and the other non-fatal ones) should be `EXPECT` instead of `ASSERT`.


- Andrew Schwartzmeyer


On May 4, 2018, 10:17 a.m., Akash Gupta wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/66961/
> -----------------------------------------------------------
> 
> (Updated May 4, 2018, 10:17 a.m.)
> 
> 
> Review request for mesos, Andrew Schwartzmeyer, Benjamin Mahler, Eric Mumau, 
> John Kordich, Joseph Wu, and Radhika Jandhyala.
> 
> 
> Bugs: MESOS-8681
>     https://issues.apache.org/jira/browse/MESOS-8681
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> The sendfile tests were using `socketpair`, but the rest of the tests
> were crossplatform. By providing a Windows `socketpair` implementations,
> the tests are now ported.
> 
> 
> Diffs
> -----
> 
>   3rdparty/stout/tests/CMakeLists.txt 
> 28674c9873b7bbccb6b990ec16b7e40a5bf4f9ec 
>   3rdparty/stout/tests/os/sendfile_tests.cpp 
> 05966ae067ae3972598da3370eb16fdce5736c21 
> 
> 
> Diff: https://reviews.apache.org/r/66961/diff/1/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Akash Gupta
> 
>

Reply via email to