----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/45015/#review126159 -----------------------------------------------------------
3rdparty/libprocess/3rdparty/stout/tests/os/filesystem_tests.cpp (lines 144 - 166) <https://reviews.apache.org/r/45015/#comment189059> I thought we concluded that actions such as `os::mkdir` and `os::touch` should be `ASSERT_*`, and the things that are being actually being tested here, i.e. `os::exists` should be `EXPECT_*`? - Michael Park On March 24, 2016, 7:58 a.m., Alex Clemmer wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/45015/ > ----------------------------------------------------------- > > (Updated March 24, 2016, 7:58 a.m.) > > > Review request for mesos, Alex Naparu, Daniel Pravat, Artem Harutyunyan, > Joris Van Remoortere, Michael Park, M Lawindi, and Yi Sun. > > > Repository: mesos > > > Description > ------- > > Currently on Windows, `os::exists` will return true if a component of a > path does not exist. For example if you have `a/fancy/path`, and you ask > `os::exists("a/fake/path")`, the result currently reports `true`. In > other words, the Windows code path only checks for the error that a file > does not exist, and ignores the error that says the path is not valid. > > This commit will fix this, and also add a test that will verify we don't > regress. > > > Diffs > ----- > > 3rdparty/libprocess/3rdparty/stout/include/stout/os/windows/exists.hpp > 9211851e4562e04045276421b359c3c78cdae7f1 > 3rdparty/libprocess/3rdparty/stout/tests/os/filesystem_tests.cpp > 4c30189bb8261ccfc699da0f31b8b1fd3e9b3c83 > > Diff: https://reviews.apache.org/r/45015/diff/ > > > Testing > ------- > > > Thanks, > > Alex Clemmer > >