> On May 23, 2018, 12:18 a.m., James Peach wrote: > > src/linux/fs.cpp > > Lines 662 (patched) > > <https://reviews.apache.org/r/67175/diff/1/?file=2024600#file2024600line662> > > > > This seems like a logical extension to `os::touch`? > > > > If we can emulate the `os::stat` API and use enum constants to make > > this more obvious at the call site: > > ``` > > os::touch(path, os::Touch::RECURSIVE, os::Touch::FILE); > > ``` > > > > cc @jieyu > > Jason Lai wrote: > Makes sense. I've dropped the function in this patch and will create > another patch in favor of your suggestion.
For `os::Touch::DIRECTORY`, I actually feel it doesn't bring much value to have `os::touch` to support both recursive and non-recursive cases (a direct call to `os::touch` would simply do the work). The semantics of touching with an extra call to `os::utime` upon existing paths also feels unnecessary and/or could introduce breaking risks whn the path is readonly and the action is supposed to be no-op. - Jason ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/67175/#review203620 ----------------------------------------------------------- On Sept. 15, 2018, 12:36 a.m., Jason Lai wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/67175/ > ----------------------------------------------------------- > > (Updated Sept. 15, 2018, 12:36 a.m.) > > > Review request for mesos, Eric Chung, Gilbert Song, Jie Yu, James Peach, and > Zhitao Li. > > > Bugs: MESOS-8257 > https://issues.apache.org/jira/browse/MESOS-8257 > > > Repository: mesos > > > Description > ------- > > Added utility function for marking slave mounts in the mount table. > > > Diffs > ----- > > src/linux/fs.hpp 502f85c4a32d8658bdd701975dd5ac3d802d308e > src/linux/fs.cpp 9055ef42edd1fb90e1026d1d603a9ba902cfc1fd > > > Diff: https://reviews.apache.org/r/67175/diff/2/ > > > Testing > ------- > > > Thanks, > > Jason Lai > >
