----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/70053/#review213187 -----------------------------------------------------------
3rdparty/stout/include/stout/os/posix/mktemp.hpp Lines 39-57 (original), 39-57 (patched) <https://reviews.apache.org/r/70053/#comment299051> From what I understand, C++11 already guarantees for us that `_path[_path.size()] == '\0'` and so we shouldn't need this +1 and '\0' logic? ``` std::string _path(path); int_fd fd = ::mkstemp(&_path[0]); ... return _path; ``` - Benjamin Mahler On Feb. 25, 2019, 9:20 p.m., Benjamin Bannier wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/70053/ > ----------------------------------------------------------- > > (Updated Feb. 25, 2019, 9:20 p.m.) > > > Review request for mesos and Benjamin Mahler. > > > Repository: mesos > > > Description > ------- > > This is a non-functional change to avoid manual memory management. By > using a `std::string` instead of an explicitly heap-allocated buffer the > used pattern might even avoid hitting the heap altogether due to the SSO > but this function might be runtime bounded by I/O anyway. > > > Diffs > ----- > > 3rdparty/stout/include/stout/os/posix/mktemp.hpp > 8dab2599f13c3e1dab109423c8a938ec16540aaf > > > Diff: https://reviews.apache.org/r/70053/diff/1/ > > > Testing > ------- > > `make check` > > > Thanks, > > Benjamin Bannier > >
