> On Jan. 7, 2016, 11:21 p.m., Neil Conway wrote: > > 3rdparty/libprocess/README.md, line 323 > > <https://reviews.apache.org/r/42030/diff/1/?file=1186438#file1186438line323> > > > > Do we really have to look at all the code paths within `doHttpGet`? By > > the time `getBlob()` returns to the caller (i.e., by the time anyone might > > satisfy the future in question), `doHttpGet` must have been called and > > returned already.
Correct me if I'm wrong, but the execution context of the callbacks registered on this future will be determined by the code path followed in `doHttpGet`. So if we want to confirm that execution remains within this process' context, we need to walk through those paths and confirm that they stay within the process. - Greg ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/42030/#review113353 ----------------------------------------------------------- On Jan. 8, 2016, 4:55 p.m., Greg Mann wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/42030/ > ----------------------------------------------------------- > > (Updated Jan. 8, 2016, 4:55 p.m.) > > > Review request for mesos, Ben Mahler, Jojy Varghese, Neil Conway, and Timothy > Chen. > > > Bugs: MESOS-4207 > https://issues.apache.org/jira/browse/MESOS-4207 > > > Repository: mesos > > > Description > ------- > > Added example of a `defer` bug to libprocess README. > > > Diffs > ----- > > 3rdparty/libprocess/README.md 6a47f6772bb7a74af368ed970af0f1c154a070e3 > > Diff: https://reviews.apache.org/r/42030/diff/ > > > Testing > ------- > > Viewed the markdown in a github branch: > https://github.com/mesosphere/mesos/tree/defer_example/3rdparty/libprocess > > Note that in addition to the inclusion of this example, further changes to > the `defer` documentation preceding the example are currently being > considered: https://reviews.apache.org/r/41618/ > > > Thanks, > > Greg Mann > >
