Martin Richard

Results 9 issues of Martin Richard

The old name is kept for backward-compatibility but will be removed for 1.0. Fixes #102

Running tests with PYTHONASYNCIODEBUG=1 generates a lot of warning messages. We should check if they're generated by asynctest or by the test, and fix them in both cases.

Documentation is pushed on readthedocs, but the latest commit on master is always the one used for the documentation. It means that the documentation may describe unreleased features, and it's...

documentation

The name CoroutineMock is confusing (see #100, #101, etc) as it's the mock of a coroutine function, not a coroutine itself. CoroutineMock should be renamed to CoroutineFunctionMock. CoroutineMock will be...

Currently, the behavior when calling a CoroutineMock with a coroutine as side_effect is undefined. It raises an exception with Python 3.6 and seems to run the coroutine with Python 3.5...

enhancement

Now that CoroutineMock relies on the loop for some features, we should let the user specify which loop shall be used. CoroutineMock(loop=None) # default, in this case, get_event_loop() is called...

Dear all, Following the bug #414, here is a patch which makes the fork + exec operation non blocking on Unix. The patch is probably not ready to land, but...

CLA signed

Hi, Subprocess creation is handled by `loop.subprocess_exec()` and `loop.subprocess_shell()`. Both uses `subprocess.Popen()`, but the later performs at least one blocking read on a pipe (on unix at least), used to...

We can rename ``settings.py`` to ``settings.py.dist`` an remove the former from the git repository to ease development.