msg245986 - (view) |
Author: Guido van Rossum (gvanrossum) *  |
Date: 2015-06-30 12:54 |
See https://github.com/python/asyncio/issues/251. I'm on vacation and may or may not find the time to actually fix this (it needs a unittest written). |
|
|
msg246248 - (view) |
Author: Adam Bartoš (Drekin) * |
Date: 2015-07-04 11:00 |
I've also been hit by this issue, see https://mail.python.org/pipermail/python-list/2015-July/693496.html and the following thread. I've spent some time trying to find where the problem is, and just after I found out it's about eof_received() returning None, I could found this issue. :-) |
|
|
msg246728 - (view) |
Author: Guido van Rossum (gvanrossum) *  |
Date: 2015-07-14 11:10 |
I've added the return True from eof_received() to the asyncio repo (https://github.com/python/asyncio/commit/ce3ad816a2ef9456b4b1c26b99dfc85ea1236811), but it still needs a unittest and merging into CPython 3.4 and up. |
|
|
msg247305 - (view) |
Author: Roundup Robot (python-dev)  |
Date: 2015-07-25 00:47 |
New changeset fc1d40a706e7 by Victor Stinner in branch '3.4': asyncio: sync with github https://hg.python.org/cpython/rev/fc1d40a706e7 |
|
|
msg247306 - (view) |
Author: STINNER Victor (vstinner) *  |
Date: 2015-07-25 00:52 |
> but it still needs a unittest and merging into CPython 3.4 and up. I did this part. By the way, running unit tests now logs two warnings on SSL tests, because returning True has no effect on SSL. We may just remove the warning at runtime and ensure that it's well documented instead. What do you think? |
|
|
msg253782 - (view) |
Author: Aymeric Augustin (aymeric.augustin) * |
Date: 2015-10-31 11:13 |
This change appears to have caused a non-obvious regression in `websockets`: https://github.com/aaugustin/websockets/issues/76 Perhaps I was relying on an implementation detail but that's annoying nonetheless. |
|
|
msg253791 - (view) |
Author: R. David Murray (r.david.murray) *  |
Date: 2015-10-31 14:35 |
Well, asyncio was still provisional, so annoyances like this were bound to happen occasionally. Victor, is there a reason this issue is still open? Is it because of the warnings question? Maybe that should be a new issue? |
|
|
msg253793 - (view) |
Author: Guido van Rossum (gvanrossum) *  |
Date: 2015-10-31 14:58 |
Well, the warning points to a deeper issue -- the fix doesn't work for TLS connections. Also, the upstream issue ( https://github.com/python/asyncio/issues/251) is still open because there's no test for this behavior. |
|
|
msg253795 - (view) |
Author: R. David Murray (r.david.murray) *  |
Date: 2015-10-31 15:02 |
Ah, I thought Victor was saying he'd done the unit test; I guess he just meant he'd done the merge. |
|
|
msg253809 - (view) |
Author: STINNER Victor (vstinner) *  |
Date: 2015-10-31 21:29 |
"Victor, is there a reason this issue is still open? Is it because of the warnings question? Maybe that should be a new issue?" I just merged from Github to CPython. I don't know if something remains to do in this issue. |
|
|
msg308817 - (view) |
Author: Andrew Svetlov (asvetlov) *  |
Date: 2017-12-20 21:11 |
Done two years ago. |
|
|