peek blocks after read on windows about mio HOT 7 CLOSED

Can you reproduce the error with Mio types (not Tokio)?

from mio.

If you look at the other issue someone did indeed reproduce it only with mio

from mio.

If you look at the other issue someone did indeed reproduce it only with mio

Then why do you include an example using Tokio in an issue for Mio?

from mio.

I am sorry this is the first time I try to contribute to the project, the line between the projects were still a bit blurry to me. It was just a simpler example for me to use. I figured I could try to help since the original issue didn't get much traction.

from mio.

Since we are not sure who's fault it is, I will close the issue for now. If someone has the same problem feel free to re-open with a proper mio example.

from mio.

I left a comment in the Tokio thread, but if there is a problem with Mio then I still like to know, so feel free to reopen when/if it's clear it a problem with Mio.

from mio.

I commented back, unsure if I have enough knowledge to create a reproduction. I didn't see any test around peek in mio though, so at the very least adding a test to confirm that mio is working as intended would be good. That would need to check that:

• There is a first readiness event on first data
• Test does a peek -> Receives data
• Test does a read -> Receives data
• Test does a peek -> Receives WouldBlock
• New data is in the socket
• There is no readiness event sent by mio

If and only if this is the case for all platform would I consider mio "bug free". I highly suspect that the last check will fail on unix platforms, but maybe not.

from mio.