That patch looks fine. I'd use a different name than 'accept_op', e.g. 'accept_future', and I'd like to delete the future again once the accept has completed. I'd also like to see a unit test, and I'd like you to sign the (online) PSF contributor form: https://www.python.org/psf/contrib/contrib-form/
If you can, please use Rietveld (codereview.appspot.com) to iterate on your patch. Thanks! On Sat, Mar 15, 2014 at 2:26 AM, Juraj Ivančić <[email protected]>wrote: > On 15.3.2014. 3:25, Guido van Rossum wrote: > > I didn't say anything about connecting the client? >> > > I misunderstood. > > > Anyway, isn't this the same situation, just a little later in the pipe's >> life? >> >> I think that the clean solution would be to make the PipeServer >> aware of its async accept operation, so that it can be cancelled in >> PipeServer.close() before the pipe is closed. >> >> >> Do you have a specific change in mind? >> > > Sure. The attached patch seems to fix the issue. > > -- --Guido van Rossum (python.org/~guido)
