Author pitrou
Recipients geertj, giampaolo.rodola, gvanrossum, pitrou, sbt, vstinner, yselivanov
Date 2015-01-02.11:35:12
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1420198513.54.0.571495650162.issue22560@psf.upfronthosting.co.za>
In-reply-to
Content
> Antoine, can you please elaborate the rationale of your patch?

The patch adds SSL support for proactor-based event loops (any event loop supporting plain sockets, actually, so it could also work for libuv etc.).

> Is the "legacy" code only used on Python 3.4 and older? Is ssl.MemoryBIO always present in Python 3.5 and newer?

Yes and yes.

> I would like to see benchmarks of memory BIO vs current code on Linux and Windows.

Do you have such benchmarks?

> Maybe we may only use memory BIO for the proactor event loop?

It sounds better to exercise the same code path under all platforms.

> Did you try your patch on Python 3.3?

No.

> How will this patch help to support STARTTLS?

Guido explained this one :-)
History
Date User Action Args
2015-01-02 11:35:13pitrousetrecipients: + pitrou, gvanrossum, geertj, vstinner, giampaolo.rodola, sbt, yselivanov
2015-01-02 11:35:13pitrousetmessageid: <1420198513.54.0.571495650162.issue22560@psf.upfronthosting.co.za>
2015-01-02 11:35:13pitroulinkissue22560 messages
2015-01-02 11:35:12pitroucreate