Message103418
> If everyone agrees on "error: [Errno 0] Error" being a legitimate alias
> for a "connection closed event" condition then I'd say the test server
> looks good, even if I think that expecting a ssl.SSLError derived
> exception would have made more sense, but I understand this is a
> OpenSSL inherited behavior.
Well, it's quite an unfortunate error display, so perhaps we could detect it and turn into something more meaningful instead.
However, I think it should remain a socket error or an OS error, because it's really the underlying socket which is closed, not the SSL wrapper.
> Anyway, I'd like to improve the SSL test server in future as I'm sure > it hides some other problems but I can do that once pyftpdlib's SSL
> server code is more mature and after 2.7 is released.
Shouldn't it become part of asyncore at some point? It's both useful (well, for people using asyncore anyway :-)) and non-trivial. |
|
Date |
User |
Action |
Args |
2010-04-17 18:11:29 | pitrou | set | recipients:
+ pitrou, jcea, janssen, vstinner, giampaolo.rodola, Arfrever, flox, dlmiles |
2010-04-17 18:11:28 | pitrou | set | messageid: <1271527888.73.0.806568088456.issue8108@psf.upfronthosting.co.za> |
2010-04-17 18:11:27 | pitrou | link | issue8108 messages |
2010-04-17 18:11:26 | pitrou | create | |
|