asyncore: suggested patch

[Robert Amesz]

Jonathan Feinberg wrote:

> rcameszREMOVETHIS at dds.removethistoo.nl (Robert Amesz) writes:
> 
>> If Jonathan Freiberg wants to make his connect-patch work on more 
>> platforms, notably Windows, he'll have to put in a bit more effort. 
> 
> Hmm.  If Robert Azesm

Congratulations. I never saw anyone garble my last name quite as badly
as that. ;-)


> would be more specific about what part of my patch didn't work, and
> in which way it failed, I would be grateful! As for
> Windows-compatibility, I developed and tested the patch in question
> on a Windows 2K box. 

Initially, I was simply concerned about the differences in error 
handling between Windows sockets and Unix sockets[*], but taking a more 
careful look at your patch *and* at asyncore.py I'm sorry to have to 
say that your patch really totally goes against the asyncore design 
principle of having a single poll-loop drive the system.

The added select() in your connection function will *block* the system. 
This means that during a long wait for a connection to be established 
(or time out) asyncore totally grinds to a halt. As you intend to make 
a link checker, which is mostly waiting for connections and not sending 
or receiving a lot of data, this is hardly the sort of behaviour you 
want.

If the only thing you want to do is abort the connection attempt after 
a certain amount of time it would be better just to simply check if the 
connection is actually established at or after a certain point in time. 
That should be straightforward enough, if you clone the loop() function 
in asyncore and put the timeout-check in the 'while map:' loop.

(It might be better still if asyncore would be extended to handle 
timeouts by calling two more callback-functions, say 
handle_connect_timeout() and handle_io_timeout().)

OTOH, it might be simpler using multi-threading and the timeoutsocket 
module.



Robert Amesz
-- 
[*] And I've just noticed some *really* strange bit of behaviour. I'll 
describe it in another thread.