Re: Application structuring with threads & network I/O - suggestions?



On Wednesday 22 February 2006 11:02, Christian Neumair wrote:
Am Freitag, den 17.02.2006, 22:04 -0500 schrieb Gorshkov:
I'm developing an small network application as a test case before I go on
to my main app - I want to make my mistakes on something small &
manageable. [...]

Unfortunately, we can't do much about it without having seen code. My
suspicion is that you just proved that C is error prone.


ummmmm ...... actually, it's not. Without getting into a major relegious war, 
NO language is error prone - *programmers* are error prone.

A bad programmer will write bad code. A good programmer will write good code. 
A great programmer will write code that makes you all wet & squishy :-)

Anyway - that's besides the point. I've solved the problem, and it seems to 
have been an interaction between the gnet library I was trying to use and 
gtk.

While searching the web for problems similar to the ones I was having, I came 
across a bug report/discussion thread somewhere that described almost exactly 
the problems I was having. I'm sorry I don't remember the details, or 
remembered to save the links. Tor  was part of the discussion, and the big 
problem was that he was unable to reproduce the bug on your systems, so there 
wasn't a lot you could have done about it.

So I bit the bullet, and went under the hood of windows. It was much less 
painfull that I thought it would be - I didn't realise that the windows 
networking stacks were based on the BSD stacks, so it didn't take me long to 
replace the gnet calls with the native windows API.

Inside of 2 hours I had a fully working application. Given that the only code 
changes *were* in my network routines, I'm pretty sure that there is 
something weird happening between the 2.8 series of Tor's port and the gnet 
libary.

Tor: I'm sorry I don't still have that code - it might have helped you 
reproduce something, but I wasn't thinking about that at the time.



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]