2007/12/19, Daniel Stenberg wrote:

> Ok, here's my simple take at a fix for systems that doesn't have
> socklen_t in their system headers.

Thanks to your patch I finally got the clue of how it could be done.
Based on your patch here's another incomplete one which takes care of
potential redefinition issues and that should be good if the configure
script defines socklen_t or not.

Should we go this route ?

A patch and a file since it seems that today I'm unable to properly diff it.

-- 
-=[Yang]=-