[Asterisk-Dev] Client development...

[Tilghman Lesher]

On Thursday 25 September 2003 19:44, Boudewijn van Ingen wrote:
> At 04:45 PM 25-09-2003 -0700, TC wrote:
>  >>I'm an experienced developer and have looked at the existing ports
>  >> of the original code. But for several reasons,
>  >
>  >the iaxclient.sf.net has nice working iax lib that is portable..
>  >is there some fundamental issue that could not be worked out on
>  > that project
>  >
>  >>I have decided that it would be
>  >>best for my needs to make a completely different design of the
>  >>iax-libraries, this time in c++, and hopefully remaining somewhat
>  >> portable.
>  >
>  >just seems a shame to start a new attempt rather than building on
>  > the effort
>
> It would seem that way to any inexperienced software developer.

Actually, no.  It would seem to be more difficult, because you're
limiting yourself to coding only in C++.  This is perhaps a reason not
to code in C++.

> Still, what I am attempting to do, is not an attempt to throw all the
> efforts of 'those who went before me' away. What I am attempting to
> do, is capturing all that happened before in better streamlined code.
> (And only for *clients* of the asterisk system, mind you, I have no
> further ambitions right now.) The reason I want to do that is mainly
> because I saw so many problems in the way that "libaix" has been
> ported to Windows so far.
>
> Furthermore, I would like to have a version of it that would be more
> easily extended by other programmers.

Embrace, extend, extinguish.  Isn't that the Microsoft way?

> To me, it seems that a properly designed C++ version of "libaix"
> could help in both those areas. But simply porting the C version into
> C++ is a hard job, without further information about the
> iax-protocol. Even without considering Windows ways to handle I/O (I
> have already concentrated that part into proper C++ classes that can
> easily be replaced)...

Ah, it's the old "see those plus signs after the C?  That means it's
better" argument.

-Tilghman