apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From r..@covalent.net
Subject Re: cvs commit: apr/locks/unix locks.c
Date Sat, 02 Dec 2000 01:36:48 GMT

> > Then please just leave it, and I'll implement the whole thing myself.  The
> > other option is for you to implement it, and I'll go in immediately and
> > change it.
> Why is it that two people ask for the simple one, don't want to do the
> harder one (and don't want to see that complexity in there), yet you will
> just go and do it anyways?

Because APR was designed to be small pieces that don't necessarily know
about each other.  There is a minimum amount of code that creates
cross-dependancies between different sections of APR.  It is not that much
more complex, and it helps to maintain the abstraction.  It is important
to me to ensure that each piece of APR can build individually without
losing much of it's functionality.  By putting all of this into one big
function in misc/unix/start.c, you remove that ability, because we rely on
misc to report on file_io.

Now, go back and read Jeff's reply:

> > That is for damn sure :)^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H
> > I'll start out with all the logic right in cfg.c and as other folks
> > have time/inclination they can consider other implementations.

I have the time/inclination.

That is why I am willing to implement something that I believe is the
correct implementation.  Personally, I think it is a bit less complex than
a single function that is basically one large:


I also know that I will be taking small pieces of APR and putting them in
my own projects.  I will most likely want to have the reporting features
that we are talking about.

Ryan Bloom                        	rbb@apache.org
406 29th St.
San Francisco, CA 94131

View raw message