httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexei Kosut <>
Subject Re: 1.3 versus 1.3.1
Date Wed, 10 Sep 1997 22:38:27 GMT
On Wed, 10 Sep 1997, Paul Sutton wrote:

> Does this mean that you don't think the patches to *enable* OS-specific
> abstractions (i.e. in os/*/os.{c,h})  should be committed for 1.3? If
> committed, it ould enable the few os abstractions required for 1.3 to be
> implemented in a consistent manner. I am worried if we don't have the
> ability to add os specific code now, when 1.3 becomes a binary realese and
> the floodgates are opened to every NT user to try Apache, we *will* get a
> lot of NT bugs. They'll all be fixed by patching the code with #ifdef
> WIN32 sections and things will get unmanageable. If the patches are in we
> can, were required, abstract Win32 code to fix the bugs.

No, it doesn't meant that i don't think they should be committed. It just
means that I don't think OS abstraction should be a design goal of 1.3 -
it never was, and it is a goal for 2.0. If it's a useful means to make
certain things work better, in a way that doesn't break other things,
then sure, but I'd perfer not to go too far down that road this late in
the 1.3 game, is all.

-- Alexei Kosut <>

View raw message