Mail Archives: cygwin-developers/2001/03/23/23:43:13
----- Original Message -----
From: "Christopher Faylor" <cgf AT redhat DOT com>
To: <cygwin-developers AT cygwin DOT com>
Sent: Saturday, March 24, 2001 2:25 PM
Subject: Re: setup revisit
> On Sat, Mar 24, 2001 at 02:19:32PM +1100, Robert Collins wrote:
> >
> >Oh. That wouldn't have affected what I was suggesting (the current
> >cygwin1.dll would be used unless an update was present - in which
case
> >it comes back to the current setup's behaviour.).
>
> And if the current cygwin1.dll is B20? Do you want to rely on that?
> Or if it is a buggy snapshot?
"unless an update was present". Then it upgrades cygwin1.dll first - and
if it can't either gives the current cygwin1.dll upgrade failure "can't
open", or [insert desired message/action].
> >> >True. I was hoping to put my nose down and hack all weekend, but
it
> >> >looks like rpm is *not* going to play nice so I'll give it a
slightly
> >> >lower priority for now.
> >>
> >> Don't let me discourage you though, Robert. If you can get this
done,
> >> it will be pretty cool. Doesn't RPM have a library like interface,
> >too?
> >>
> >
> >I'm lookinginto that. The RPM issue I'm fighting with currently is
> >porting sleepycat's db to raw win32 as a static library. (It doesn't
> >build with gcc unless you build it for cygwin :[).
> >
> >Once that's out of the way I can actually try to build rpm natively.
>
> The last I remember, rpm needed chroot() to function. That's why I
> asked Corinna to implement this in Cygwin.
>
Thanks... Sounds like I'll be ripping code from there for use...
Rob
- Raw text -