[ksh93-integration-discuss] Re: diffs file in the source (was ksh93-integration prototype004 2007-01-27 webref+diff)

[Jan Setje-Eilers]

 <11762.1170465777 at athyra>Mike Kupfer writes:
> >>>>> "Danek" == Danek Duvall <danek.duvall at sun.com> writes:
> 
> Danek> On Thu, Feb 01, 2007 at 08:19:19PM -0800, Jan Setje-Eilers wrote:
> 
> >> To be honest, I'm starting to regret not finding the time to get an
> >> exception to the single delta rule and getting the history in that
> >> way.
> 
> Danek> FWIW, I think I would have preferred that.
> 
> I expect that to be a little awkward.  As soon as you get multiple deltas
> (trying to get the local changes right), you then have to do a dance to
> get from 
> 
>     (gate rev + latest upstream + local changes in many deltas)  
> 
> to 
> 
>     (gate rev + latest upstream + local changes in 1 delta)
> 
> I'd feel better about it if we had (better) tool support for that
> operation.

 I don't think anyone was suggesting allowing actually multiple
deltas. The case in which we might have considered talking about an
exception would have involved a create (so no pre-existing gate rev),
in essence introducing one bit of external history from the beginning.

 That said, I really don't see that those would be useful for all that
long after integration; 6 months perhaps. For major putbacks folks
keep around project gates for that sort of time anyway. So this is all
probably a non-issue and we should simply not introduce dead code.

-jan