> > Magnus Hagander wrote:
> >> Do we really want to run cosmetic cleanups on a stable branch?
>
> > Agreed, it is not a great idea, but if we don't, then 8.1.X and CVS
> > HEAD will not match indenting, and patches generated by 8.1.X users
> > will not apply cleanly to CVS HEAD. And if we don't run it
> at all, we
> > then will have CVS HEAD with columns > 80 and incorrect
> typedef indentations.
>
> I agree with Bruce here: better to keep 8.1 and HEAD matching
> as best we can. I've already had problems with back-patching
> because the comment indentation in 8.0 and 8.1 is so
> completely different --- manually redoing a patch because
> patch can't figure it out is no fun and a likely source of
> errors. Having to do it an extra time for 8.1 vs HEAD would
> increase the pain and risk that much more.
I didn't consider the patch-conflict issue. With that in mind, yeah, it
seems reasonable to do it.
> One of the reasons I wanted Bruce to post the proposed diff
> was so that we could eyeball-verify that it's only hitting
> comments. I think it's worth a little more trouble to check
> the results given that we plan to run it against 8.1.
Sounds like a good idae.
//Magnus
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
