Re: shadow variables - pg15 edition

On Thu, 25 Aug 2022 at 02:00, Justin Pryzby <pryzby(at)telsasoft(dot)com> wrote:
>
> On Wed, Aug 24, 2022 at 10:47:31PM +1200, David Rowley wrote:
> > I was hoping we'd already caught all of the #1s in 421892a19, but I
> > caught a few of those in some of your other patches. One you'd done
> > another way and some you'd done the rescope but just put it in the
> > wrong patch. The others had not been done yet. I just pushed
> > f959bf9a5 to fix those ones.
>
> This fixed pg_get_statisticsobj_worker() but not pg_get_indexdef_worker() nor
> pg_get_partkeydef_worker().

The latter two can't be fixed in the same way as
pg_get_statisticsobj_worker(), which is why I left them alone. We can
deal with those when getting onto the next category of warnings, which
I believe should be the "Repurpose" category. If you look at the
shadow_analysis spreadsheet then you can see how I've categorised
each. I'm not pretending those are all 100% accurate. Various cases
the choice of category was subjective. My aim here is to fix as many
of the warnings as possible in the safest way possible for the
particular warning. This is why pg_get_statisticsobj_worker() wasn't
fixed in the same pass as pg_get_indexdef_worker() and
pg_get_partkeydef_worker().

David