Greetings,
* Tom Lane (tgl@sss.pgh.pa.us) wrote:
> Michael Banck <michael.banck@credativ.de> writes:
> > Am Freitag, den 19.10.2018, 10:36 -0400 schrieb Stephen Frost:
> >> I just saw this committed and I'm trying to figure out why we are
> >> creating yet-another-list when it comes to deciding what should be
> >> checksum'd and what shouldn't be.
>
> > To be fair, the list in src/backend/replication/basebackup.c was a copy-
> > paste from the one in pg_verify_checksums (or from other parts of the
> > online activation patch).
>
> Seems like a job for a new(?) module in src/common/.
Yeah, that would be nice... Even nicer would be a way for non-core PG
tools to be able to use it (we have basically the same thing in
pgbackrest, unsurprisingly), though I realize that's a much larger step.
Thanks!
Stephen