Re: Assertion for logically decoding multi inserts into the catalog

> On 31 Jul 2019, at 19:20, Heikki Linnakangas <hlinnaka@iki.fi> wrote:

> This patch makes the assertion more strict than it was before. I don't see how it could possibly make a regression failure go away. On the contrary. So, huh?

Yeah, this is clearly fat-fingered, the intent is to only run the Assert in
case XLH_INSERT_CONTAINS_NEW_TUPLE is set in xlrec->flags, as it only applies
under that condition.  The attached is tested in both in the multi-insert patch
and on HEAD, but I wish I could figure out a better way to express this Assert.

cheers ./daniel