Clint Wylie

Hi @FrankChen021 thanks for having a look, sorry I haven't finished filling out the PR description for #12753 yet, it might have been able to answer some of your questions....

> Looks like it is failing code coverage, weird, no code was added / removed. Tried restarting those jobs, lets see. Ah, the coverage always runs, even when the tests...

Apologies that it has been so slow getting traction on this PR, I have started to look things over and will try to do a full review soon.

>@clintropolis Did you have a chance to review this PR? Is there any blockage? Anything I can help with? Sorry I haven't yet, only blockage has been on my side...

I think the compaction integration tests might be legitimately failing (i restarted them all last night and still not passing). I haven't looked closer yet to see what the issue...

it looks like a handful of integration tests perhaps have legitimate failures, though they all look similar-ish: exceptions that occur from some underlying network failure, so hopefully it isn't something...

> One way I can come up with is to define some macros such as double_latest for different data types at the SQL layer. @gianm @clintropolis Do you have any...

>Hi @clintropolis , Thanks for your suggestion. I'll try to solve it. Depending on how big of a change this is, it might be worth splitting out a separate PR...

>Hi @clintropolis @suneet-s , Could you review this PR at any time you're convenient ? Since this PR is a little large, I think the SQL problem could be separated...

drive by comment re `INFORMATION_SCHEMA`, i think it is a bit of a standard, https://en.wikipedia.org/wiki/Information_schema, so we need to be considerate about how we modify it I think. (I'll try...