ajwerner

The OIDs don't imply the widths. I guess you can make a char(2) type or whatever width you want but if you don't name the width I guess we do...

TFTR! bors r+

bors r-

bors r+

I think the deeper principle here relates to all functions which take `regclass` or `oid` arguments. We today accept those as strings but when we serialize them, we need to...

The action item here is to take some of the `seqexpr` logic and generalize it fully such that we track references in expressions by IDs and serialize them with the...

I suppose this is fine. We already have logic in the legacy schema changer to retry transient errors: https://github.com/cockroachdb/cockroach/blob/d106bd6525e547372cb96c9082a22b19ee966bde/pkg/sql/schema_changer.go#L2627-L2631 It seems that here we're in the importer, so I guess...

I prefer that, if it works, because the user will get visibility, there will be backoff, and we'll eventually succeed.

@dt can you chime in here? I know sql teams own import or whatever, but let's be real, you know the answers and we don't. I do see progress checkpointing...

I'm taking this off the schema board.