No error is thrown if selectExactlyOne selects 0 rows when used in a lateral

[tetchel]

I've created a repository with a repro and details here: https://github.com/tetchel/zap-lateral-bug

[jawj]

Hmm. I guess the idea of selectExactlyOne is that you use it when you absolutely know that there is one matching row — e.g. because you're joining on a NOT NULL foreign key.

That said, you're right that this is unexpected and inconsistent behaviour, and that qualifies it as a great big bug. I'll see if I can fix it!

[purge]

Ha, I just experienced this! you think you absolutely know there's one matching row but RLS makes that uncertain :)

[tetchel]

yeah, for us we'd removed a NOT NULL from a FK column and did not think to update all the laterals using that column as a join.