On Fri, Jan 22, 2021, at 16:42, Tom Lane wrote: >"Joel Jacobson" <j...@compiler.org> writes: >> I ran this query (on a patched database) to see if there are still any >> catalog tables without primary keys: >> ... >> pg_depend >> pg_shdepend > >Yeah, this is noted in the patch's own regression tests.
Thanks. Looks good. >> Wouldn't it be possible to add primary keys to these two as well? > >Neither of the existing indexes is suitable, not being unique. > >We could imagine adding a unique index across the whole column set, >but that would be an awfully large price to pay for neatnik-ism. >Also, at least for pg_depend (less sure about pg_shdepend), some code >cleanup would be required, because I don't think that we try very >hard to avoid making duplicate dependency entries. On the whole >I feel this'd be counterproductive. > >regards, tom lane I see, and I agree with you. I'm very happy with this patch. I've already found great use of it in a tool I'm working on: https://github.com/truthly/pg-pit/blob/master/FUNCTIONS/add_primary_keys.sql#L23 Many thanks to all of you for all the great work! Can't wait to use this functionality in production. /Joel
