Greg Smith <g...@2ndquadrant.com> writes:
> Yeah, you're right that is the proper way to handle this. As the
> problem isn't that serious once you're aware of it, I don't see a large
> amount of motivation to work on that now though, if that's what it will
> take to fix. And Tomas seemed satisfied with a workaround too. I just
> added a TODO item pointing to your suggested implementation so it's more
> likely people will stumble onto the relevant trivia here on their own,
> and maybe some day a patch will get written to implement that idea too.
Well, if we're leaving TODO crumbs, what I had in mind was:
1. Use try_relation_open() not relation_open() to start with.
2. On failure return, do something like
if (relation_recently_dead(relid))
... return NULL ...
else
... throw error ...
relation_recently_dead() would probably be a few dozen lines of code,
but it could be shared among all places where we want to do something
like this.
regards, tom lane
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
