David E. Wheeler
David E. Wheeler
From my testing it seems to only be an issue for 5.22, 5.24, and 5.40.
Thanks to @perennialmind's comment for a reasonable work around. Would be nice to be able to specify mappings of some kind, though, to avoid finicky configs like [this](/theory/justatheory/blob/master/Makefile#L13).
Still an issue, I've updated my blog publish script from the broken link above to [this script](/theory/justatheory/blob/main/bin/publish). Sure wish I could specify mappings explicitly and call it once!
If you trigger the event from within pgTAP, the only way I know of to use it to `LISTEN` for such events is to use [dblink](https://www.postgresql.org/docs/current/dblink.html) within the test to...
Odd. Your example works perfectly for me on Postgres 17.5.
I agree 1 is worth trying. I suspect it means updating the use of `pg_proc.prokind` in a `WHERE` clause.
Oh! ISTR that I've been thinking about `has_functions()` as applying to all functions, and then one can do separate calls to `is_procedure()`, `is_aggregate()`, `is_window()` to determine each one specifically. But...
> I’m a bit on the fence about that. There’s a problem with this approach -- it’s five times slower than using `SELECT toString(...) FROM ...` and getting the same...
I spent some time trying to use `toString()` in the queries built by pg_clickhouse to convert to strings on the back end. I got it working nicely for Postgres `NUMERIC`...
> A more serious issue, however, is that this PR breaks shared-library builds Does `-fPIC` help?