Re: Confused by result of pg_catalog.format_type()
От | Erik Jones |
---|---|
Тема | Re: Confused by result of pg_catalog.format_type() |
Дата | |
Msg-id | 5E8468D4-11AF-4904-AA2B-19387E0514CE@myemma.com обсуждение исходный текст |
Ответ на | Re: Confused by result of pg_catalog.format_type() (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-general |
On Apr 24, 2008, at 5:10 PM, Tom Lane wrote: > Erik Jones <erik@myemma.com> writes: >> Ah, pg_catalog.format_type(prorettype, null) = '"trigger"', thanks. > > It's probably fair to ask what it is you want to accomplish here, > because comparing format_type's output to a constant seems awfully > fragile. Aside from the quotes (which I believe 8.3 won't emit in > this particular case) the output can vary depending on search_path > and perhaps other factors. > > If you're trying to identify trigger functions I'd suggest > > where prorettype = 'pg_catalog.trigger'::pg_catalog.regtype > > as being the most bulletproof formulation, and probably faster too. Ah, thank you for the advice! I'm writing python unittest assertions for testing db state -- extremely useful for testing some custom client db tools I'm writing for partitioning and migrations (both data and schema). I'd known about, and used, the regclass oid type, but had only learned about it from seeing someone else's example. I should've read the chapter on system information function a little more thoroughly :) My assertion methods are about to become a lot shorter... Erik Jones DBA | Emma® erik@myemma.com 800.595.4401 or 615.292.5888 615.292.0777 (fax) Emma helps organizations everywhere communicate & market in style. Visit us online at http://www.myemma.com
В списке pgsql-general по дате отправления: