Re: [HACKERS] LZTEXT for rule plan stings
От | Don Baccus |
---|---|
Тема | Re: [HACKERS] LZTEXT for rule plan stings |
Дата | |
Msg-id | 3.0.1.32.20000225132940.01098760@mail.pacifier.com обсуждение исходный текст |
Ответ на | LZTEXT for rule plan stings (wieck@debis.com (Jan Wieck)) |
Ответы |
Re: [HACKERS] LZTEXT for rule plan stings
|
Список | pgsql-hackers |
At 09:36 PM 2/25/00 +0100, Jan Wieck wrote: > But it requires an initdb and we're in BETA. So I better ask > if someone complains. Well...here's an example of a view that worked in 6.5, with an 8KB block size, that fails in 7.0 unless I build with a 16KB block size: create view ec_products_displayable as select * from ec_products where active_p='t'; Impressively large, eh? :) I was kinda grossed out when Postgres choked on it, to be honest. ec_products in this case has quite a few columns... You know, I've investigated further and the rule string itself is no where near 8KB. More like 1KB. So there is more to the story than just the string itself. Still, it should help because most of my failing views were just a bit over 8KB. One still fails with a 16KB block size, though! Fortunately it's not currently used in the web tool kit. Anyway, it seems to me that we need SOME solution to this problem. It is going to be hard to convince users that views like the one above are really too complex for Postgres to handle. I still like the idea of "text" being implemented under the hood as lzText for a quick 7.1 release if that idea works out ... - Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert Serviceand other goodies at http://donb.photo.net.
В списке pgsql-hackers по дате отправления: