Re: How to keep format of views source code as entered?

2021-01-13 Thread Paul Förster
Hi Laurenz, > On 14. Jan, 2021, at 04:59, Laurenz Albe wrote: > > If PostgreSQL were to store the original text, either that text would become > wrong, or you would have to forbid renaming of anything that is referenced > by a view. this is why views, procedures, functions and packages can beco

Re: How to keep format of views source code as entered?

2021-01-13 Thread David G. Johnston
On Wed, Jan 13, 2021 at 6:40 PM Adam Brusselback wrote: > It has been a major annoyance for views with complex subqueries or where > clauses, the PG representation is absolutely unreadable. > The path to a solution here is to write a schema-to-text system that presents the derived output in a m

Re: How to keep format of views source code as entered?

2021-01-13 Thread Laurenz Albe
On Wed, 2021-01-13 at 20:39 -0500, Adam Brusselback wrote: > > Admittedly, the system probably should be made to save the text, should > > someone wish to write such a patch. > > It has been a major annoyance for views with complex subqueries or where > clauses, the PG representation is absolut

Re: How to keep format of views source code as entered?

2021-01-13 Thread Adam Brusselback
> Admittedly, the system probably should be made to save the text, should someone wish to write such a patch. Just wanted to throw $0.02 behind this idea if anyone does want to take it up later. Using a source control system is better obviously. But even if you use source control it is still incre

Docker Image with Python support ?

2021-01-13 Thread Markur Sens
Is there any available Postgres image that has been compiled with the --with-python flag? I could use and modify (comment out this) https://github.com/docker-library/postgres/blob/03e769531fff4c97cb755e4a608b24935c27/13/alpine/Dockerfile but looks like too much for a simple flag.

Re: Define hash partition for certain column values

2021-01-13 Thread Голубева Яна
Sorry, I think I 've described my case not precisely enough. "Randomly" is not pure random in my case. My solution is planning to be used on different servers with different DBs. The initial data of the base table depends on the DB. But I know that the key value of the new rows is increasing. Not m

Re: System (and DB) upgrade problem.

2021-01-13 Thread Laurenz Albe
On Wed, 2021-01-13 at 07:41 +0100, W.P. wrote: > I am upgrading Fedora 24 to (now) 26, PostgreSQL stopped starting (as > expected), the message from systemctl was to do "postgresql-setup > --upgrade". > > Did installed the tool, loaunched. > > But it fails with (attached logs) messages, I am mi