Re: Loss of some parts of the function definition - Mailing list pgsql-hackers

From Jim Nasby
Subject Re: Loss of some parts of the function definition
Date
Msg-id 5543E9FF.4060501@BlueTreble.com
Whole thread Raw
In response to Loss of some parts of the function definition  (Sergey Grinko <sergey.grinko@gmail.com>)
Responses Re: Loss of some parts of the function definition  (Sergey Grinko <sergey.grinko@gmail.com>)
List pgsql-hackers
On 4/30/15 6:44 AM, Sergey Grinko wrote:
> Now create a script in the application of its function parameters and
> return values can be declared using %TYPE.
> However, when you save the script is stored inside the server only what
> is considered his body. Thus, we obtain:
...

We actually mung things a lot worse when it comes to views, so I'm 
curious why you're only worried about the problems with stored functions?

FWIW, I think the best 'solution' to this right now is to actually keep 
your original definitions as files in your VCS and use something like 
sqitch for deployment. Taken to it's logical extreme, that means that 
the only thing you ever 'patch' is an actual table (via ALTER TABLE), or 
indexes. Everything else essentially gets treated like regular code.

That's still not terribly satisfying since unlike other forms of 
software you now have all that definition both in your VCS and the 
database itself, but ISTM that's a much bigger problem than the small 
amount of info we lose from stored functions...
-- 
Jim Nasby, Data Architect, Blue Treble Consulting
Data in Trouble? Get it in Treble! http://BlueTreble.com



pgsql-hackers by date:

Previous
From: Jim Nasby
Date:
Subject: Re: CTE optimization fence on the todo list?
Next
From: Peter Geoghegan
Date:
Subject: Re: INSERT ... ON CONFLICT UPDATE/IGNORE 4.0