Re: Re: Outstanding patches - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Re: Outstanding patches
Date
Msg-id 21086.989424423@sss.pgh.pa.us
Whole thread Raw
In response to Re: Re: Outstanding patches  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-hackers
Bruce Momjian <pgman@candle.pha.pa.us> writes:
>> No, I don't think that's very related; that's a simple matter of
>> implementing an ALTER FUNCTION command.  The other thing will require
>> figuring out how to do dependency tracking.

> Got it.  Let me ask, if they change the column type, would they use
> ALTER FUNCTION to then update to match the new column type.  As I
> understand it, the problem is that this does not happen automatically,
> right?

My vision of ALTER FUNCTION is that it would let you change the function
body, and perhaps also the function language and attributes (isCachable,
isStrict).  It would NOT allow you to change the function's parameter
types or return type, because that potentially breaks things that depend
on the function.  To do that, you should have to create a new function.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Re: Outstanding patches
Next
From: Bruce Momjian
Date:
Subject: Re: Re: New Linux xfs/reiser file systems