"Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov> wrote:
> Florian Pflug <fgp(at)phlo(dot)org> wrote:
>> Hm, OK I see your point now I believe. This is not so much about
>> wanting to put things into BEFORe triggers which doesn't really
>> fit there, but instead avoiding weeks of debugging if someones
>> messes up.
> I would say that is the overriding concern.
I'm going to have to argue with myself -- really, the *biggest*
concern is that the current situation allows data to be quietly
mangled through non-obvious action-at-a-distance. The damage might
not be discovered for a very long time. Debugging the cause of the
mangling, once noticed, is the *next* most significant concern.
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2011-08-02 20:47:18|
|Subject: Transient plans versus the SPI API|
|Previous:||From: Kevin Grittner||Date: 2011-08-02 18:32:10|
|Subject: Re: WIP fix proposal for bug #6123|