Tom Lane wrote:
> Hmmm ... I'm less than excited about documenting agreed-to-be-broken
> behavior in exquisite detail. Doing so will just encourage people to
> rely on it. It's particularly unfriendly to document it in detail
> without pointing out that it's likely to change.
>
> My inclination is to forget about these doc changes and spend our time
> on fixing the behavior instead.
>
The open question for this one then is the right way for exceptions to
act. Marti, since you've looked at this a bit already, any thoughts on
what that should look like? You initially commented that what happens
was surprising, but I'm not clear on what you expected to happen instead.
--
Greg Smith 2ndQuadrant US greg@2ndQuadrant.com Baltimore, MD
PostgreSQL Training, Services and Support www.2ndQuadrant.us