Greetings,
* Robert Haas (robertmhaas@gmail.com) wrote:
> Unless somebody can prove convincingly that this argument is wrong and
> that there are no other possible problems either, and memorialize that
> argument in the form of a detailed comment, I think we should reject
> this patch. http://postgr.es/m/CA+Tgmoa7TJA6-MvjWdcb_bouwFKx9apnU+Ok9m94TkTZTYmqjw@mail.gmail.com
> from earlier this morning is good evidence for the proposition that we
> must be careful to document the reasons for the changes we make, even
> seemingly minor ones, if we don't want developers to be guessing in
> ten years whether those changes were actually safe and correct.
Based on Robert's feedback (which, given his comments, I agree with),
I'm going to mark this as rejected. The approach for dealing with this
seems to be what Alvaro was getting at above where we should have a way
for an extension to get control to handle cleaning things up during
DROP EXTENSION, or call the appropriate event triggers before we drop
the functions, or, well, something else, but certainly not by just
making this change.
Thanks!
Stephen