On Thu, Aug 28, 2008 at 8:15 PM, Christophe
<xof@thebuild.com> wrote:
On Aug 28, 2008, at 6:10 PM, Matthew Dennis wrote:
I'm not sure I follow. Couldn't you track which statements were prepared that called a function and either reprepare (just like reindex, recheck, etc) or in the case of dropping a function, refuse to drop it because something depends on it?
Yes, sorry about that - it should have been obvious the execute you were talking about from the context. In any case, you don't track dependencies there for the same reason you don't track what SQL my application sends. The plpgsql execute statement, as I understand it, means "take this string and execute like a client sent it to you". It's explicit in the definition/description/documentation of the function that you aren't executing a "static" thing that would have dependencies you would track. However, when you used execute, you should get no different repsonse than what a client would get under the same circumstances.
