Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com> wrote:
> (sorry for repeatedly replying to self. I'll go for a coffee after
> this...)
That's so nice of you to try to make me feel better for the serious
brain fade I suffered yesterday. ;-)
> On 08.06.2011 14:18, Heikki Linnakangas wrote:
>> Committed after adjusting that comment. I did a lot of other
>> cosmetic changes too, please double-check that I didn't screw up
>> anything.
On a first read-through, all your changes look like improvements to
me. I'll do some testing and give it a closer read.
> Also, it would be nice to have some regression tests for this. I
> don't think any of the existing tests exercise these new functions
> (except for the fast-exit, which isn't very interesting).
I'll see about posting regression tests for this, as well as looking
into the questions in your earlier posts -- particularly about the
heap truncation in vacuum. I'm pretty sure that vacuum can't clean
up anything where we're still holding predicate locks because of
visibility rules, but I'll take another look to be sure. I hope to
do all that today, but I don't want to push to the point where I'm
making dumb mistakes again.
-Kevin