Tom Lane <tgl@sss.pgh.pa.us> writes:
> Applied with some further hacking.
Thanks!
> Hmm, that leads me to wonder exactly how extensively the regression
> tests test event triggers, because it sure looked to me like there
> were multiple bugs left in this version.
The first versions of the event triggers patch series used to include a
large regression test files with some level of covering for each and
every supported command. It was used this way because there was some
command specific code at the time.
It was apparently not testing the right things and was judged to be too
much tests for no clear benefit, thus thrown away. Then with the never
stopping refactoring of the patch series, no specific regression tests
ever found their way back in.
It's not clear to me what tests to add exactly, will find some time to
read the current code in more details and figure out what's easy enough
to cover and not covered already.
>> Finally, I've been surprised to find out that those cases are only
>> triggering for "ddl_event_start" (and not "ddl_event_end"), I think
>> that's a bug we should be fixing:
>
> Agreed, I fixed it.
Thanks!
Regards,
--
Dimitri Fontaine
http://2ndQuadrant.fr PostgreSQL : Expertise, Formation et Support