* Peter Eisentraut (peter.eisentraut@2ndquadrant.com) wrote:
> On 5/6/16 3:11 PM, Stephen Frost wrote:
> >These are just new tests..?
>
> This is a matter of degree, but I think there is a difference
> between new test cases and a whole new test suite.
To be clear, I've been calling it a 'test suite' because there ended up
being over 1000 tests added, but it's all using the exinsting
infrastructure of the TAP testing system, just as we have for other
commands (pg_ctl, initdb, etc).
> >I assumed that would be welcome during post
> >feature-freeze, and certainly no one raised any concerns about adding
> >these tests during the discussion prior to my commiting them.
>
> I didn't see that discussion and I still can't find it.
The new tests were discussed on the thread related to the changes to
pg_dump. The first patch, which included just 300 or so tests, was
here:
http://www.postgresql.org/message-id/20160425043909.GW10850@tamriel.snowman.net
There was a bit of back-and-forth on that thread and new patches were
posted there, as well as indication that I was planning to push them a
couple days before I did.
Honestly, over the next couple of months between feature-freeze and
release, I'd like to add even more tests, and not just to pg_dump but
also to other commands that don't have very good testing today (psql, in
particular, but pg_dumpall needs more also, and there's more to do with
pg_dump too).
I certainly understand the concern raised about making the buildfarm go
red right before a release, though I believe that's now been dealt with.
When it comes to packaging, if adding tests using the existing test
infrastructure (the TAP system) causes a problem there, then I think we
need to address that issue rather than not add new tests. Packagers
also always have the option to not enable the tap tests, if there really
is an issue there which can't be addressed.
Thanks!
Stephen