> On 22 Feb 2023, at 21:55, Tom Lane wrote:
>
> Daniel Gustafsson writes:
>>> On 22 Feb 2023, at 21:33, Andres Freund wrote:
>>> On 2023-02-22 15:10:11 +0100, Daniel Gustafsson wrote:
Rebased patch to handle breakage of v2 due to bd8d453e9b.
>
>>> I think we probably should just apply thi
Daniel Gustafsson writes:
>> On 22 Feb 2023, at 21:33, Andres Freund wrote:
>> On 2023-02-22 15:10:11 +0100, Daniel Gustafsson wrote:
>>> Rebased patch to handle breakage of v2 due to bd8d453e9b.
>> I think we probably should just apply this? The current behaviour doesn't
>> seem
>> right, and
> On 22 Feb 2023, at 21:33, Andres Freund wrote:
> On 2023-02-22 15:10:11 +0100, Daniel Gustafsson wrote:
>> Rebased patch to handle breakage of v2 due to bd8d453e9b.
>
> I think we probably should just apply this? The current behaviour doesn't seem
> right, and I don't see a downside of the new
Hi,
On 2023-02-22 15:10:11 +0100, Daniel Gustafsson wrote:
> > On 26 Nov 2022, at 22:46, Daniel Gustafsson wrote:
>
> > I've moved the statuses[i] check before the differ check, such that there
> > is a
> > separate block for this not mixed up with the differs check and printing.
>
> Rebased p
> On 26 Nov 2022, at 22:46, Daniel Gustafsson wrote:
> I've moved the statuses[i] check before the differ check, such that there is a
> separate block for this not mixed up with the differs check and printing.
Rebased patch to handle breakage of v2 due to bd8d453e9b.
--
Daniel Gustafsson
v3-
> On 26 Nov 2022, at 21:55, Andres Freund wrote:
> On 2022-11-26 21:11:39 +0100, Daniel Gustafsson wrote:
>> The attached makes child failures an error condition for the test as a belts
>> and suspenders type check. Thoughts?
>
> I wonder if it's the right thing to treat a failed psql that's the
Hi,
On 2022-11-26 21:11:39 +0100, Daniel Gustafsson wrote:
> In the thread about TAP format out in pg_regress, Andres pointed out [0] that
> we allow a test to pass even if the test child process failed. While its
> probably pretty rare to have a test pass if the process failed, this brings a
> r
In the thread about TAP format out in pg_regress, Andres pointed out [0] that
we allow a test to pass even if the test child process failed. While its
probably pretty rare to have a test pass if the process failed, this brings a
risk for false positives (and it seems questionable that any regress
