|From:||Dag Lem <dag(at)nimrod(dot)no>|
|To:||Paul Ramsey <pramsey(at)cleverelephant(dot)ca>|
|Cc:||PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>|
|Subject:||Re: daitch_mokotoff module|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
Dag Lem <dag(at)nimrod(dot)no> writes:
> Tomas Vondra <tomas(dot)vondra(at)enterprisedb(dot)com> writes:
>> On 2/7/23 18:08, Paul Ramsey wrote:
>>>> On Feb 7, 2023, at 6:47 AM, Dag Lem <dag(at)nimrod(dot)no> wrote:
>>>> I just went by to check the status of the patch, and I noticed that
>>>> you've added yourself as reviewer earlier - great!
>>>> Please tell me if there is anything I can do to help bring this across
>>>> the finish line.
>>> Honestly, I had set it to Ready for Committer, but then I went to
>>> run regression one more time and my regression blew up. I found I
>>> couldn't enable the UTF tests without things failing. And I don't
>>> blame you! I think my installation is probably out-of-alignment in
>>> some way, but I didn't want to flip the Ready flag without having
>>> run everything through to completion, so I flipped it back. Also,
>>> are the UTF tests enabled by default? It wasn't clear to me that
>>> they were?
>> The utf8 tests are enabled depending on the encoding returned by
>> getdatabaseencoding(). Systems with other encodings will simply use the
>> alternate .out file. And it works perfectly fine for me.
>> IMHO it's ready for committer.
> Yes, the UTF-8 tests follow the current best practice as has been
> explained to me earlier. The following patch exemplifies this:
Can you please have a look at this again?
|Next Message||Amit Kapila||2023-03-06 11:14:59||Re: [PATCH] Use indexes on the subscriber when REPLICA IDENTITY is full on the publisher|
|Previous Message||Jim Jones||2023-03-06 10:50:54||Re: [PATCH] Add CANONICAL option to xmlserialize|