Re: confusing / inefficient "need_transcoding" handling in copy

Hi,

In <Z1ukEe2d7ml6-oaZ(at)paquier(dot)xyz>
"Re: confusing / inefficient "need_transcoding" handling in copy" on Fri, 13 Dec 2024 12:03:45 +0900,
Michael Paquier <michael(at)paquier(dot)xyz> wrote:

>> OK. I've added valid cases too by using LATIN1 as you
>> suggested.
>
> I may have missed something but v3 does not use that for a valid
> conversion?

Oh, sorry... I attached wrong patch...
I attach the v4 patch that includes this case.

>> Oh! I didn't know the "XXX_1.out" feature.
>
> You have missed the inclusion of an alternate output, which should be
> something like that to bypass the test rather than failing:
> --- /dev/null
> +++ b/src/test/regress/expected/copyencoding_1.out
> @@ -0,0 +1,7 @@
> +--
> +-- Test cases for COPY encoding
> +--
> +-- skip test if not UTF8 server encoding
> +SELECT getdatabaseencoding() <> 'UTF8' AS skip_test \gset
> +\if :skip_test
> +\quit
>
> I guess that you have the file, forgot a `git add`.

I did "git add" but I attached a wrong file...

The v4 patch includes this too.

Thanks,
--
kou