Re: Small issues with CREATE TABLE COMPRESSION

On Thu, May 06, 2021 at 09:33:53PM +0530, Dilip Kumar wrote:
> On Thu, May 6, 2021 at 5:42 PM Michael Paquier <michael(at)paquier(dot)xyz> wrote:
> >
> > On Thu, May 06, 2021 at 10:45:53AM +0530, Dilip Kumar wrote:
> > > Thanks, Robert and Michael for your input. I will try to understand
> > > how it is done in the example shared by you and come up with the test
> > > once I get time. I assume this is not something urgent.
> >
> > Thanks. FWIW, I'd rather see this gap closed asap, as features should
> > come with proper tests IMO.
>
> I have done this please find the attached patch.

No objections to take the approach to mark the lz4-related test with a
special flag and skip them. I have three comments:
- It would be good to document this new flag. See the comment block
on top of %dump_test_schema_runs.
- There should be a test for --no-toast-compression. You can add a
new command in %pgdump_runs, then unmatch the expected output with the
option.
- I would add one test case with COMPRESSION pglz somewhere to check
after the case of ALTER TABLE COMPRESSION commands not generated as
this depends on default_toast_compression. A second test I'd add is a
materialized view with a column switched to use lz4 as compression
method with an extra ALTER command in create_sql.

> I don't have much idea about the MSVC script, but I will try to see
> some other parameters and fix this.

Thanks! I can dive into that if that's an issue. Let's make things
compatible with what upstream provides, meaning that we should have
some documentation pointing to the location of their deliverables,
equally to what we do for the Perl and OpenSSL dependencies for
example.
--
Michael