|From:||Michael Paquier <michael(at)paquier(dot)xyz>|
|To:||Chapman Flack <chap(at)anastigmatix(dot)net>|
|Cc:||Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Craig Ringer <craig(at)2ndquadrant(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>|
|Subject:||Re: configure's checks for --enable-tap-tests are insufficient|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
On Mon, Mar 19, 2018 at 10:54:59PM -0400, Chapman Flack wrote:
> Ehh, I'm with Craig. The *perl project* identifies which packages (and
> versions) are part of perl core in each perl version ... not Red Hat.
> Red Hat's practice seems super reckless given that there's an actual
> module (itself part of core since 5.8.9) that lets you check what
> modules you can safely assume are there. I'm sure I'm not the only one
> to have written code relying on it....
At the same time I don't think that there is much we can do in trying to
reduce the split in those distributions. Even if it were to be changed
in the future, that won't affect existing deployments, past OS base
images and maintenance releases. So in order to simplify the life of
people running the regression test suites, I would be on board with Tom
in adding more checks at configure level as it simplifies the life of
anybody running the TAP tests on those distributions.
|Next Message||Etsuro Fujita||2018-03-20 03:32:51||Re: inserts into partitioned table may cause crash|
|Previous Message||Tom Lane||2018-03-20 03:07:13||Re: pg_get_functiondef forgets about most GUC_LIST_INPUT GUCs|