pgsql: Remove user_relns() SRF from regression tests.

From: Andres Freund <andres(at)anarazel(dot)de>
To: pgsql-committers(at)postgresql(dot)org
Subject: pgsql: Remove user_relns() SRF from regression tests.
Date: 2016-09-13 02:39:47
Message-ID: E1bjddT-0007Xu-3Y@gemulon.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers

Remove user_relns() SRF from regression tests.

The output of the function changes whenever previous (or, as in this
case, concurrent) tests leave a table in place. That causes unneeded
churn.

This should fix failures due to the tests added bfe16d1a5, like on
lapwing, caused by the tsrf test running concurrently with misc. Those
could also have been addressed by using temp tables, but that test has
annoyed me before.

Discussion: <27626(dot)1473729905(at)sss(dot)pgh(dot)pa(dot)us>

Branch
------
master

Details
-------
http://git.postgresql.org/pg/commitdiff/0dba54f1666ead71c54ce100b39efda67596d297

Modified Files
--------------
src/test/regress/input/create_function_2.source | 9 --
src/test/regress/input/misc.source | 3 -
src/test/regress/output/create_function_2.source | 8 --
src/test/regress/output/misc.source | 138 -----------------------
4 files changed, 158 deletions(-)

Browse pgsql-committers by date

  From Date Subject
Next Message Robert Haas 2016-09-13 13:25:16 pgsql: Have heapam.h include lockdefs.h rather than lock.h.
Previous Message Andres Freund 2016-09-13 02:04:32 Re: pgsql: Address portability issues in bfe16d1a5 test output.