expected/sequence_1.out obsolete?

Started by Peter Eisentrautabout 11 years ago4 messages
#1Peter Eisentraut
peter_e@gmx.net

expected/sequence_1.out hasn't been updated at least since

commit d90ced8bb22194cbb45f58beb0961251103aeff5
Date: Thu Oct 3 16:17:18 2013 -0400

and nobody appears to have complained.

Can it be removed?

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

#2Tom Lane
tgl@sss.pgh.pa.us
In reply to: Peter Eisentraut (#1)
Re: expected/sequence_1.out obsolete?

Peter Eisentraut <peter_e@gmx.net> writes:

expected/sequence_1.out hasn't been updated at least since
commit d90ced8bb22194cbb45f58beb0961251103aeff5
Date: Thu Oct 3 16:17:18 2013 -0400
and nobody appears to have complained.

Can it be removed?

No, it needs to be fixed. The alternate output from "select * from
foo_seq" will appear occasionally (from memory, if a checkpoint starts
or perhaps completes during the test).

We might have noticed this already if the buildfarm weren't so noisy
lately :-(

An alternative answer would be to remove log_cnt from the set of
columns displayed, but I think that would be taking away part of the
point of that particular test.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

#3Michael Paquier
michael.paquier@gmail.com
In reply to: Peter Eisentraut (#1)
1 attachment(s)
Re: expected/sequence_1.out obsolete?

On Wed, Oct 22, 2014 at 4:46 AM, Peter Eisentraut <peter_e@gmx.net> wrote:

expected/sequence_1.out hasn't been updated at least since

commit d90ced8bb22194cbb45f58beb0961251103aeff5
Date: Thu Oct 3 16:17:18 2013 -0400

and nobody appears to have complained.

Can it be removed?

I don't think that this is a good idea, have a look at bb3f839 explaining
that this alternate output may happen when a checkpoint kicks in. Simple
patch is attached.
--
Michael

Attachments:

20141021_regression_sequence.patchtext/x-patch; charset=US-ASCII; name=20141021_regression_sequence.patchDownload
diff --git a/src/test/regress/expected/sequence_1.out b/src/test/regress/expected/sequence_1.out
index 124967e..e426f64 100644
--- a/src/test/regress/expected/sequence_1.out
+++ b/src/test/regress/expected/sequence_1.out
@@ -91,6 +91,8 @@ SELECT nextval('serialTest2_f6_seq');
 
 -- basic sequence operations using both text and oid references
 CREATE SEQUENCE sequence_test;
+CREATE SEQUENCE IF NOT EXISTS sequence_test;
+NOTICE:  relation "sequence_test" already exists, skipping
 SELECT nextval('sequence_test'::text);
  nextval 
 ---------
@@ -163,6 +165,9 @@ SELECT nextval('sequence_test'::text);
       99
 (1 row)
 
+DISCARD SEQUENCES;
+SELECT currval('sequence_test'::regclass);
+ERROR:  currval of sequence "sequence_test" is not yet defined in this session
 DROP SEQUENCE sequence_test;
 -- renaming sequences
 CREATE SEQUENCE foo_seq;
@@ -341,6 +346,9 @@ SELECT lastval();
       99
 (1 row)
 
+DISCARD SEQUENCES;
+SELECT lastval();
+ERROR:  lastval is not yet defined in this session
 CREATE SEQUENCE seq2;
 SELECT nextval('seq2');
  nextval 
#4Tom Lane
tgl@sss.pgh.pa.us
In reply to: Michael Paquier (#3)
Re: expected/sequence_1.out obsolete?

Michael Paquier <michael.paquier@gmail.com> writes:

I don't think that this is a good idea, have a look at bb3f839 explaining
that this alternate output may happen when a checkpoint kicks in. Simple
patch is attached.

Pushed, thanks. (The 9.4 branch was broken too, but differently :-()

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers