Steve,
I presume the differences in errors I'm seeing on winnie are because of the
notice issue.
pg 9.3 'z_crash' => 0,
'z_fail' => 1,
'z_pass' => 64
pg 9.2
'z_crash' => 0,
'z_fail' => 4,
'z_pass' => 61
pg 9.1
'z_crash' => 0,
'z_fail' => 1,
'z_pass' => 64
I haven't looked closely at them to see. I thought I had notice checking
disabled though so perhaps something different since I do recall
Before they were all passing.
-----Original Message-----
From: postgis-devel-bounces@lists.osgeo.org
[mailto:postgis-devel-bounces@lists.osgeo.org] On Behalf Of Paul Ramsey
Sent: Thursday, September 05, 2013 6:58 PM
To: PostGIS Development Discussion
Subject: Re: [postgis-devel] [OT] postgresql raise notice is giving me a
major headache
Not really, you can see that we had to write our own regression harness
(run_test) as soon as we supported multiple versions of PgSQL.
P
--
Paul Ramsey
http://cleverelephant.ca
http://postgis.net
On Thursday, September 5, 2013 at 3:51 PM, Stephen Woodbridge wrote:
Hi all,
We have a bunch of functions in pgrouting that output a lot of
informational NOTICE messages to the user.In our automated tests, we collect the command output and compare it to
a reference file. The headache is the no two versions of postgresql
output the same format for these messages. So a lot of our test fail
with false positive errors. We are testing on pg versions 9.0, 9.1 and
9.2 and I guess we should be adding 9.3 soon also.I guess the obvious answer is to not compare the notice messages. But I
don't like that answer.Any better thoughts on this?
-Steve
_______________________________________________
postgis-devel mailing list
postgis-devel@lists.osgeo.org (mailto:postgis-devel@lists.osgeo.org)
http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel
_______________________________________________
postgis-devel mailing list
postgis-devel@lists.osgeo.org
http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel