Re: [opendap-tech] [libdap 3.18.0] DMRTest: 177 178 179 180 failed

[Aleksander Vines]

Hi,

By libdap 3.18.1, I assume you mean the master branch on github? I can not find any release for this version yet.

It still seems to fail on the same tests.

I'm building with libxml2 2.9.4.

You can see the entire build output at https://travis-ci.org/nansencenter/libdap4-feedstock

Thanks,

Aleksander

From: James Gallagher <jgall...@opendap.org>
To: Aleksander Vines <aleksand...@nersc.no>
Cc: James Gallagher <jgall...@opendap.org>, Tech OPeNDAP <openda...@opendap.org>
Sent: 2016-08-22 22:30
Subject: Re: [opendap-tech] [libdap 3.18.0] DMRTest: 177 178 179 180 failed

On Jul 7, 2016, at 04:52, Aleksander Vines <aleksand...@nersc.no> wrote:

forgot to attach the files.

Thanks. Sorry for the delay - travel, vacation and travel...

Can you try libdap 3.18.1? This problem may be fixed since OSX and Debian 12 (travis) show the code passing all tests:

176: trans test_array_1.xml scale(x,0x8fffffffffffffff)  test_array_1.xml.7.func_base ok
177: trans test_array_5.xml scale(a,0.001)  test_array_5.xml.1.func_base ok
178: trans test_array_5.xml scale(b,0.001)  test_array_5.xml.2.func_base ok
179: trans test_array_5.xml scale(c,0.001)  test_array_5.xml.3.func_base ok
180: trans test_array_5.xml scale(d,0.001)  test_array_5.xml.4.func_base ok

The problem is that in the constrained DMR that’s printed by the dmr-test test driver the order of the <Dimension …> elements is reversed (given what’s in the baseline). It’s not really an error, because order of those elements does not matter, but the test does not know it’s OK. Arrgh. And, I don’t know why the order is different. 

If you see this problem in 3.18.1, I’ll look into it. This could be an issue with libxml2, so if you find the version you’re using, that would help. I am using 2.9.2 on OSX and on Debian 12 I’m using whatever libxml2-dev will provide.

Thanks,

James

---

Aleksander Vines

Programmer

IT | Scientific Data Management

Nansen Environmental and Remote Sensing Center

(+47) 465 05 891 | aleksand...@nersc.no

From: Aleksander Vines <aleksand...@nersc.no>
To: <openda...@opendap.org>
Sent: 2016-07-07 12:50
Subject: [libdap 3.18.0] DMRTest: 177 178 179 180 failed

Hi,

I've encountered 4 fails during make test, and I can not figure out what the failure really is (from DMRTest.log).

The build output is found here: https://travis-ci.org/nansencenter/libdap4-feedstock/builds/143006280

Attached is the DMRTest.log and config.log.

I'm trying to build older versions to see when the issue occured, so far version 3.17.3 reports exactly the same. In 3.17.2 one of the tests succeeds (so there are 3 failures). 

3.17.2. this one succeeds: 167: trans test_array_5.xml scale(b,0.001)  test_array_5.xml.2.func_base ok

---

Aleksander Vines

Programmer

IT | Scientific Data Management

Nansen Environmental and Remote Sensing Center

(+47) 465 05 891 | aleksand...@nersc.no

---

Nansen Environmental and Remote Sensing Center
Address: Thormøhlens gate 47, NO-5006 Bergen, NORWAY
Website: www.nersc.no, Email: po...@nersc.no
Work: +47 55 20 58 00, Fax: +47 55 20 58 01

--
You received this message because you are subscribed to the Google Groups "OPeNDAP Tech" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opendap-tech...@opendap.org.
To post to this group, send email to openda...@opendap.org.
Visit this group at https://groups.google.com/a/opendap.org/group/opendap-tech/.
For more options, visit https://groups.google.com/a/opendap.org/d/optout.
<DMRTest.log><config.log>

--

James Gallagher
jgall...@opendap.org

---

Nansen Environmental and Remote Sensing Center
Address: Thormøhlens gate 47, NO-5006 Bergen, NORWAY
Website: www.nersc.no, Email: po...@nersc.no
Work: +47 55 20 58 00, Fax: +47 55 20 58 01

[James Gallagher]

On Aug 23, 2016, at 07:47, Aleksander Vines <aleksand...@nersc.no> wrote:

Hi,

By libdap 3.18.1, I assume you mean the master branch on github? I can not find any release for this version yet.

Yes, I do mean the master branch. 

It still seems to fail on the same tests.

OK, I’ll look into this a time permits.

Thanks,

James

--

James Gallagher
jgall...@opendap.org