New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix vector plots #1560

Merged
merged 9 commits into from Sep 24, 2015

Conversation

Projects
None yet
5 participants
@aashish24
Contributor

aashish24 commented Sep 19, 2015

WIP: I have observed tests failing on my machine but the difference seems small. I just wanted to see how they run on buildbot machine.

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 19, 2015

Ref: #653

@aashish24 aashish24 force-pushed the fix_vector_plots branch from 4e31235 to a50fe9d Sep 21, 2015

@doutriaux1

This comment has been minimized.

Member

doutriaux1 commented Sep 21, 2015

@aashish24 still lots of failures here. What changed?

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 22, 2015

@doutriaux1 thats puzzling me. On my machine, I am getting only vector tests failure. Although I am using the patched VTK

@aashish24 aashish24 force-pushed the fix_vector_plots branch from f3bdc79 to 99ba107 Sep 22, 2015

@aashish24 aashish24 force-pushed the master branch from bfe2642 to 3283d6c Sep 22, 2015

@aashish24 aashish24 force-pushed the fix_vector_plots branch 5 times, most recently from 3f97702 to 4589d40 Sep 22, 2015

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 22, 2015

@doutriaux1 still the same. For some reason, I am not seeing these failing tests on my machine.

@@ -27,14 +30,45 @@ def plot(self, data1, data2, tmpl, grid, transform):
data1 = self._context().trimData2D(data1)
data2 = self._context().trimData2D(data2)
scale = 1.0
lat = data1.getLatitude()[:]

This comment has been minimized.

@sankhesh

sankhesh Sep 23, 2015

Contributor

data1 could be NoneType

vcs_no_xtra_elts fails because of that. (https://open.cdash.org/testDetails.php?test=375905204&build=4024522)

@aashish24 aashish24 force-pushed the fix_vector_plots branch from 5040d2f to bd9ae1f Sep 24, 2015

@aashish24 aashish24 force-pushed the fix_vector_plots branch from bd9ae1f to df7ef7b Sep 24, 2015

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

@doutriaux1 this branch should be good to now. The problem (failing tests) was with VTK bug. I will fix the doWrap method in another PR.

@aashish24 aashish24 force-pushed the fix_vector_plots branch from a0aecd6 to 5fed2d3 Sep 24, 2015

@aashish24 aashish24 force-pushed the fix_vector_plots branch from 5fed2d3 to 1ff5fbd Sep 24, 2015

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

@doutriaux1

This comment has been minimized.

Member

doutriaux1 commented Sep 24, 2015

@aashish24 waiting for the bots to be done to confirm merge. Thanks for pushing this in.

@doutriaux1

This comment has been minimized.

Member

doutriaux1 commented Sep 24, 2015

@aashish24 4 failures on oceanonly, but I think it is because master was merged in by bots.

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

Right. there was an issue with the test-data. The only real test failing is test_no_extra_elements

@aashish24 aashish24 force-pushed the fix_vector_plots branch 2 times, most recently from 8693236 to 6e386fa Sep 24, 2015

@aashish24 aashish24 force-pushed the fix_vector_plots branch from 6e386fa to 7ac099b Sep 24, 2015

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

@doutriaux1 most of the tests are passing now. The multiple format one and label background one is failing for some other reason.

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

@doutriaux1 this branch is looking good now. Not sure why the label and the click test is failing.

https://open.cdash.org/testDetails.php?test=376700167&build=4028520

@durack1

This comment has been minimized.

Member

durack1 commented Sep 24, 2015

@aashish24 the reason is due to problems with the test image, so compare the baseline:
test_vcs_click_info
With the test output:
test
You'll see that there is bleeding of color outside the box, and the y-axis labels in particular are offset, look at 90N.. There are also problems with the x-axis labels and their placement too..

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

@doutriaux1 could it be caused by my changes? thanks @durack1

@durack1

This comment has been minimized.

Member

durack1 commented Sep 24, 2015

@aashish24 I've been seeing this issue in a number of PRs #1540, #1560, #1557 - and it only seems to be a problem on garant - so an OS-specific or machine-specific issue?

@durack1

This comment has been minimized.

Member

durack1 commented Sep 24, 2015

@aashish24 @doutriaux1 I'm wondering if garant is just not syncing correctly with the updated source and as it's behind we're getting these failures - looks like the case with #1564..

@jbeezley

This comment has been minimized.

Contributor

jbeezley commented Sep 24, 2015

It doesn't merge in master like the others. Is that the issue?

@durack1

This comment has been minimized.

Member

durack1 commented Sep 24, 2015

@jbeezley it seems that in particular with #1564 FFmpeg issues are only occurring on garant, whereas everything else passes.. So like it fails to pull across all the PR changes or something? Dunno?

@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

thanks @durack1 for the information. In that case, I think this branch looks good to me. Can someone review it @durack1 @sankhesh @jbeezley @chaosphere2112

@@ -0,0 +1,24 @@
import vcs, cdms2, numpy, os, sys

This comment has been minimized.

@sankhesh

sankhesh Sep 24, 2015

Contributor

This file doesn't comply with PEP8 style.

This comment has been minimized.

@aashish24

aashish24 Sep 24, 2015

Contributor

@sankhesh but I don't see flake8 test failing. what I am missing?

This comment has been minimized.

@jbeezley

jbeezley Sep 24, 2015

Contributor

We probably don't check the test scripts.

This comment has been minimized.

@aashish24

aashish24 Sep 24, 2015

Contributor

Ah I see. Thanks @jbeezley. I will push a fix to follow the PEP8 stype in my next branch as I need to take care of doWrap as well.

doutriaux1 added a commit that referenced this pull request Sep 24, 2015

@doutriaux1 doutriaux1 merged commit 7241b33 into master Sep 24, 2015

6 of 9 checks passed

cont-int/LLNL/Darwin-Mac 10.10.3 running 'ctest -j4 -D Experimental' (Thu Sep 24 08:57:41 2015)
Details
cont-int/LLNL/Linux-annie Ubuntu 15.04 (FULL/MESA) running 'ctest -j15 -D Experimental' (Thu Sep 24 07:45:11 2015)
Details
continuous-integration/kitware-buildbot/uvcdat-garant-linux-release/ Build done.
Details
cont-int/LLNL/Darwin-Mac1 10.10.5 (NOGUI) running 'rm -rf /Users/doutriaux1/bot/build' (Thu Sep 24 07:00:13 2015)
Details
cont-int/LLNL/Darwin-Mac2 10.10.5 (FULL) running 'rm -rf /Users/doutriaux1/bot/build' (Thu Sep 24 07:00:14 2015)
Details
cont-int/LLNL/Linux-crunchy RH6 (FULL) running 'ctest -j12 -D Experimental' (Thu Sep 24 07:25:50 2015)
Details
cont-int/LLNL/Linux-oceanonly RH6 (MESA/NOGUI) running 'ctest -j12 -D Experimental' (Thu Sep 24 07:20:32 2015)
Details
continuous-integration/travis-ci/pr The Travis CI build passed
Details
continuous-integration/travis-ci/push The Travis CI build passed
Details
@aashish24

This comment has been minimized.

Contributor

aashish24 commented Sep 24, 2015

thanks @doutriaux1

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment