Skip to content

Conversation

@pp-mo
Copy link
Member

@pp-mo pp-mo commented Mar 31, 2020

Tidied and pulled in the code that I wrote for the recent demo, mainly supporting the mesh subset operations based on gridded/UGRID.

I included everything from there, and added at least one simple test for each function.

This needs a new image test-iris-imagehash and a new testfile in iris-test-data,
so it won't pass tests until both of these PRs are in :

WIP
still todo: fix iris-test-data SHA in Travis, when SciTools/iris-test-data#57 is done.

@pp-mo
Copy link
Member Author

pp-mo commented Mar 31, 2020

I don't understand why no tests are running.
@trexfeathers @bjlittle @lbdreyer do we know why this is ?

Tests are still running on my fork : https://travis-ci.org/github/pp-mo/iris/builds/669350910
(failing because iris-test-data + test-iris-imagehash need updating)

Tests are also not running in test-iris-imagehash, either ...

@lbdreyer
Copy link
Member

I think they are running (see here for the results of the travis run for this PR), but for some reason travis isn't sending the results back to github?

@pp-mo
Copy link
Member Author

pp-mo commented Apr 1, 2020

@lbdreyer I think they are running ... travis isn't sending the results back to github?

Thanks @lbdreyer
That is somewhat comforting, but still pretty odd.
I'd guess that GitHub would be at fault, since it is at least triggering them (if I understand how it works, which I don't)
I'll look into it ..

@pp-mo pp-mo mentioned this pull request Apr 1, 2020
@pp-mo
Copy link
Member Author

pp-mo commented Apr 1, 2020

Update: the testing problem is not a "glitch".
New PR with extra commit on a different branch-name just repeats the problem : #3693
Plus, same is happening in test-iris-imagehash : SciTools/test-iris-imagehash#32 (comment)

@pp-mo
Copy link
Member Author

pp-mo commented Apr 1, 2020

Created PR with small changes to this, in own repo (since merged, so appears here) : pp-mo#60

In this, on own fork, same thing of the PR not linking to the Travis tests.

@pp-mo
Copy link
Member Author

pp-mo commented Apr 1, 2020

pp-mo#60 on own fork, same thing of the PR not linking to the Travis tests.

Something has changed, I'm struggling to t know how to chase this...
dask+pandas Github projects/repos are not having these problems,
But our own test-iris-imagehash is, even though it has a totally different + much simpler .travis.yml.
I guess we may need to look for something similar between iris + test-iris-imagehash travis configs?

Copy link
Contributor

@trexfeathers trexfeathers left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry @pp-mo I haven't quite finished looking at this. But I do know that you'll need to update the test data ref in travis.yml to the latest commit from iris-test-data PR 57. Catch you next week!

@trexfeathers
Copy link
Contributor

pp-mo#60 on own fork, same thing of the PR not linking to the Travis tests.

Something has changed, I'm struggling to t know how to chase this...
dask+pandas Github projects/repos are not having these problems,
But our own test-iris-imagehash is, even though it has a totally different + much simpler .travis.yml.
I guess we may need to look for something similar between iris + test-iris-imagehash travis configs?

I just re-ran the Travis tests and now I've got that information back on this PR. In order to re-run the tests I had to log out of Travis and log back in since the restart button wasn't there the first time. I don't know whether it was the log-in or the re-run that has fixed it for me.

@pp-mo
Copy link
Member Author

pp-mo commented Apr 3, 2020

I just re-ran the Travis tests and now I've got that information back on this PR

Thanks for persevering -- this is great news!
I was really worried that fixing this might long+complicated.

@pp-mo
Copy link
Member Author

pp-mo commented Apr 3, 2020

In theory, if I now just install the latest iris-test-data SHA in .travis.yml, this will then pass ...

@trexfeathers
Copy link
Contributor

This all looks great, thanks @pp-mo!

@trexfeathers trexfeathers merged commit 5eed12b into SciTools:ng-vat Apr 6, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants