Improve test suite
danielfromearth opened this issue · comments
The test suite is missing checks on the layout of the output, such as whether variables are being included or excluded from the output and how differences within each group appear.
@danielfromearth
Hello, do you have any sample files that can be used for the tests, or just as a reference?
If you opened this issue, then it means my PR didn't test the layout of the outputs either. I have not been able to check the output because I couldn't find a good reference set of files.
Yep, @kokroo, exactly. I will work next on creating a good (and small-sized) sample file(s) to include as a reference, and hopefully for use in an automated test too.
In the meantime, if you are looking for netCDF files to test, one option could be to download a couple from Earthdata https://search.earthdata.nasa.gov/search. Of course, there are plenty of other sources of netCDF files, but that has lots of different types and so it could provide many different, and perhaps difficult, test cases. (If you do test it with some of those files, I'd really love to hear about what you use and how it goes).
@kokroo, I just added (in PR #15) to the develop branch a test that checks the text file output against a pre-generated "golden file". When you have a chance, I think it would be good for you to try updating your feature branch with the latest in develop, and then updating the code as necessary so it passes the new tests. How does that sound?
@danielfromearth Hello Daniel, thank you so much for creating these.
I will try to get it done in 1-3 days.
Closing this to make a new release with the latest, and I'm going to upload it to PyPI.
If the tests present new problems, we can create a new issue.
Closing this to make a new release with the latest, and I'm going to upload it to PyPI.
If the tests present new problems, we can create a new issue.
@danielfromearth
Hello Daniel, sorry, I got caught up with real life and could not get back on this timely. Regarding PRs, I just read this ( nasa/radbelt#47 (comment)) on another NASA repo. I am still glad to submit PRs but I just wanted to give you a heads-up and confirm if you can accept them.
Thanks!
Hi @kokroo, no worries! This is a good question about PRs — I'm looking further into it now, and am hoping to have more information about it by late next week.
Hi @kokroo, I've spoken with several NASA folks and it seems there is no issue with external collaborators contributing to a NASA repository licensed with the NASA Open Source Agreement (NOSA). Such contributions are very much welcome!
The few things to still note are from this Wikipedia page describing NOSA: NOSA is "approved as an open source license by the Open Source Initiative (OSI)", and NOSA apparently expects each contribution to be an "original creation".
Hi @kokroo, I've spoken with several NASA folks and it seems there is no issue with external collaborators contributing to a NASA repository licensed with the NASA Open Source Agreement (NOSA). Such contributions are very much welcome!
The few things to still note are from this Wikipedia page describing NOSA: NOSA is "approved as an open source license by the Open Source Initiative (OSI)", and NOSA apparently expects each contribution to be an "original creation".
@danielfromearth Thank you so much for following up on this Daniel. I can send in my PRs now!
@lpsinger Radbelt is also under NOSA if I am not wrong and Daniel here received information that it's okay to accept contributions. I am really confused now since OnAIR asked me for a contributor agreement. What gives? :D
At least at my NASA center (Goddard), for software that is released under NOSA, the lawyers require us to collect hand-signed contributor license agreements from external contributors, and they have to process and approve the contributor license agreements. I have attempted this process several times over the past few years, and in not one instance has Goddard's legal counsel ever approved and processed a contributor license agreement for one of my projects.
This package may be different because it looks like it was released at Langley, and they may have more tractable lawyers there.
Hi @kokroo, sorry for the delay in responding. You and @lpsinger have brought up some important questions that I haven't yet received clear guidance about, so I'm looking into this further and am waiting to hear confirmation of the approach that my NASA center (Langley) follows. Will let you know as soon as I know!
Hi @kokroo, so I've gotten some unfortunate news. As of now, the Langley Research Center does require a signed Contributor License Agreement as well. I haven't found out what exactly that looks like, so I will try to find out more about that now.
@danielfromearth
Hey Daniel, thanks for following up on this!
The amazing folks handling OnAIR at https://github.com/nasa/OnAIR/ have all of that setup already, perhaps they could guide you in the right direction?
Thanks @kokroo, for pointing out the OnAIR repository, that's helpful. Sorry again for the delay. I am still working to get answers to all the questions that need to addressed for Langley regarding CLAs, but I'll keep providing further updates as I get them!
@danielfromearth Hello Daniel, I hope you are doing well.
Any updates on this situation?
I have prepared a new PR!
Hi @kokroo, yes — as tracked in this issue, the license for ncompare is now Apache License 2.0, so please go ahead and open the new PR!