Use Case #9 happy path - pearson correlation fails

[HelenClifton]

Expected behavior

Ran the use case#9 happy path using the GUI.

Actual behavior

Pearson correlation failed

Traceback:
UC9 happy path traceback.txt

Steps to reproduce the problem

See:
steps to reproduce.docx

Specifications

cate-0.9.0-dev.7
Windows 7 Professional

[HelenClifton]

#309 is similar. It may or may not be related.

[JanisGailis]

The CLI Happy Path script for UC09 runs as expected on Linux both from current master (1.0.0.dev1) and from 0.9.dev7 tag.

My guess is that somehow when the time range in the GUI is selected, one end is excluded from it, while from the CLI it is inclusive. This would explain why pearson_correlation fails with that message.

[JanisGailis]

I disagree with the ops label. The CLI happy paths are working as expected, hence, I don't think the problem is in ops. I will be able to do some debugging of this late on Monday, as I'm travelling for most of it.

[JanisGailis]

I can not reproduce this on the GUI of Cate 0.9.dev7. I followed the steps in the document and successfully created the correlation dataset.

[kbernat]

Couldn't reproduce either. Used core and gui from git-master branch (1.0.0.dev1)

[HelenClifton]

Tried this with cate 1.0.0-dev2 and could not reproduce.
This can be closed.

[HelenClifton]

@forman, @mzuehlke, @JanisGailis, @kbernat : It's back in cate-2.0.0-dev.1.
Traceback is here
issue440_traceback.txt
Steps to reproduce:
issue440_steps_to_reproduce.docx

[papesci]

Dear All, i have executed all the steps successfully on a Linux machine (ubuntu).
Please find attached the workspace
UC9.tar.gz

[HelenClifton]

@forman, @mzuehlke @JanisGailis @kbernat OK... confirmed that it was operator error ...
I'd set end time constraint to 2007-03-01 instead of 2007-03-31
This can be closed again :)