Skip to content
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

Plotting an already plotted dataset causes the new plot to only show model #1546

Closed
m2cci-NMZ opened this issue May 18, 2020 · 8 comments · Fixed by #1556
Closed

Plotting an already plotted dataset causes the new plot to only show model #1546

m2cci-NMZ opened this issue May 18, 2020 · 8 comments · Fixed by #1556
Assignees
Labels
Critical High priority Defect Bug or undesirable behaviour
Milestone

Comments

@m2cci-NMZ
Copy link
Contributor

When a dataset is already plotted, plotting the same dataset in a different Fitpage only shows the model in the plot.

@krzywon
Copy link
Contributor

krzywon commented May 18, 2020

This seems very similar to #1454. Was it not fully fixed?

@m2cci-NMZ
Copy link
Contributor Author

Indeed, @rozyczko fixed it in PR #1481, but somehow the commit didn't make it to ESS_GUI branch? This is strange as it shows as merged.

@krzywon
Copy link
Contributor

krzywon commented May 18, 2020

Looking at the git history for DataExplorer.py, the change made in PR #1481 was removed to fix an issue with 2D plots in PR #1508.

@butlerpd
Copy link
Member

wow - nice catch @krzywon!!.. but I am really getting confused!  How is this possible? Moreover, this kind of thing seems to be happening not that infrequently any more (a new PR overwriting an older one)  I thought the whole point of version control software was to prevent this kind of thing?  How come the PR merged the old code back in and overwriting the new? Clearly there is something about version control I don't understand here. Most importantly, what can we do to avoid this continuing to happen in the future?

@krzywon
Copy link
Contributor

krzywon commented May 18, 2020

@butlerpd I think the code added in #1481 to force a single 1D plot caused a different issue with 2D plots that was addressed in #1508. This was likely more an issue with how the original PR was tested and not with version control.

@rozyczko
Copy link
Member

Indeed this seems to have been an issue with the unit testing (or lack of thereof) for this bug.
One thing to remember is to always write unit tests, especially for bugs that got fixed, this would allow us to see failures immediately.
We used to be pretty stringent wrt unit tests for all functionality but got slightly less so the less time we had, especially before releases.
I think we should insist (as we originally did) on having unit tests for everything new that goes into the code. Part of the code review is also making sure that there is a (working) unit test committed.

@rozyczko
Copy link
Member

Fixed the logic and added tests. Hopefully this sorts things out for all the edge cases.

@rozyczko rozyczko assigned rozyczko and unassigned m2cci-NMZ May 28, 2020
m2cci-NMZ added a commit that referenced this issue May 29, 2020
Fixed the edge cases and added a beefy unit test. #1546
@krzywon
Copy link
Contributor

krzywon commented May 29, 2020

Was this resolved when the PR was merged? We should close the ticket if that is the case.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Critical High priority Defect Bug or undesirable behaviour
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants