Colors (as in color-color or hardness-intensity diagrams)

[matteobachetti]

Depends on #754

[pep8speaks]

Hello @matteobachetti! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2024-01-15 10:50:12 UTC

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (1fc9b90) 79.07% compared to head (115b9e6) 93.98%.

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #781       +/-   ##
===========================================
+ Coverage   79.07%   93.98%   +14.91%     
===========================================
  Files          43       43               
  Lines        8697     8762       +65     
===========================================
+ Hits         6877     8235     +1358     
+ Misses       1820      527     -1293     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[dhuppenkothen]

Great work! The usual nitpicky comments you expect from me, and a couple of questions. :)

[dhuppenkothen]

top-level import?

[dhuppenkothen]

My main comment is that I'm not convinced by the behaviour in TimeSeries that suggests that event lists have dt=0. That seems very counterintuitive to me. Either event lists are also time series, in which case dt should be set to either the time resolution of the instrument, or to the intervals between events. I think by definition time series should not be allowed to have dt<=0, because that kind of defeats the purpose of a class that explicitly has one variable be ordered time. I would like to see if there's maybe a different mechanism to make that distinction that is also simple but doesn't break basic assumptions about how time series work?

[matteobachetti]

@dhuppenkothen the problem is that some missions do not declare a time resolution. What do you use for dt then? 0 means simply that there is no known time resolution, or that it's probably irrelevant for the purpose of the analysis.

[matteobachetti]

Ok, I changed the logic to eliminate some of the confusion, and eliminated method-level imports as promised ;)

[matteobachetti]

Note to self: fix the return value in the docstring of get_intensity_evolution Done

[dhuppenkothen]

LGTM! 👍