FIX: Broken Function.get_value_opt for N-Dimensional Functions

[giovaniceotto]

Pull request type

• Code changes (bugfix, features)

Checklist

• Tests for the changes have been added (if needed)
• Docs have been reviewed and added / updated
• Lint (black rocketpy/ tests/) has passed locally
• All tests (pytest --runslow) have passed locally
• CHANGELOG.md has been updated (if relevant)

Current behavior

As reported by @styris00, the following code shows an error with Function.get_value_opt when used with 2D functions.

import numpy as np
from rocketpy import Function

arr = np.array([(1, 1, 2), (3, 6, 8),
            (8, 8, 9)])

myFun = Function(
    arr,
    interpolation="shepard",
    extrapolation="natural",
    title="myFun"
)

myFun.set_get_value_opt()

print(myFun.get_value(1, 1)
print(myFun.get_value_opt(1, 1))

Output is:

2.0
1.0

New behavior

The bug was fixed.

Breaking change

• No

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (0ef9849) 70.91% compared to head (8d6f244) 70.95%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #492      +/-   ##
==========================================
+ Coverage   70.91%   70.95%   +0.04%     
==========================================
  Files          55       55              
  Lines        9261     9262       +1     
==========================================
+ Hits         6567     6572       +5     
+ Misses       2694     2690       -4     

Flag	Coverage Δ
unittests	70.95% <100.00%> (+0.04%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[Gui-FernandesBR]

I wonder when this bug was introduced. The tests haven't been modified recently.

[giovaniceotto]

I wonder when this bug was introduced. The tests haven't been modified recently.

I didn't have enough time to track it down (the snake case commit makes things harder). However, the old test was never going to catch the bug.

What was going on is that, considering x, y, z, get_value_opt was returning y instead of z. And, in the test, y was equal to z, so it passed.

[giovaniceotto]

Are we going to make this into a hotfix?
@MateusStano

[MateusStano]

Are we going to make this into a hotfix?
@MateusStano

Yes we are!