Shape union

[KaiSzuttor]

A union of shapes.

[codecov]

Codecov Report

Merging #3493 into python will increase coverage by <1%.
The diff coverage is 94%.

@@           Coverage Diff           @@
##           python   #3493    +/-   ##
=======================================
+ Coverage      87%     87%   +<1%     
=======================================
  Files         534     537     +3     
  Lines       24367   24433    +66     
=======================================
+ Hits        21216   21275    +59     
- Misses       3151    3158     +7

Impacted Files	Coverage Δ
src/core/shapes/Shape.hpp	100% <100%> (ø)	⬆️
src/script_interface/shapes/initialize.cpp	100% <100%> (ø)	⬆️
src/core/shapes/Union.hpp	100% <100%> (ø)
src/core/shapes/unit_tests/Union_test.cpp	100% <100%> (ø)
src/script_interface/shapes/Union.hpp	79% <79%> (ø)
src/core/grid_based_algorithms/lb_boundaries.cpp	97% <97%> (ø)	⬆️
src/core/electrostatics_magnetostatics/p3m.cpp	85% <0%> (-1%)	⬇️
.../grid_based_algorithms/lbboundaries/LBBoundary.hpp	100% <0%> (ø)	⬆️
... and 1 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ed9cbaf...2e0260e. Read the comment docs.

[jngrad]

Looks good, but it would be nice to have a minimal documentation for this feature. For example, the calculate_dist() method could use a docstring like /** Calculate the distance to the closest shape */. In the Sphinx docs, file constraints.rst, one could make section 11.2 a subsection of section 11.1 (and make 11.2.1 to 11.2.4 subsubsections of 11.2, but keep 11.2.5-11.2.6 as subsections of 11.1), and have a new section 11.3 to describe the union.

[fweik]

@jngrad this should have a formal definition of what is calculated (which is not hard), but first this has to be actually feature complete ^^

[jngrad]

@fweik of course. I saw your comment when writing my review. I was a little confused at first as to what was calculated, and thought the union was a variant, until I read through the boost accumulator code ^^

[fweik]

I think this should be called Union, it's clear from the context that this is about Shapes.

[fweik]

LGTM. The change on the bondary logic need to be applied to #2701 (Walberla).

[KaiSzuttor]

@fweik I've simplified the LB boundary initialization a bit via boost.