Multiple rms closest

[znicholls]

Pull request

Please confirm that this pull request has done the following:

• Tests added
• Documentation added (where applicable)
• Example added (either to an existing notebook or as a new notebook, where applicable)
• Description in CHANGELOG.rst added

Adding to CHANGELOG.rst

Please add a single line in the changelog notes similar to one of the following:

- (`#XX <https://github.com/GranthamImperial/silicone/pull/XX>`_) Added feature which does something
- (`#XX <https://github.com/GranthamImperial/silicone/pull/XX>`_) Fixed bug identified in (`#YY <https://github.com/GranthamImperial/silicone/issues/YY>`_)

[znicholls]

@Rlamboll still a draft but wanted to get your thoughts at this stage before I write an example notebook

[codecov-commenter]

Codecov Report

Merging #145 (34adf58) into master (8d6bbbc) will decrease coverage by 0.62%.
The diff coverage is 88.00%.

@@            Coverage Diff             @@
##           master     #145      +/-   ##
==========================================
- Coverage   96.98%   96.35%   -0.63%     
==========================================
  Files          23       23              
  Lines        1159     1206      +47     
  Branches      252      257       +5     
==========================================
+ Hits         1124     1162      +38     
- Misses         16       21       +5     
- Partials       19       23       +4     

Impacted Files	Coverage Δ
src/silicone/database_crunchers/rms_closest.py	89.51% <88.00%> (-5.29%)	⬇️

---

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 8d6bbbc...34adf58. Read the comment docs.

[Rlamboll]

Hi Zeb, I like the idea of the tool and the syntax for infill multiple seems fine. I find the interrupting loops in the code quite hard to follow, so more comments would be useful, as well as more documentation since it's unusual for a cruncher to have this sort of ability. I'd also do more tests, like a test where it's possible to return the wrong answer for two of the variables without inventing new data - currently you are choosing the right pathway from only one option on the N2O two-pathway use case!

[znicholls]

interrupting loops

Is this a synonym for nested loops?

like a test where it's possible to return the wrong answer for two of the variables without inventing new data

Ok cool will have a think. If you have any more you want to suggest please fire away (if you want to make a PR into this one with the use cases of interest that could also work well).

[Rlamboll]

No, by interrupting I mean loops featuring continues and breaks. I don't belong to the hard school of "never use these, put it all in an if statement or control your loops" but a few comments would go a long way in increasing readibility.

I'm quite busy now and writing tests for this would be low on my to-do list but I can do it in the distant future.

[znicholls]

No, by interrupting I mean loops featuring continues and breaks. I don't belong to the hard school of "never use these, put it all in an if statement or control your loops" but a few comments would go a long way in increasing readibility.

Ye cool got it

I'm quite busy now and writing tests for this would be low on my to-do list but I can do it in the distant future.

Yep let's leave this sitting as a proof of concept until we find time again