Fix ScoredPhrase Bug

[apmoriarty]

Fixes #2441

• added a test suite for ContentFunctionsDescriptor
• corrected how fields are parsed from a scored function when calling fieldsAndTerms

[drewfarris]

It would be nice to add some additional commentary in the comments, but fundamentally this code looks ok.

For future work, it would be nice to consider restructuring the code to avoid a bunch of nested if/else statements. I'm curious how sonar would rate the cyclomatic complexity of this class.

[apmoriarty]

It would be nice to add some additional commentary in the comments, but fundamentally this code looks ok.

For future work, it would be nice to consider restructuring the code to avoid a bunch of nested if/else statements. I'm curious how sonar would rate the cyclomatic complexity of this class.

I originally wanted to break apart that method but decided not to for the sake of keeping the change set small. Since you've asked, I'm more than happy to split out the scored phrase logic into it's own method. Sonar Lint shows the method complexity at 100 -- far too much.

[apmoriarty]

It would be nice to add some additional commentary in the comments, but fundamentally this code looks ok.
For future work, it would be nice to consider restructuring the code to avoid a bunch of nested if/else statements. I'm curious how sonar would rate the cyclomatic complexity of this class.

I originally wanted to break apart that method but decided not to for the sake of keeping the change set small. Since you've asked, I'm more than happy to split out the scored phrase logic into it's own method. Sonar Lint shows the method complexity at 100 -- far too much.

I took a look at this and to do it right would require a massive expansion of scope. Wrote #2456 to track that.