Commit f2f484680f155e89139e5651955ff057660bd2aa in lucene-solr's branch refs/heads/master from Christine Poerschke
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=f2f4846 ]

LUCENE-7205: Remove repeated nl.getLength() calls in (Boolean|DisjunctionMax|FuzzyLikeThis)QueryBuilder.

Commit 87d7de0d9892eab8462f08b876af3481f5b0ccc3 in lucene-solr's branch refs/heads/branch_6x from Christine Poerschke
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=87d7de0 ]

LUCENE-7205: Remove repeated nl.getLength() calls in (Boolean|DisjunctionMax|FuzzyLikeThis)QueryBuilder.

Commit 0cdb3cafb2f1bcb975be85cbc63ab124ddba31a8 in lucene-solr's branch refs/heads/branch_5x from Christine Poerschke
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=0cdb3ca ]

LUCENE-7205: Remove repeated nl.getLength() calls in (DisjunctionMax|FuzzyLikeThis)QueryBuilder and Boolean(Query|Filter)Builder.

What prompted this? I would expect NodeList.getLength() implementations to be trivial getters; no?

I agree, the NodeList.getLength() implementations should be a trivial getter. With multiple/many child nodes in the node list that would be repeated methods calls nonetheless, all returning the same result since the list doesn't change, so this is meant to be very minor optimisation (the extra `int nlLen` presumably doesn't really cost since even when called within the for-loop the getLength() result needs to be stored?).

Personally I don't think we should add extra lines of code (a form of cost in readability) where there is negligible benefit to optimize. But I'm not standing in your way.

Extra lines of code as a form of cost in readability is a fair point. Will keep that in mind going forward (and have no further optimisations like this planned at the moment). Thanks for your input!