C#: Respect order of LGTM_INDEX_FILTERS in buildless extraction
#15325
Conversation
That is, using `exclude:**/*\ninclude:**/*` should include everything.
| for (int i = pathFilters.Count - 1; i >= 0; i--) | ||
| { | ||
| var pathFilter = pathFilters[i]; |
There was a problem hiding this comment.
Minor: I wonder if pathFilters.Reverse() or foreach (var pathFilter in pathFilters.Reverse<PathFilter>()) would be clearer, rather than switching to indices.
There was a problem hiding this comment.
I actually looked on stackoverflow for this, which said that using the inline .Reverse in the foreach was quite inefficient.
There was a problem hiding this comment.
I am not sure off the top of my head if having the .Reverse in the foreach would get evaluated with every iteration, but replacing pathFilters.Sort with pathFilters.Reverse should at least be more efficient than .Sort (although likely not as efficient as using the indices of course).
There was a problem hiding this comment.
I can swap the reverse loop with a Reverse() on the list if you like. Personally I've seen enough reverse iterations of lists with explicit indexing that it just clicks 🤷
There was a problem hiding this comment.
I don't feel strongly about this, so feel free to just leave it as is 👍🏻
csharp/extractor/Semmle.Extraction.CSharp.DependencyFetching/FilePathFilter.cs
Outdated
Show resolved
Hide resolved
Co-authored-by: Michael B. Gale <mbg@github.com>
That is, using
exclude:**/*\ninclude:**/*should include everything (but that wouldn't happen with the old code)