My PR is failing due to the
syntax_highlighting_not_quadratic test, but nothing in the PR seems non-linear. Anyone have any idea why this is happening?
I don't see anything wrong with the PR, maybe the test is too strict.
I need to look closer into it, but it definitely is quadratic. The PR adds
is_public check, so presumably that is linear, rather than O(1)?
Playing with the test locally and looking at https://github.com/rust-analyzer/rust-analyzer/blob/master/docs/dev/README.md#profiling might help to confirm this
Or we could also try plotting the time without the PR changes, just to check that it's really linear right now.
Good suggestion! It looks like it is quadratic rn?