Context: we have a bunch of test data in the linter repo that doesn't analyze cleanly (by design) and that creates noise when we want to publish.
When I learned about .pubignore I thought I was golden and in fact, doubly so, since I don't want to publish this test data anyway but I was surprised to see it still producing warnings when I dart pub lish -n.
Is this by design? Am I just holding things wrong?
Thanks!