Trevor Elliott
Trevor Elliott
What do you think about adding the `cache.aborted` stat into the two tests that were modified? That way we could see any of the cases where the poisoning caused `maybeCacheGlobalStateAndFiles`...
It looks like the test failure is due to a cached file not raising the same duplicate type_member error. If you want to commit the test diff and make an...
Looks like tests are failing the formatting step, could you run `cargo fmt` and commit the results?
This looks like it might be a duplicate of #8930.
`sorbet-runtime` and `memoist` behave similarly in that they wrap an existing method definition with additional functionality (`sig` in `sorbet-runtime` wraps the method definition with runtime type checking, and `memoize` in...
Fixed in #8291
That's a great workaround, thank you!
You can put the [`private` keyword before the `sig`](https://sorbet.run/#%23%20typed%3A%20true%0Aclass%20A%0A%20%20extend%20T%3A%3ASig%0A%0A%20%20private%0A%20%20sig%20%7Breturns%28Integer%29%7D%0A%20%20attr_reader%20%3Areader%0A%0A%20%20def%20foo%0A%20%20%20%20self.reader%0A%20%20end%0Aend%0A%0AA.new.reader) to get the behavior you're looking for, though it would be nice if the attr rewriter supported the annotation between the...
The failures are due to changing the serialization of the `Block` and `NumBlock` Nodes. If the last commit is the right path forward, the `style` parameter will need to be...
I approve this license change