Josh Berdine
Josh Berdine
It would be good to test that, if ocamlformat was designed with an attempt to mimic the indentation of ocp-indent. The basic implementation approaches are very different, and it would...
I think perhaps my reply didn't come across as intended, I'm sorry for that. I certainly agree that there is a ton of experience codified into ocp-indent, and don't want...
Ouch. But I'm not personally all that happy with the ocamlformat output in this case.
Retitled as regression testing has been in place for a while, but more thorough automated testing remains.
The issue itself is still completely relevant, even if the implementation strategy explored so far hasn't gone the whole way.
One example could be something like: ```ocaml let inc x = trace ~call:(fun {pf} -> pf "%d" x) ~retn:(fun {pf} -> pf "%d") @@ fun () -> x + 1...
Yes, parens might be hard. An intermediate point could be to have a small set of tokens that are allowed to move, like parens, and still leave us in a...
Is this, or a slightly weaker version that allows motion relative to parens, be something that ought to be added to the 1.0 wishlist? So far, this is the only...
Yes, changing the margin on-the-fly is a no-go, `Format` has way too much internal state to make that close to feasible. I suspect that this issue is due to how...
Do you mean that e.g.: ``` (* 5. xxxx xxxx xxxx xxxx xxxx *) ``` should be preserved, while currently it formats to: ``` (* 5. xxxx xxxx xxxx xxxx...