Ellie Hermaszewska

@dschrempf That's probably worthy of it's own issue

Yeah, that is quite clunky and not necessarily without runtime overhead or a clarity cost. Perhaps a neat way of allowing the user to signal the wrapping would be to...

@johnkslang I think typically all of those would go to stderr as they're all "diagnostic output" https://pubs.opengroup.org/onlinepubs/009695399/functions/stderr.html

These issues seem relevant, they mention compass failure and using GPS heading on multirotors: https://github.com/iNavFlight/inav/issues/991 https://github.com/iNavFlight/inav/issues/1323 https://github.com/iNavFlight/inav/pull/2802

It would be a shame for this to stall over argument format bikeshedding.

It seems a shame to have to disable that useful feature in projects which do rely on whitespace in quasiquotes though.

Sure, I don't think it's a high priority, in fact I only bumped into this when stylish-haskell removed whitespace in a qq which I didn't want anyway :). It might...

Splitting into 26 subdirectories should give ~1000 files in the largest case. Pretty ugly though. No modern filesystem should have any issue with that number however, nor git. On Thu,...

Probably best to implement `displayException` for `NoHackageTarBallFound` to include this information and set the exception handler in cabal2nix to use that when reporting exceptions at the top level.

`displayException` would have to be implemented in hackage-db, and then maybe something like https://hackage.haskell.org/package/uncaught-exception-0.1.0/docs/Control-Exception-Uncaught.html#v:setDisplayExceptionHandler used in cabal2nix. A less invasive fix might be to just catch `NoHackageTarBallFound` in `cabal2nix` and...