Simon Hengel

@echatav do you think it's possible to reproduce your issue to the point where it does not require a database?

@vaclavsvejcar I think your issue is due to `:set -XQuasiQuotes` triggering https://gitlab.haskell.org/ghc/ghc/-/issues/20670.

@vaclavsvejcar as a workaround, I think you can't set `-XQuasiQuotes` in Haddock examples, but you can set it in `doctest/Main.hs` instead.

@vaclavsvejcar you are welcome. Ideally, we would enable `-XQuasiQuotes` by default (as we already do for `-XTemplateHaskell`). However, this will render e.g. `[i|i

@echatav I don't have any spare cycles to investigate this myself right now. If you (or somebody else) can provide a somewhat more minimal example that reproduces the issue then...

@chshersh hey👋 I'll try to look at it and see if I can help, but it may take a week or so until I get to it. If it gets...

@chshersh in general, if you can provide exact steps on how to reproduce, this will save me a lot of time. Here is what I tried: ``` $ git rev-parse...

> Even proper solution would be to have a way to turn off/configure module discovery per executable/test-suite/benchmark, in addition to filtering out main-is files. This is already possible, no? Specifically,...

Regarding filtering out `main-is`, generally that makes sense (or at the very least it should not be harmful). But the code may get messy. I would want to see (or...

I'm not sure if it is still decidable in a post-processing step.