Merijn Verstraaten

> As much as I'd like for ""bla語" :: ByteString to be a compile-time error, I'd rather have a runtime error telling me why something is wrong rather than having...

This issue should probably be closed since, as of 1.2, criterion now *does* have the ability to generate data per run, see the new `perRunEnv` and `perBatchEnv` functions.

I'll put it on my todo list, unless I can sucker someone else into it...

I can look into a bit if you've got any clues where in the codebase I can find the rendering for those parentheses.

Ah, it actually looks like this breaks due to an interaction with ExtraToolTip. I'll have to figure out what exactly goes wrong.

I am seeing the exact same termination with SIGSEGV behaviour with Scalene 1.5.4 and Python 3.8.12

I'm not sure what's unclear: If the number of errors that are *not* typed holes is >0, then don't report "errors" from typed holes. If number of errors from typed...

The distinction is that the HLS has the tools to detect the difference between typed holes and other warnings properly, any client is limited to doing lousy textual/regex matching on...

Yeah, I made a homebrew package which is here: https://github.com/nlesc/homebrew-nlesc It's, uh, beta quality ;)

On further thinking I don't think this is enough, there is no way for a user to know that the `IO` they produce via `MonadUnliftIO` will be run in the...