Jaro

So the inline version does not actually take advantage of those rules. Here is the relevant part of the core dump of the inline version: ```Haskell $wgo4_sciy ww_sciw w_scit =...

Yes, and I also found out that the [`D.B.Char8.dropWhile`](https://github.com/haskell/bytestring/blob/8c631dfdaaffdc5a42f94c66231577270d6301b1/Data/ByteString/Char8.hs#L522-L530) doesn't have the same rules as `D.B.dropWhile` and it is annotated with `INLINE [1]`, so it also won't inline before applying...

It's also in the wayback machine: https://web.archive.org/web/20180701172843/http://www.cse.unsw.edu.au:80/~chak/papers/papers.html

I've added these benchmarks: ```haskell , bgroup "pack" [ bench "not fused" $ nf S.pack (replicate nRepl 0) , bench "fused" $ nf (S.pack . replicate nRepl) 0 ] ,...

I was kind of waiting for a response on [this thread](https://github.com/haskell/bytestring/pull/494#discussion_r820141174): > As the benchmarks show, if no fusion happens then there is no benefit in inlining (it even costs...

I've done a bit of investigation into what would be necessary, here are my findings: All the relevant documentation is here: https://docs.microsoft.com/en-us/windows/console/ I collected this data by commenting out the...

I haven't looked too long at terminfo because it was pretty easy to use the mingw package. I agree that it would be nicer to use the native API. I...

OK, then I think the best course of action is to first consolidate all the unix dependent code behind such an abstract `NativeHandle` interface. Then we don't really need to...

I'm not working on anything at the moment. My thought was that the platform independent part would use `NativeHandle` to represent the native handle type. But maybe it's possible to...

@chessai try passing the `-v3` flag to cabal.