hoogle VS grin

Dimensions behave somewhat like a "type system" for math. These dimensional-analysis tricks act like the trick you see in Haskell sometimes, where you can easily guess an implementation of an expression once you know it's type (or e.g. search by type signature https://hoogle.haskell.org/ )

Haskell Spotlight makes vscode a client for hoogle. It isn't too different than jumping into your browser and type https://hoogle.haskell.org/. The main advantage is that you have everything in one place

https://hackage.haskell.org/package/base and https://hoogle.haskell.org are automatically up to date and better searchable than almost any other reference of any other programming language. maintaining a redundant reference that needs to be kept up to date manually is simply stupid.

Ideally like this: https://zio.dev/reference/#concurrency

Or this: https://hoogle.haskell.org/?hoogle=fork

And yeah I get it, it's hard to Google for punctuation operators in languages because it doesn't give useful search results (but not impossible, for example, Haskell has a search engine for documentation that handles symbols/punctuation).

{ "customLocalFormatters.formatters": [ { "command": "make format", "languages": ["haskell"] } ], "emeraldwalk.runonsave": { "commands": [ { "match": "*.hs", "isAsync": true, "cmd": "make retag retag_file=${file}" } ] }, "ghcid.command": "make ghcid", "goto-documentation.customDocs": { "hs": "https://hoogle.haskell.org/?hoogle=${query}" } }

You had a look at Hoogle?

https://hoogle.haskell.org/

For some type signatures there is (are) only one (or only a few) meaningful implementation(s).

You could take a look at GRIN, which is an IR especially made for optimizing functional languages before passing them to something like LLVM. (keep in mind that that project is still very early in development) See also their paper.

There is the GRIN optimizer that is implemented based on Urban Boqist PhD thesis. It works really well with LLVM because GRIN removes all indirect calls so it allows LLVM to do its job really well. But Boq GRIN PhD does not tell how to support the complex GHC primop and RTS features. That's why I have another direction as well.