comparison vs other ryu libraries?

[timotheecour]

according to disruptek/ryu#1 (comment),

@alaviss is writing a nimish version of Ryu that should deprecate this repo. So, the goal of this effort is to flesh out tests and benchmarks so that we can vet future implementations. I expect that the next implementation will be much friendlier to inclusion in the stdlib, for example.

but looks like this repo mentions:

This converter aims for correctness over performance, and is used as a reference to evaluate other algorithms, such as ryū. It should not be used in production due to slow arithmetic regarding larger float sizes.

have you worked on another ryu implementation?

refs libmir/mir-algorithm#330 (comment)

[alaviss]

I stopped working on ryu after @Clyybber told me he was working on ryu and/or dragonbox.

It has been 5 months since then, not sure if he has made any progress.

[Clyybber]

My ryu port (https://github.com/Clyybber/nimryu/tree/master/ryu) is finished as far as functionality goes, but could probably be nimified a bit more. The dragonbox port I haven't finished yet; it's very heavy templatized C++ code.