01mf02
feat nushell in the performance comparison
Hey team
Nushell having file parsing (including JSON) capabilities and being also written in Rust, I think it would be interested to add it to the performance comparison
Object
https://github.com/01mf02/jaq/blob/e126bb48ce21939254270b3bb55bf7dcb0589468/README.md?plain=1#L131-L157
References
- https://www.nushell.sh/
- https://github.com/nushell/nushell
Interesting, yes, but nushell implements a completely different language. It cannot run jq programs. So how should we compare it?
I was thinking about comparing only the comparable features from the list at
https://github.com/01mf02/jaq/blob/e126bb48ce21939254270b3bb55bf7dcb0589468/README.md?plain=1#L156-L186
(and leaving out the features that aren't present from jq/jaq in nushell)
@Malix-Labs, I think that this is out of scope for the benchmark. If I do not limit the benchmarked programs to jq interpreters, then what kind of languages/interpreters should I benchmark? Rust? Python? Julia? This would dilute the purpose of the benchmark and make it too large. If Nushell would be a jq interpreter, then OK. But it's not.
The point of this benchmark is to evaluate the same programs with different interpreters that run the same programs. I could easily come up with equivalent Rust programs (and, with some effort, probably Nushell programs) for each of these programs in the benchmark, but in the end, this has little explanatory power.
i feel like nushell is a special case, especially since it is taking jq market share https://www.nushell.sh/cookbook/jq_v_nushell.html but i understand for sure