Eric Woolsey

@rapcmia if you've got one that works I would be very appreciative.

Would also love to have support for tokio. Would be a game changer for my project. This is a massively cool project, thanks so much for your efforts.

Not really related to your PR, but a higher path version of `wasm-bindgen` needs to be added since we depend on features that require a version > 0.2.0. My build...

Any plans to update to blink.cmp?

@mehalter thanks for the well thought out response! That makes total sense! I'll give a try to the astro community implementation. PS thanks for all your work on this project!...

We're currently using `anvil` for [optimism/tests](https://github.com/ethereum-optimism/tests) work and we need this feature, so if there are no objections I'll probably start working on it right away.

@zerosnacks let me know if this looks good to you: https://github.com/foundry-rs/foundry/pull/8749

Slow query plan: ``` HashAggregate (cost=205051.45..205055.19 rows=214 width=40) (actual time=3244.923..3279.311 rows=11 loops=1) Group Key: (time_bucket('1 day'::interval, event."time")) Batches: 1 Memory Usage: 40kB -> Gather (cost=1000.44..202209.13 rows=162418 width=48) (actual time=11.204..3203.753 rows=166764...

> I was wondering, why does you first query have `GROUP BY bucket, price;`? If you're calculating `first(price, time)`, shouldn't it be `GROUP BY bucket`? Sorry that was a typo...

>This family of optimizations for DISTINCT ON is usually called "Skip Scan". IIRC we don't have something like that for time_bucket, unfortunately. @akuzm Forgive me I know nothing about postgres...