Twap Wasm Bindings

[mattverse]

Sub component of : #2187

What is the purpose of the change

This PR adds wasm bindings for query on two APIs GetArithmeticTWAP and GetArithmeticTwapToNow.

This PR invlolves adding twap keeper to the wasm query plugin to use the API provided by the twap keeper.

Brief Changelog

• Add wasm bindings for twap APIs(GetArithmeticTWAP, GetArithmeticTWAPToNow)

Testing and Verifying

Testing is TBD and to be sorted out, as of right now, do not have exact test cases for testing.

Documentation and Release Note

• Does this pull request introduce a new feature or user-facing behavior changes? (yes / no)
• Is a relevant changelog entry added to the Unreleased section in CHANGELOG.md? (yes / no)
• How is the feature or change documented? (not applicable / specification (x/<module>/spec/) / Osmosis docs repo / not documented)

[p0mvn]

utACK

[mattverse]

Figuring plans for testing w/ exact numbers for TWAP

[mattverse]

Converting to draft.

Current status for this PR is that I've added test codes for new queries in custom bindings test, whilst getting error Error parsing into type osmo_reflect::msg::QueryMsg: unknown variant arithmetic_twap, expected one of full_denom, pool_state, spot_price, estimate_swap: query wasm contract failed. Further investigation needed in order to figure out where this is coming from / why / where this is happening

[mattverse]

Confirmed that this is happening bc we don't have the .wasm files thats used for tests updated with the bindings(rust side). Replacing those files with the updated .wasm(after the rust side code gets reviewed and merged) files should fix the broken tests

[ValarDragon]

This LGTM!

[nicolaslara]

This looks good to me.

My only comment is that it's not clear which queries should have their own bindings and which should go through StargateQuery with convenience wrappers in https://github.com/osmosis-labs/osmosis-rust. I don't see manipulation of the query result to avoid non-determinism, in which case adding it to the stargate query whitelist should have the same effect.

[ValarDragon]

wdym

[ValarDragon]

Oh I see, the query errors in this case, interesting

[ValarDragon]

here's the error message: looking for a time thats too old, not in the historical index. Try storing the accumulator value.

[ValarDragon]

ah the error is correct, this is a test error here =p

Committing fix here.

[ValarDragon]

Hrmm actually its not so clean, this is actually a deeper issue with the binding using Unix time for the time passing here.

The reason its not working at the start time is because its being passed in unix time in miliseocnds. However the state record has greater precision.

Trying to see more about how to fix this

[ValarDragon]

So I think what we need to do is make the TWAP binding parse from UnixMiliseconds, and then in the state machine we write at UnixMilisecond precision.

For this PR, can we change the binding to be from Unix milisecond and then call it good from that? Then in the test if we set ctx.BlockTime to be at a milisecond boundary, this should be fixed. Once we do #2209 should also get fixed state machine side as well.

[ValarDragon]

@mattverse please check my latest commit to fix the time threading.

if you agree, please make docs for the binding, for start time being in Unix time miliseconds. (and more importantly make the docs rust side)

Also communicate to Mars that we can change the time input API to be whatever they'd prefer.

Right now there will be an edge case with StartTime for a TWAP, that will get fixed in a subsequent commit pre-release. Namely that if start time = pool creation time, and the block time is not milisecond aligned, it may not work. Setting it to be 1 milisecond after pool creation time would fix this for now as workaround, until next week we change it state machine to not be nanosecond precise in state machine storage.

(The state machine computes the correct result, its that this binding API doesn't offer sufficient precision)

[mattverse]

Agreed with using Unix time millisecond for now for precision, updated documentation

[ValarDragon]

(I rebased this after I merged the twap branch)

[ValarDragon]

ah we forgot to make this milli in the version we sent to mars

[czarcas7ic]

I can spin one up fairly quickly if this needs to be changed