multi: Add fee estimate oracle for tatanka #2769
Conversation
|
A quick note: A tatanka only provides fee estimates for its chains. We might consider providing a fee estimate even if Tatanka does not run a particular chain, mainly for client consumption. The same applies to fiat rates. |
| // sources. Fee estimate values are in atoms for dcr, gwei for ethereum, | ||
| // satoshis for bitcoin and bitcoin clone blockchains (per byte sat), or the | ||
| // lowest non-divisible unit in other non-Bitcoin blockchains. |
There was a problem hiding this comment.
They are all in the DEX AtomicUnit of the asset, right?
There was a problem hiding this comment.
@ukane-philemon I though wei for eth was fine. It looks like it is returning wei.
There was a problem hiding this comment.
They are all in the DEX AtomicUnit of the asset, right?
Not sure what you mean, but they are all in their smallest denom, except eth (gwei). But a review from @JoeGruffins suggests wei so yes.
There was a problem hiding this comment.
We use "atomic units" to refer to the integer units used with asset.Wallet and asset.Backend. How asset.Wallet and asset.Backend scale the "atomic units" internally is of no concern outside of the respective wallet packages, really. Well, and here I guess. That said, we already know that there are some problems with this system. Specifically, Polygon and other evm blockchains can have fee rates smaller than 1 gwei per gas. This means that for swaps, we have to lock at least 1 gwei per gas, even if the actual network rate is substantially lower. The rate actually assessed is limited by the block's base fee rate, of course, but fee rates can determine minimum lot sizes and mm bot spreads etc. Luckily, 1 gwei / gas is very, very small even for ethereum, so the effects are not really noticeable. I do think it's worth discussing alternative ways to encode fee rates and maybe values.
There was a problem hiding this comment.
Specifically, Polygon and other evm blockchains can have fee rates smaller than 1 gwei per gas. This means that for swaps, we have to lock at least 1 gwei per gas, even if the actual network rate is substantially lower. The rate actually assessed is limited by the block's base fee rate, of course, but fee rates can determine minimum lot sizes and mm bot spreads etc. Luckily, 1 gwei / gas is very, very small even for ethereum, so the effects are not really noticeable
I think we should stick to gwei for eth for now.
| // sources. Fee estimate values are in atoms for dcr, gwei for ethereum, | ||
| // satoshis for bitcoin and bitcoin clone blockchains (per byte sat), or the | ||
| // lowest non-divisible unit in other non-Bitcoin blockchains. |
There was a problem hiding this comment.
@ukane-philemon I though wei for eth was fine. It looks like it is returning wei.
Signed-off-by: Philemon Ukane <ukanephilemon@gmail.com>
Signed-off-by: Philemon Ukane <ukanephilemon@gmail.com>
Signed-off-by: Philemon Ukane <ukanephilemon@gmail.com>
ukane-philemon
force-pushed
the
tatanka-external-wallet-fees
branch
from
534bdaa
to
a00d151
Compare
This PR adds the
txfeepkg that provided external fee estimates fortatanka. To test this PR, please change the demo client network totestnetormainnet.Chains with
mainnetandtestnetcoverage are:btc,doge,dcr,eth,ltc,dash,bch,zcash. Some fee estimate sources require Free API keys that can be generated on the provider's site.Closes #2726