FeeCurrency directory

[pahor167]

Description

Introduction of FeeCurrencyDictionary for L2 geth client.

Tested

Unit tests

[karlb]

According to https://github.com/celo-org/celo-blockchain-planning/issues/172#issuecomment-2066858102, we should extract an IFeeCurrencyDirectory interface from this contract. I think that should include

• getCurrencies
• getCurrencyConfig
• getExchangeRate
• CurrencyConfig

If we don't want to expose the CurrencyConfig, we should skip getCurrencyConfig and CurrencyConfig and add a getIntrinsicGas(address token) instead.

[pahor167]

According to celo-org/celo-blockchain-planning#172 (comment), we should extract an IFeeCurrencyDirectory interface from this contract. I think that should include

• getCurrencies
• getCurrencyConfig
• getExchangeRate
• CurrencyConfig

If we don't want to expose the CurrencyConfig, we should skip getCurrencyConfig and CurrencyConfig and add a getIntrinsicGas(address token) instead.

According to celo-org/celo-blockchain-planning#172 (comment), we should extract an IFeeCurrencyDirectory interface from this contract. I think that should include

• getCurrencies
• getCurrencyConfig
• getExchangeRate
• CurrencyConfig

If we don't want to expose the CurrencyConfig, we should skip getCurrencyConfig and CurrencyConfig and add a getIntrinsicGas(address token) instead.

Interface added

[pahor167]

The currencyIdentifier feels like something that does not belong into the FeeCurrencyDirectory, since it depends on the oracle. I could image oracles having all kind of identifiers:

• token address
• some other address
• string
• uint id
• two of the above (one for "from" and one for "to")
• no id since it only provides values for a single currency

I would suggest not including it and relying on the oracle having an interface that takes the token address. If it doesn't we could add a contract in between to bridge the interfaces. Or would that be too gas expensive or too cumbersome in a different way?

Apart from that, this looks good to me.

The currencyIdentifier feels like something that does not belong into the FeeCurrencyDirectory, since it depends on the oracle. I could image oracles having all kind of identifiers:

• token address
• some other address
• string
• uint id
• two of the above (one for "from" and one for "to")
• no id since it only provides values for a single currency

I would suggest not including it and relying on the oracle having an interface that takes the token address. If it doesn't we could add a contract in between to bridge the interfaces. Or would that be too gas expensive or too cumbersome in a different way?

Apart from that, this looks good to me.

We will have further discussion regarding IOracle interface, for now I'm leaving it as it is

[openzeppelin-code[bot]]

FeeCurrency directory

Generated at commit: 5a4987d5f15ec0542de7db4273a2d6ca56c184bf

🚨 Report Summary

	Severity Level	Results
Contracts	Critical High Medium Low Note Total	2 3 0 15 40 60
Dependencies	Critical High Medium Low Note Total	0 0 0 0 0 0

---

For more details view the full report in OpenZeppelin Code Inspector