feat: store ipfsCID as a public variable

[smol-ninja]

Closes #281.

Note

• ipfsCID is included in the MerkleLockup.ConstructorParams struct.
• ipfsCID is stored as string because of it's nondeterministic CID length.

[andreivladbrg]

To keep the createMerkleLockup API unaffected

not sure what you mean. unaffected compared to what?

i haven't properly reviewed it yet, but it doesn't make sense for me to not include it in the struct

[smol-ninja]

not sure what you mean. unaffected compared to what?

// current API
function createMerkleLockupLL(
    MerkleLockup.ConstructorParams memory baseParams,
    ISablierV2LockupLinear lockupLinear,
    LockupLinear.Durations memory streamDurations,
    string memory ipfsCID,
    uint256 aggregateAmount,
    uint256 recipientsCount
)
    external
    returns (ISablierV2MerkleLockupLL merkleLockupLL);

vs

// API if we include ipfsCID in baseParams
function createMerkleLockupLL(
    MerkleLockup.ConstructorParams memory baseParams,
    ISablierV2LockupLinear lockupLinear,
    LockupLinear.Durations memory streamDurations,
    uint256 aggregateAmount,
    uint256 recipientsCount
)
    external
    returns (ISablierV2MerkleLockupLL merkleLockupLL);

By not including ipfsCID in the baseParams, the createMerkleLockupLL API doesn't change. That means these changes wouldn't require integrators and Sablier UI to update their API calls.

What do you think about this?

[andreivladbrg]

That means these changes wouldn't require integrators and Sablier UI to update their API calls.
// current API
function createMerkleLockupLL(
MerkleLockup.ConstructorParams memory baseParams,
ISablierV2LockupLinear lockupLinear,
LockupLinear.Durations memory streamDurations,
string memory ipfsCID,
uint256 aggregateAmount,
uint256 recipientsCount
)
external
returns (ISablierV2MerkleLockupLL merkleLockupLL);

this "current" version is not actually current 😅. we are still developing on staging and it has not yet been used in any UI, nor have any integrators started using it

the real current API, is already changed since we've added the struct in the first place, meaning the UI already needs to be updated. so, this is why it doesn't make sense to not add it in the struct

v2-periphery/src/interfaces/ISablierV2MerkleStreamerFactory.sol

Lines 51 to 63 in 53e2590

	function createMerkleStreamerLL(
	address initialAdmin,
	ISablierV2LockupLinear lockupLinear,
	IERC20 asset,
	bytes32 merkleRoot,
	uint40 expiration,
	LockupLinear.Durations memory streamDurations,
	bool cancelable,
	bool transferable,
	string memory ipfsCID,
	uint256 aggregateAmount,
	uint256 recipientsCount
	)

[smol-ninja]

You are right. I think I got lost in an alternate parallel reality where we have already shipped the staging by now 🤣

I'll update the PR today.

[smol-ninja]

@andreivladbrg we can merge this one. I have removed its dependency from #282.

[andreivladbrg]

lgtm, expect for this one question