@upala/group-manager v0.1.1

Group manager lib

The lib behind group manager CLI.

Draft Developer notes

All individual scores are published to the DB and can be accesses by DApps. DB will also have a web UX for group managers to view scores. Base scores and score bundle hashes are stored on-chain. The goal of this iteration is to let future managers to play around with the protocol and understand better their needs.

Interaction and references:

• Uses deployments for contracts ABIs and addresses.
• Queries baseScore and bundleHash from Upala subgraph.
• Reads and writes individual scores to DB.
• Manages groups on-chain through Upala groups.
• Uses metadata scheme from Upala groups.

Merkle Tree Pool (future)

Merkle pool generates merkle tree from UpalaID-score pairs. The root of this three is then stored on-chain as score bundle hash. Users provide merkle proof to verify their scores. The procedure of updating the scores requires three steps. Merkle tree pool will be implemented after Signed Scores Pool. Can be skiped untill then.

Announcements

Changes that require commit-reveal scheme. All commits are made through the same commitHash(bytes32 hash) function. The way hash is calculated is what differs for different changes.

todo desribe attackWindow, executionWindow.

commitBaseScore(newScore, optional secret)

TBD

• hash = keccak256(abi.encodePacked("setBaseScore", newScore, secret)) - calculate commitment hash
• call commitHash(bytes32 hash) function of the group contract
• save timestamp

set-base-score

Increasing base score doesn't require commit-reveal scheme and works for any pool-type. Group manager can increase baseScore at any time. Decreasing score requires a commit. The workflow for the command is as follows.

• querry graph node for the current baseScore (Graph "what is current base score for the group address?")
• question: "Your current base score is %baseScore% what is the new value?:"
• if new value is higher, call increaseBaseScore(uint newBotReward) function of the pool.
• if score is lower, try to retrieve latest commit for setBaseScore from local storage (todo). if nothing found "Error: post a commit before decreasing the score"
• if there is a commit, call setBaseScore(uint botReward, bytes32 secret) function of the pool

commitRootDeletion()

TBD hash = keccak256(abi.encodePacked("deleteRoot", newRoot, secret)) sends transaction: commitHash(bytes32 hash)

deleteRoot(bytes32 root, bytes32 secret)

TBD send transaction: deleteRoot(bytes32 root, bytes32 secret)

commitWithdrawal(newScore, recipient, amount, optional secret)

TBD

• hash = keccak256(abi.encodePacked("withdrawFromPool", recipient, amount, secret))
• send transaction: commitHash(bytes32 hash)

withdrawFromPool(address recipient, uint amount, bytes32 secret)

TBD send transaction: withdrawFromPool(address recipient, uint amount, bytes32 secret)

temp

Commits: firstArgument (commitType), timestamp, other arguments, secret score bundle hashes (all published hashes) for merkle - current unpublished tree and root

Uniswap Merkle Distributor notes

The tools use Uniswap Merkle Distributor lib. This is their README.