openrank-sdk

Preparing environment

datasets repo can be used to get up to speed, since it contains some example datasets, like Degen Tipping DB or OSO GitHub DB. It contains the script for installing the SDK, as well as an example config.toml needed to run openrank-sdk methods. To clone the datasets repo:

git clone https://github.com/openrankprotocol/datasets.git
cd ./datasets

Then run ./install-sdk.sh script:

./install-sdk.sh

The above command will install the cargo (Rust package manager) and openrank-sdk binary. It will also generate a new keypair that will be used for signing messages using generate-keypair command.

The generated secret key should be added to local .env file that will be used by OpenRankSDK. So, create .env file, and copy-paste the output from the generate-keypair command:

SECRET_KEY="b0f6d4b7865e1128eebfe4eb37b96522d2e58cbd7892c7e0759907c5f4c6ede4"
# ADDRESS: b79aafc95c8866e65ed51a7856e75587feb481ff

If you wish for you address to be whitelisted, send us a request at devs@karma3labs.com.

Datasets

If you wish to create new datasets to run the compute on, they should be prepared in the following format, e.g.:

• For Seed trust:

i,v
1,0.1
2,0.5
3,0.4

• For local trust

i,j,v
1,2,30
1,3,40
2,1,10
2,3,50
3,1,5
3,2,25

Usage:

TrustUpdate - Updating a bulk of Trust scores to a specific namespace. It will expect a csv file with i,j,v entries where i and j are string with arbitrary values, and v is integer value:

openrank-sdk trust-update [TRUST_DB_FILE_PATH] [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

SeedUpdate - Updating a bulk of Seed scores to a specific namespace. It will expect a csv file with i,v entries where i is a string with arbitrary value and v is integer value:

openrank-sdk seed-update [SEED_DB_FILE_PATH] [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

ComputeRequest - Request a compute in a domain specified inside OPENRANK_SDK_CONFIG_PATH file. The hash of ComputeRequest TX will be returned:

openrank-sdk compute-request [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

Get Results - Get results of a specific compute request identified by it's TX hash:

openrank-sdk get-results [ComputeRequest_TX_HASH] [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

Get Results and Check Integrity - Get the results of a specific compute request, and perform the convergence check of top X amount of scores. The scores will be compared against a predefined test vector with TEST_VECTOR_PATH path:

openrank-sdk get-results-and-check-integrity [ComputeRequest_TX_HASH] [OPENRANK_SDK_CONFIG_PATH] [TEST_VECTOR_PATH]

Get Compute Results - Get ComputeResult object given its identifier:

openrank-sdk get_compute_result [ComputeRequest_TX_HASH] [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

Get Compute Results TXs - Get TXs contained in ComputeResult object given:

openrank-sdk get_compute_result-txs [ComputeRequest_TX_HASH] [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

Get TX - Get arbitrary TX object given its kind and hash:

openrank-sdk get_tx [!kind]:[TX_HASH] [OPENRANK_SDK_CONFIG_PATH] [*OUTPUT_PATH]

! - kind should be in format: "compute_request", "compute_commitment", etc. E.g. for how to fetch ComputeRequest TX:

openrank-sdk get_tx compute_request:3d967111e0e244f62822f64d914ac3c032db85b2284ebc8f5a8bb4fd1273ff74 ./config.toml ./out.json

Where:

TRUST_DB_FILE_PATH = Path to Trust DB file, a csv file with i,j,v header

SEED_DB_FILE_PATH = Path to Seed DB file, a csv file with i,v header

OUTPUT_PATH = Path to of the file to write result of the JsonRPC call

ComputeRequest_TX_HASH = TX Hash of ComputeRequest TX that was submitted to the protocol

TX_HASH = Hash of the TX being requested

kind = Kind of the TX: "trust_update", "seed_update", "compute_request", "compute_assignment", "compute_commitment", "compute_verification".

TEST_VECTOR_PATH = Path to a file that will contain a vector of scores in csv format (i,v entries), and will be used for comparing with the scores resulted from the compute

Parameters marked with * are optional