Subtensor API Python package

A python wrapper around the @polkadot/api node library to query the bittensor chain.

Install from pypi

pip install subtensorapi==1.1.0

Running the CLI

Usage

View usage
python3 -m subtensorapi --help

Specify chain endpoint url
python3 -m subtensorapi [sync_and_save|blockAtReg_and_save|sync_and_save_historical|sync_and_save_historical_difficulty] --endpoint_url ENDPOINT_URL
Specify filename to save to python3 -m subtensorapi [sync_and_save|blockAtReg_and_save|sync_and_save_historical|sync_and_save_historical_difficulty] --filename FILENAME.json
Specify blockhash to sync the chain at, default "latest" python3 -m subtensorapi [sync_and_save|blockAtReg_and_save|sync_and_save_historical|sync_and_save_historical_difficulty] --block_hash BLOCK_HASH

Sync And Save Metagraph

Pulls the neurons storage map and saves it to a JSON file.
View usage
python3 -m subtensorapi sync_and_save --help
Run sync of metagraph and save to JSON file
python3 -m subtensorapi sync_and_save -f metagraph.json

Sync And Save Neurons from Metagraph at historical blocks

Pulls the neurons storage map at each block for each UID and saves it to a JSON file.
View usage
python3 -m subtensorapi sync_and_save_historical --help
Run sync of metagraph at blocks 2706651 2706652 2706653 and UIDs 1 2 3 and save to JSON file
python3 -m subtensorapi sync_and_save_historical -i 1 2 3 -b 2706651 2706652 2706653 -f history.json

Grab blockAtRegistration

Pulls the blockAtRegistration storage map and saves it to a JSON file.
View usage python3 -m subtensorapi blockAtReg_and_save --help

Sync And Save Difficulty at historical blocks

Pulls the difficulty storage value at each block and saves it to a JSON file.
View usage
python3 -m subtensorapi sync_and_save_historical_difficulty --help
Run sync of metagraph at blocks 2706651 2706652 2706653 and save to JSON file
python3 -m subtensorapi sync_and_save_historical_difficulty -b 2706651 2706652 2706653 -f history_diff.json

Using the library

FastSync class

Setup of a FastSync instance

from subtensorapi import FastSync

# check if fast sync is available on the platform
FastSync.verify_fast_sync_support() 

# specify the chain endpoint_url  
fast_sync: FastSync = FastSync(endpoint_url)

Run the metagraph sync

Pulls the neurons storage map.

# specify block_hash to sync at. Default is "latest"
block_hash = "0xb2fa081[...]"
# run the sync command and save to JSON file at block_hash
fast_sync.sync_and_save(block_hash)
# load neurons in from JSON file
neurons = fast_sync.load_neurons()

Run the historical metagraph sync

Pulls the neurons storage map for each block and each UID

# specify blockNumbers to sync at. Default is "latest"
blockNumbers = ["latest", 2706652]
# specify UIDs to sync
UIDs = [1, 2, 3, 4095]
# run the sync command and save to JSON file at block_hash
fast_sync.sync_and_save_historical(blockNumbers, UIDs)
# load neurons in from JSON file
historical_neurons = fast_sync.load_historical_neurons()

Run the historical difficulty sync

Pulls the difficulty storage value for each block

# specify blockNumbers to sync at. Default is "latest"
blockNumbers = ["latest", 2706652]
# run the sync command and save to JSON file at block_hash
fast_sync.sync_and_save_historical_difficulty(blockNumbers)
# load neurons in from JSON file
historical_diff = fast_sync.load_historical_difficulty()

Run blockAtRegistration pull

Pulls the blockAtRegistration storage map.

# specify block_hash to sync at. Default is "latest"
block_hash = "0xb2fa081[...]"
# run the pull command and save to JSON file at block_hash
fast_sync.get_blockAtRegistration_for_all_and_save(block_hash)
# load blockAtRegistration_all from JSON file
blockAtRegistration_all = fast_sync.load_blockAtRegistration_for_all()

Using the FastSync class without writing to a file

You can redirect the file writing to a pipe instead of a file.
This removes the need to write to disk and provides a speedup.

Example:

# all neurons
all_neurons: List[SimpleNamespace] = fast_sync.sync_fd(block_hash)
# historical neurons
historical_neurons: Dict[str, Dict[str, SimpleNamespace]] = fast_sync.sync_historical_fd(blockNumbers, UIDs)
# blockAtRegistration_all
blockAtRegistration_all: List[int] = fast_sync.get_blockAtRegistration_for_all_fd(block_hash)
# historical difficulty
historical_diff: Dict[str, int] = fast_sync.sync_historical_difficulty_fd(blockNumbers)