PyCardano is a Cardano library written in Python. It allows users to create and sign transactions without depending on third-party Cardano serialization tools, such as cardano-cli and cardano-serialization-lib, making it a lightweight library, which is simple and fast to set up in all types of environments.
Current goal of this project is to enable developers to write off-chain code and tests in pure Python for Plutus DApps. Nevertheless, we see the potential in expanding this project to a full Cardano node client, which could be beneficial for faster R&D iterations.
- Shelly address
- Transaction builder
- Transaction signing
- Multi-asset
- Chain backend integration
- Fee calculation
- UTxO selection
- Native script
- Native token
- Metadata
- Plutus script
- Mnemonic
- Byron Address
- Reward withdraw
- HD Wallet
- Staking certificates
- Protocol proposal update
Install the library using pip:
pip install pycardano
https://pycardano.readthedocs.io/en/latest/
A full stack testnet DApp is hosted on replit: https://pycardano.cffls.repl.co/
To learn more details, go to the DApp page.
"""Build a transaction using transaction builder"""
from pycardano import *
# Use testnet
network = Network.TESTNET
# Read keys to memory
# Assume there is a payment.skey file sitting in current directory
psk = PaymentSigningKey.load("payment.skey")
# Assume there is a stake.skey file sitting in current directory
ssk = StakeSigningKey.load("stake.skey")
pvk = PaymentVerificationKey.from_signing_key(psk)
svk = StakeVerificationKey.from_signing_key(ssk)
# Derive an address from payment verification key and stake verification key
address = Address(pvk.hash(), svk.hash(), network)
# Create a BlockFrost chain context
context = BlockFrostChainContext("your_blockfrost_project_id", network)
# Create a transaction builder
builder = TransactionBuilder(context)
# Tell the builder that transaction input will come from a specific address, assuming that there are some ADA and native
# assets sitting at this address. "add_input_address" could be called multiple times with different address.
builder.add_input_address(address)
# Get all UTxOs currently sitting at this address
utxos = context.utxos(str(address))
# We can also tell the builder to include a specific UTxO in the transaction.
# Similarly, "add_input" could be called multiple times.
builder.add_input(utxos[0])
# Send 1.5 ADA and a native asset (CHOC) in quantity of 2000 to an address.
builder.add_output(
TransactionOutput(
Address.from_primitive("addr_test1vrm9x2zsux7va6w892g38tvchnzahvcd9tykqf3ygnmwtaqyfg52x"),
Value.from_primitive(
[1500000,
{
bytes.fromhex("57fca08abbaddee36da742a839f7d83a7e1d2419f1507fcbf3916522"): # Policy ID
{
b'CHOC': 2000
}
}]
)
)
)
# We can add multiple outputs, similar to what we can do with inputs.
# Send 2 ADA and a native asset (CHOC) in quantity of 200 to ourselves
builder.add_output(
TransactionOutput(
address,
Value.from_primitive(
[2000000,
{
bytes.fromhex("57fca08abbaddee36da742a839f7d83a7e1d2419f1507fcbf3916522"): # Policy ID
{
b'CHOC': 200
}
}]
)
)
)
# Build a finalized transaction body with the change returning to the address we own
tx_body = builder.build(change_address=address)
# Sign the transaction body hash using the payment signing key
signature = psk.sign(tx_body.hash())
# Add verification key and the signature to the witness set
vk_witnesses = [VerificationKeyWitness(pvk, signature)]
# Create final signed transaction
signed_tx = Transaction(tx_body, TransactionWitnessSet(vkey_witnesses=vk_witnesses))
# Submit signed transaction to the network
context.submit_tx(signed_tx.to_cbor())See more usages under examples.
Clone the repository:
git clone https://github.com/cffls/pycardano.git
PyCardano uses poetry to manage its dependencies. Install poetry for osx / linux / bashonwindows:
curl -sSL https://raw.githubusercontent.com/python-poetry/poetry/master/get-poetry.py | python -
Go to poetry installation for more details.
Change directory into the repo, install all dependencies using poetry, and you are all set!
cd pycardano && poetry install
When testing or running any program, it is recommended to enter
a poetry shell in which all python dependencies are automatically
configured: poetry shell.
PyCardano uses pytest for unit testing.
Run all tests:
make test
Run all tests in a specific test file:
poetry run pytest test/pycardano/test_transaction.py
Run a specific test function:
poetry run pytest -k "test_transaction_body"
Run a specific test function in a test file:
poetry run pytest test/pycardano/test_transaction.py -k "test_transaction_body"
We use Coverage to calculate the test coverage.
Test coverage could be generated by: make cov
A html report could be generated and opened in browser by: make cov-html
The package uses Google style docstring.
Code could be formatted with command: make format
The code style could be checked by flake8: make qa
The majority of package documentation is created by the docstrings in python files. We use sphinx with Read the Docs theme to generate the html pages.
Build docs and open the docs in browser:
make docs