The Swiss Army Knife of the Bitcoin protocol.
This Python3 library provides an easy interface to the bitcoin data structures. This is based on https://github.com/Simplexum/python-bitcointx, but is focused only on fixing bugs and adjusting for asimov blockchain.
The RPC interface,
bitcointx.rpc, is designed to work with Bitcoin Core v0.16.0.
Older versions may work but there do exist some incompatibilities.
Everything consensus critical is found in the modules under bitcointx.core. This rule is followed pretty strictly, for instance chain parameters are split into consensus critical and non-consensus-critical.
bitcointx.core - Basic core definitions, datastructures, and
bitcointx.core.key - ECC pubkeys
bitcointx.core.script - Scripts and opcodes
bitcointx.core.scripteval - Script evaluation/verification
bitcointx.core.serialize - Serialization
Note that this code may not be fully consensus-compatible with current bitcoin core codebase. Corner cases that is not relevant to creating valid bitcoin transactions is unlikely to be considered. See also note on VerifyScript usage below.
Non-consensus critical modules include the following:
bitcointx - Chain selection
bitcointx.base58 - Base58 encoding
bitcointx.rpc - Bitcoin Core RPC interface support
bitcointx.wallet - Wallet-related code, currently Bitcoin address and
private key support
Effort has been made to follow the Satoshi source relatively closely, for instance Python code and classes that duplicate the functionality of corresponding Satoshi C++ code uses the same naming conventions: CTransaction, CPubKey, nValue etc. Otherwise Python naming conventions are followed.
Mutable vs. Immutable objects
Like the Bitcoin Core codebase CTransaction is immutable and CMutableTransaction is mutable; unlike the Bitcoin Core codebase this distinction also applies to COutPoint, CTxIn, CTxOut.
Rather confusingly Bitcoin Core shows transaction and block hashes as little-endian hex rather than the big-endian the rest of the world uses for SHA256. python-bitcointx provides the convenience functions x() and lx() in bitcointx.core to convert from big-endian and little-endian hex to raw bytes to accomodate this. In addition see b2x() and b2lx() for conversion from bytes to big/little-endian hex.
Note on VerifyScript() usage
It is good to use VerifyScript to pre-screen the transaction inputs that you create, before passing the transaction to bitcoind, or for debugging purposes.
But! Bitcoin Core should always remain the authoritative source on bitcoin transaction inputs validity.
Script evaluation code of VerifyScript() is NOT in sync with Bitcoin Core code, and lacks some features. While some effort was made to make it behave closer to the code in Bitcoin Core, full compatibility is far away, and most likely will not be ever achieved.
WARNING: DO NOT rely on VerifyScript() in deciding if certain signed transaction input is valid. In some corner cases (non-standard signature encoding, unhandled script evaluation flags, etc) it may deem something invalid that bitcoind would accept as valid. More importanty, it could accept something as valid that bitcoind would deem invalid.
Module import style
While not always good style, it's often convenient for quick scripts if
import * can be used. To support that all the modules have
examples/ directory. For instance this example creates a transaction
spending a pay-to-script-hash transaction output:
$ PYTHONPATH=. examples/spend-pay-to-script-hash-txout.py
Selecting the chain to use
Do the following:
Where NAME is one of 'testnet', 'mainnet', or 'regtest'. The chain currently selected is a global variable that changes behavior everywhere, just like in the Satoshi codebase.
To use alternative chain parameters:
examples/litecoin-alt-p2sh-prefix.py for an example usage.
Under bitcointx/tests using test data from Bitcoin Core. To run them:
python -m unittest discover && python3 -m unittest discover
Alternately, if Tox (see https://tox.readthedocs.org/) is available on your system, you can run unit tests for multiple Python versions:
Currently, the following implementations are tried (any not installed are skipped):
* CPython 3.4
* CPython 3.5
HTML coverage reports can then be found in the htmlcov/ subdirectory.
Sphinx documentation is in the "doc" subdirectory. Run "make help" from there to see how to build. You will need the Python "sphinx" package installed.
Currently this is just API documentation generated from the code and docstrings. Higher level written docs would be useful, perhaps starting with much of this README. Pages are written in reStructuredText and linked from index.rst.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for asimov_lib-0.10.4.dev0-py3-none-any.whl