merkle-py

Python implementation of Merkle tree

This library defines my special implementation in Python of the notorious Merkle trees. Feel free to use it (with the appropriate credits).

Other implementations include: Go, Scala and TypeScript.

Usage

$ pip install merkle-py

Here are some simple examples of how it works:

from merklepy import MerkleTree, MerkleTreeOptions, build_hash_function, SHA_256

options1 = MerkleTreeOptions(doubleHash= True, engine='sha-256', sort=True)
tree1 = MerkleTree(options1)

# Build a tree from the raw data
proofs1 = tree1.add_leaves(True, '1', '2', '3')
root_hash = tree1.get_root_hash()
assert(tree1.depth() == 1)

json = tree1.to_json()

# Build another identical tree from the JSON of the first one
tree2 = tree_from(json)
assert(tree1.size() == tree2.size())
sha256 = build_hash_function(SHA_256)
assert(tree2.size() == proofs1[0].size)
assert(tree2.validate_proof(proofs1[0], sha256('1'), root_hash) == True)

# Enrich with new hashed data
proofs2 = tree2.add_leaves(False,
    bytes.fromhex('1234567890abcdef1234567890abcdef1234567890abcdef1234567890abcdef'),
    bytes.fromhex('abcdef0123456789abcdef0123456789abcdef0123456789abcdef0123456789')
)
assert(tree2.size() == 5)
assert(tree2.depth() == 2)

# Because the size of the tree has changed, and so has the root hash
assert(proofs1[0].to_string() != proofs2[0].to_string() and tree2.validate_proof(proofs1[0], sha256('1'), root_hash) == False)

Important note

As you can see from the examples above, for a continuously growing Merkle tree, proofs may not work at all time. You may need either a new proof from the latest tree, or rebuild the old tree, hence the size attribute passed within the MerkleProof instance. If you don't use a sorted tree and keep a record of the leaves' hashes in the order they were included in the tree, this allows you to rebuild the corresponding tree and therefore use any proof at any time.
In other words, this implementation is either not made for a growing tree, or should take this behaviour into account when issuing and verifying proofs.

Tests

$ git clone https://github.com/cyrildever/merkle-trees.git
$ cd merkle-trees/packages/py/
$ pip install -e .
$ python3 -m unittest discover

License

This library is distributed under a MIT license.
See the LICENSE file.