- A.py my implementation (it could be wrong its just my understanding) https://github.com/LucaPaterlini/merkletreeImplementations
- B.py taken from the merkletools pip package https://github.com/Tierion/pymerkletools/issues
- C.py Evan Kozliner implentation https://github.com/evankozliner/merkle-tree
- D.py tutorial example https://github.com/JaeDukSeo/Simple-Merkle-Tree-in-Python
- E.py lib package merkletree https://github.com/jvsteiner/merkletree
- F.py lib package pymerkle(python 3) https://github.com/FoteinosMearg/pymerkle
N.B. none of those different implentations returns the same output on the same input
Assuming we have to be able to redo the same operation in order to allways get the same root corresponding to the same array in input with the element in the same order there are no more then 1 valid solution to be a merkle tree that can tested for each hashing function. Then, witch of those is the right one?
As well if we consider the standard the bitcoin implentation, its written by its author that it have a potential flow due to the wrong handling of the odd lists https://github.com/bitcoin/bitcoin/blob/bccb4d29a8080bf1ecda1fc235415a11d903a680/src/consensus/merkle.cpp
A1.py shows this inmplentation instead
Isn't it better to avoid the Second preimage attack as proposed on wiki https://en.wikipedia.org/wiki/Merkle_tree by adding a byte at the begin of each level of hashing to certify the dept of the structure?(this way the implicit limit of the list is 2**(2**8) = 115792089237316195423570985008687907853269984665640564039457584007913129639936L (still a big number)
P.S. I am in the comptia (https://www.comptia.org/) council regarding blokchain and my first tought was to currectly address and define what a merkle tree is and how to calculate its root
#Answers
Feel free to open inssues with your answers on this github repo. Tk in advance for your feedback, LP