Hashlib

Suppose we have a file and want a unique value that represents the contents of that file. With hashlib, we can use hash algorithms like SHA256 and MD5 to get hashes.

For hashlib, we are trying to get unique hashes, and these may be slow to compute. Collections like the dictionary use different hashes that are faster to compute.

Example

This Python example program first creates a binary file called "example.txt" that contains some data. It then tries to hash this file's contents.

Version 1 We can compute a hash by creating the required hash object, calling update() with bytes, and then calling hexdigest().

Version 2 If we read in a file as binary (with the "b" flag) we can pass its data to the update() method as well.

Version 3 With file_digest, we have a more efficient way to compute a hash for a file object.

Version 4 If we want to use another hash algorithm like BLAKE2, that is possible by passing in a "blake2s" string to file_digest.

Version 5 The MD5 hash algorithm is not secure, but it is still sometimes used, and we can compute MD5 hashes with file_digest as well.

import hashlib

# Create the needed example file.
with open("example.txt", "wb") as f:
    f.write(b"Some example text")

# Version 1: get sha256 hash from bytes.
m = hashlib.sha256()
m.update(b"Some example text")
print(m.hexdigest())

# Version 2: get sha256 hash from file using update.
with open("example.txt", "rb") as f:
    m2 = hashlib.sha256()
    data = f.read()
    m2.update(data)
    print(m2.hexdigest())

# Version 3: get sha256 hash from file with file_digest.
with open("example.txt", "rb") as f:
    m3 = hashlib.file_digest(f, "sha256")
    print(m3.hexdigest())

# Version 4: use blake2s hash.
with open("example.txt", "rb") as f:
    m4 = hashlib.file_digest(f, "blake2s")
    print(m4.hexdigest())

# Version 5: use md5 hash.
with open("example.txt", "rb") as f:
    m5 = hashlib.file_digest(f, "md5")
    print(m5.hexdigest())ba94b1c49abbd67b58019d6295f070913f499a774173c1b951b28525b0fb7193
ba94b1c49abbd67b58019d6295f070913f499a774173c1b951b28525b0fb7193
ba94b1c49abbd67b58019d6295f070913f499a774173c1b951b28525b0fb7193
cd154ac1356589b52ee8ad35899e36129bd62dcbf741dc7c08748fd10900e555
5b3a9b7b92bd8217bf5ffbd301043cea

Hashing the contents of files lets us compute 2 files for equality without knowing their contents. This can be a performance optimization, or help with ensuring data has not been tampered with.