core.crypto.token ================= .. py:module:: core.crypto.token Attributes ---------- .. autoapisummary:: core.crypto.token.RANDOM_TOKEN_LENGTH Functions --------- .. autoapisummary:: core.crypto.token.random_token core.crypto.token.stored_random_token Module Contents --------------- .. py:data:: RANDOM_TOKEN_LENGTH :value: 64 .. py:function:: random_token(nbytes: int = 512) -> str Generates an unguessable token. Generates a random string with the given number of bytes (may not be lower than 512) and hashes the result to get a token with a consistent length of 64. Why hashing? We could of course just create a random token with a length of 64, but that would leak the random numbers we actually create. This can be a bit of a problem if the random generator you use turns out to have some vulnerability. By hashing a larger number we hide the result of our random generator. Doesn't generating a hash from a larger number limit the number of tokens? Yes it does. The number of different tokens is 2^256 after hashing, which is a number larger than all the atoms on earth (approx. 2^166). So there is a chance of a collision occuring, but it is *very* unlikely to *ever* happen. More information: ``_ ``_ ``_ .. py:function:: stored_random_token(namespace: str, name: str) -> str A random token that is only created once per boot of the host (assuming the host deletes all files in the /tmp folder). This method should only be used for development and is not meant for general use!