Skip to content

v0.1

Pre-release
Pre-release

Choose a tag to compare

View all tags
@AdomasMal AdomasMal released this 05 Oct 08:36
v0.1
dc11b36
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Verified
Learn about vigilant mode.

Pseudocode

Hash (string s)

        h1 ← 0

        pow ← 257

        for i ← 0 to s length do

                h1 ← (h1 + s[i] ASCII code * pow) % (1e9 + 7)

                pow ← (pow * 257) % (1e9 + 7)

        h2 ← 0

        pow ← 263

        for i ← 0 to s length do

                h2 ← (h2 + s[i] ASCII code * pow) % (1e9 + 9)

                pow ← (pow * 263) % (1e9 + 9)

        hash ← ""

        random number generator 1 seed ← h1

        random number generator 2 seed ← h2

        for i ← 0 to 32 do

                hash ← hash + random number generator 1 number from 0 to f

                hash ← hash + random number generator 2 number from 0 to f

        return hash

Testing

Input and outup size

Input of any size can be hashed and output is always a 256 bit number in hexadecimal format:

  • Hash of a 1 character string:
    40344f3f1c0ea114cef5ace64671d745b842e43bc0eff51bd63a0b13d98dfd8e
  • Hash of a 10000 chacarter string:
    a16fa28de23ff7d4ea9648a984e380fa69c6c0d0c5ffc197e1b4aa3a4cb9bef1

Hash output is deterministic

For instance hashing an empty file will always return
8899bbdd99dd88dd6699aa667744ee00ff446677cccc88779966eedd115511aa

Hashing performance

Hashing times rise somewhat linearly.
Tested by hashing different number of lines from constitution of Lithuania:
lineHashing

Collision resistance

0 collisions found by testing 100000 pairs of random text strings - 25000 of 10, 100, 500 and 1000 characters long.

Avalanche effect

100000 pairs of string with 1 character difference between them - 25000 of 10, 100, 500 and 1000 characters long where tested:

  • Bitwise difference:
    • Average: 50.01%
    • Minimum: 36.72%
    • Maximum: 63.28%
  • Hexwise difference:
    • Average: 93.77%
    • Minimum: 78.13%
    • Maximum: 100%

Conclusion

By this limited testing it can be concluded that:

  • The hashing function can process inputs of variable length and output a 256 bit number in hexadecimal form.
  • Hashing times rise somewhat linearly while increasing input size.
  • The hashing function is somewhat collision resistant as no collisions were found while testing 100000 pairs of random strings.
  • While hashing function's hexwise difference while hashing random strings with 1 character difference is high, bitwise difference is a bit low as difference equal or above 50% is preferred.
Assets 2
Loading