Implementing Challenge Solver in other language

[Dr-Blank]

I asked copilot to translate the challenge_solver.rs file and it came up with something. However, I am not sure how this code works; I understand that it needs to work out the nonce that was used given the prefix and the taget_hex by repeatedly hashing the "{prefix}{nonce}" string and comparing it to target.
It would be nice if you could provide some test cases that I can run to see if the implementation is correct as right now there is no way for me to verify.

[tranxuanthang]

I understand that it needs to work out the nonce that was used given the prefix and the taget_hex by repeatedly hashing the "{prefix}{nonce}" string and comparing it to target.

You understood it correctly. In short, you must find an unknown nonce so that the the sha256 hash of {prefix}{nonce} is smaller than the target.

The current target's hex:

000000FF00000000000000000000000000000000000000000000000000000000

Is present in an u8 array as:

[0, 0, 0, 255, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]

For example, the following prefix:

kyhYEjsjkxcxHRliu3oOLQUodVumo6ur

Then the first nonce (from zero) that is valid is:

42660891

Because the sha256 of kyhYEjsjkxcxHRliu3oOLQUodVumo6ur42660891 is:

000000C17F529D77D963888A368D9C13B5F0A9FC89E4BCBB5B213DF3AE4CA660

Or, in u8 array:

[0, 0, 0, 193, 127, 82, 157, 119, 217, 99, 136, 138, 54, 141, 156, 19, 181, 240, 169, 252, 137, 228, 188, 187, 91, 33, 61, 243, 174, 76, 166, 96]

Is smaller than the current target.

So you got the Publish Token with the value kyhYEjsjkxcxHRliu3oOLQUodVumo6ur:42660891

For high-level languages like Python, you may just hash it to a hex string and compare the 2 strings directly:

>>> '000000FF00000000000000000000000000000000000000000000000000000000' > '000000C17F529D77D963888A368D9C13B5F0A9FC89E4BCBB5B213DF3AE4CA660'
True

[Dr-Blank]

I just realized that the target string is the same for all challenges, that answers my question that the prefixes are random strings.

[tranxuanthang]

making sure target starts with zeros would require your server to do the work as well

Target is predefined and rarely being changed. The client has to do the heavy work: brute force to find the valid nonce number. The server only needs to verify it against the predefined target.

the target string is the same for all challenges

One note is that target might be changed in the future, so it's best that you use the target string in the /api/request-challenge response when solving challenge instead of hard coding it.

[Dr-Blank]

Thank you for all the help! I learnt something about cryptography while making a wrapper for this. I also learnt how slow python is 🥹. Your rust implementation takes seconds to compute the proof of work will my implementation takes a good 30 seconds even with multiple threads and processes.

[tranxuanthang]

You may implement the performance-critical part of your library using another low-level language, such as C, C++ or Rust. I cannot help you more as my Python knowledge is limited, but something like pyo3 is a good starting point, or maybe wasmer-python for easy cross-platform support while still fast enough.

[Dr-Blank]

Yes, Having it in level language would help. Thank you for the guidance 🙌