You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently SSH.NET only supports MD5, with the big-ish struggles right not being that FingerPrint being just a bite array, so something we can't easily add new algorithm support to.
We're currently calculating our own fingerprint for SHA256 support, but ideally having the library support this out of the box would be best.
To prevent breaking backwards comparability we can:
Make additional properties per algorithm (not the best, easy though, accessible, matches current "convention")
Make a new property that allows us to look up by algorithm, all algorithms need to be checked up-front and stored, but we can add more
Make a method that takes an algorithm that we'll return the thumbprint for, internal implementation may cache this
What probably makes the most sense? Or does everyone just roll their own and it's not really a big deal?
The text was updated successfully, but these errors were encountered:
Note that you can already compute the SHA256 fingerprint as of SSH.NET 2020.0.2. The fingerprint is nothing more than a hash of the host key. Here's how you can compute it:
The pull request I submitted merely addresses discoverability (thanks to autocompletion) and standardization (hex format for MD5 and unpadded base64 format for SHA256).
Currently SSH.NET only supports MD5, with the big-ish struggles right not being that
FingerPrint
being just a bite array, so something we can't easily add new algorithm support to.We're currently calculating our own fingerprint for SHA256 support, but ideally having the library support this out of the box would be best.
To prevent breaking backwards comparability we can:
What probably makes the most sense? Or does everyone just roll their own and it's not really a big deal?
The text was updated successfully, but these errors were encountered: