secretstream support? about enacl HOT 5 CLOSED

This wouldn't be too hard.

The interface in general would follow that of crypto where you initialize a context/state on the C NIF side and use an NIF resource to track that state along different calls. Once you have this, the API should be somewhat easy to stamp out.

Testing the API is equally easy if you generate messages in a list and process them.

It is a new kind of API though, so some work needs to be done in order to pull it off.

from enacl.

Update on this one:

I'm working toward a cleaner repository on the C side of things. This significantly simplifies the implementation of this issue. I'm probably going to slipstream it in once the dust settles from the cleanup.

from enacl.

This should now be possible to add. The general implementation follows that of either multi-part signatures or generic hashing.

from enacl.

I implemented it in #44

from enacl.

This has now been merged. Is slated for next release, once we have some EQC tests for the new API.

from enacl.