NSA-LGDev2
commented
5 years ago Probably a good idea. In the interim, things I'm looking at right now include:
- multi-threading the sync process
- abandoning prctl
In the future, I may be able to support http 1.1 pipelining. Pushing more down into C is a possibility.
Would help to have more understanding on roadmap, particularly what few features/contributions are top-of-mind and what is well thought-out thus not worth reinventing.
I'm personally looking towards a complete C API, then bindings for node.js & beyond would be in reach.