On Tue, Aug 30, 2011 at 05:54:42PM -0400, Nick Mathewson wrote: > Neat! APIwise, is there also a way to just pass in an SSL_CTX or use > a factory function to create SSL objects? This kind of "config > structure" trick is neat, but a lot of time, I need to do some pretty > heavy unanticipated SSL hackery. >
Trivial change; In this case I was laying out the structure - but I will take this into account. > Also, this is a yet another good sign that Mark's "libevhtp" might be > more maintainable going forward than our current evhttp.c http > implementation. People should definitely try it out, look at the > source code, etc: once it's feature-complete and passes all of > Libevent's current http unit tests, I'd like to start talking about > whether and how we can eventually migrate towards it as a drop-in > seamless source-compatible evhttp replacement, and away from having to > maintain the current evhttp.c code indefinitely. If this is a bad > idea for some reason, we should figure that out soon. > On this note, a few people have already contributed and I thank you. One of the big things I intend to do next is to get client support built, and if anyone wants to help out with that I will buy you a drink of your choice. *********************************************************************** To unsubscribe, send an e-mail to majord...@freehaven.net with unsubscribe libevent-users in the body.
