0

I'm trying to build a complete web caching proxy using Boost Asio and LibCURL, I've already built the server and everything works fine. It receives http requests (GET, POST, upload using POST ...) correctly and also it sends back the responses to the browser for e.g correctly.

Now, I want to extend it, so it can handles https requests. I read about it in LibCURL web site http://curl.haxx.se/libcurl/c/libcurl-tutorial.html (proxy section), I understood how it works and I have a clear idea how it should be done. But I didn't find a good documentation about how proxies handle https requests. and:

  • what are the possible messages (information, format, length ...) exchanged by the source application and the proxy ?
  • things to consider.
  • ...

Thanks in advance :-) .

Yacine Hebbal
  • 394
  • 3
  • 16

1 Answers1

1

You will receive the CONNECT command in plain text, and respond to it ditto, then the communications after that will be encrypted. If your proxy is to be an SSL endpoint, which is highly problematic given that HTTPS requires a certificate that matches the target host-address, you will then need to enter SSL mode on both connections. More probably you should just start copying bytes in both directions without attempting to process the contents.

user207421
  • 305,947
  • 44
  • 307
  • 483