Hi Denis, Well seen! This is a small typo but it highlighted several issues in my update of existing tests. The OpenSSL engine was identified as GnuTLS and it caused MHD to allow calls to GnuTLS-specific options that should be rejected when using OpenSSL. As I forgot to change a few occurences of these options in existing tests, these tests failed after fixing the typo.
I fixed all of them and they pass again. Thanks again, Gauthier > Hi Gauthier, > > Nice work! I think this would be a great addition to libmicrohttpd. From > an embedded perspective it would be great to be able to link to openssl or > perhaps wolfssl (https://www.wolfssl.com/products/wolfssl/) to have a > single SSL implementation on the platform. I like the way the TLS API > cleanly defines the features that MHD needs from an SSL library. > > Found a small typo: > diff --git a/src/microhttpd/tls_openssl.c b/src/microhttpd/tls_openssl.c > index fee4917..67d9162 100644 > --- a/src/microhttpd/tls_openssl.c > +++ b/src/microhttpd/tls_openssl.c > @@ -906,8 +906,8 @@ MHD_TLS_openssl_session_write (struct MHD_TLS_Session > * session, > const struct MHD_TLS_Engine tls_engine_openssl = > { > - "GnuTLS", > - MHD_TLS_ENGINE_TYPE_GNUTLS, > + "OpenSSL", > + MHD_TLS_ENGINE_TYPE_OPENSSL, > MHD_TLS_openssl_has_feature, > MHD_TLS_openssl_init_context, > > > Regards, > Denis
