On May 20, 2017 11:02 AM, "Oleg Kalnichevski" <[email protected]> wrote:
On Sat, 2017-05-20 at 10:52 -0700, Gary Gregory wrote: > On Sat, May 20, 2017 at 10:12 AM, Oleg Kalnichevski <[email protected] > > > wrote: > > > On Sat, 2017-05-20 at 09:52 -0700, Gary Gregory wrote: > > > OK, cool, my next issue (in the same method) is that the verifier > > > is > > > NOT > > > called for all requests, which is a bug for my use case. I can > > > see > > > that it > > > is only called for HttpEntityEnclosingRequest so I cannot verify > > > all > > > requests coming in! > > > > > > > Expect-continue handshake applies to entity enclosing methods only. > > That is why I think changes you have made to HttpClient 4.6 is a > > mistake. > > > > Hi Oleg, > > OK, I see, so that change does not site well with 100 (Continue) > Status. > That should be back on the table then. > > But here, I'm not talking about 100 (Continue) Status, I'm talking > about > using HttpAsyncExpectationVerifier for the general case of > "verifying" > anything that comes in, not just the 100 use case. > > It's not clear to me based on the Javadoc for > HttpAsyncExpectationVerifier > that it exists solely to implement 100 (Continue) Status behavior. > > May you please clarify? > Gary, HttpAsyncExpectationVerifier was designed to solely implement the expect-continue handshake and thus it only applies to the entity closing methods. I looked at the javadocs. Yes, they do not state that clearly enough. Hi Oleg, Thank you for the clarification. What do you think about providing a similar hook for general purpose validation of incoming requests? Gary Oleg --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
