Carter Kozak

Results 73 issues of Carter Kozak

==COMMIT_MSG== Implement experimental NTLM proxy auth support ==COMMIT_MSG==

do not merge
autorelease

Dialogue didn't behave well in systems using classloaders which must be unloaded later. ## After this PR ==COMMIT_MSG== Dialogue attempts not to retain context classloaders ==COMMIT_MSG== ## Possible downsides? The...

autorelease

This works around a degenerate case in which closure isn't negotiated in a timely manner, and the client consumes far more bytes than desired in `SSLSocketInputRecord.deplete`. ## After this PR...

autorelease

From https://github.com/undertow-io/undertow/pull/875 1. Thread T1 calls decrementRequests. Shutdown is false, so this thread will follow the 'else' path in preparation for decrementAndGet 2. Thread T2 calls shutdown. Increments from 1...

Updates the allMethodsMatcher to lookup templates using the match method instead of get in order to normalize the input. Previously registration would throw an exception when routes were added for...

This change results in adherance to the servlet 4.0 specification in which any path match is used prior to extension matches. From https://github.com/undertow-io/undertow/pull/1038 includes both 1f36d68628b0398edc1302f714aa88364acd5e5b and fc1db053d3e3a8a5948a891f7842be7ce74b2dfb

I think using `.isTrue()` on any kind of complex assertion is probably a codesmell, so might be worth replacing stuff like this: ```diff -assertThat(result.getTestResults().stream().allMatch( - testResult -> testResult.getType().equals(RegressionTestResultType.NO_REGRESSION))).isTrue(); +assertThat(result.getTestResults()).allMatch( +...

## What happened? Several of our handlers require a small slice of what is available on InvocationHandler: * `TracingInvocationEventHandler` only uses `java.reflect.Method.getName()` for operation names * `MetricsInvocationEventHandler` uses `getStartTimeNanos()` and...

This benchmark class is equivalent to JsonStdReadVanilla but disables `CANONICALIZE_FIELD_NAMES`. I don't have much preference whether this is merged, I can see an argument that this variant is already covered...