The test client can now be used against a server that has properly
configured certificates, instead of just the test server. To reach the
test server, the client needs these arguments:
--use_test_ca=true --server_host_override=foo.test.google.fr
Client no longer has any required arguments, although for any given
setup needing to specify at least one argument is highly likely.
The arguments have been improved in general to hopefully be more
orthogonal and match those of other language's test clients.
This change loses asynchronous notification of channel state-change and
a way to wait until the channel is actually connected. Both of these are
expected to be added back as part of a health API. The important
distinction from Service is that ChannelImpl never permanently fails and
can revert from being started to connecting again.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=83875407
No major refactorings/simplifications were done. Only gRPC v1 support
infrastructure was removed.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=82737436
The test was going to use a queue like ping_pong, but using a mock
proved much simpler. Thus, I also updated ping_pong to use the simpler
model, because it is useful for the two tests to be similar.
InProcessTransportTest failed for empty_stream due to gRPC v2 issues,
and so instead of ignoring emptyStream() I found the broken tests that
were preventing swapping to gRPC v2 and ignored them instead.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=82733943
mvn exec:exec doesn't kill the child process when Maven is killed. Now
we just run ourselves, but have Maven do as much heavy lifting as able.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=81736090
Any place that force-sets the protocol to 2 or assumes the old value
is now removed. Unfortunately, it seems InProcessTransportTest has
some non-obvious dependency on gRPC v1.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=81597524
Previously, it was creating a client socket to determine when the server
was bound to the port. This causes the Netty server to create and then
quickly fail a client channel which was generating noisy error logs.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=80546045
The client now checks certificates and performs hostname verification.
Tests check certificates, so the server uses a cert that the client
trusts.
Only the client portion of SslContextFactory was previously used.
Applications that want to ignore certificates (i.e., for testing) can
use io.netty.handler.ssl.util.InsecureTrustManagerFactory instead.
The MOE configuration was already failing to work, and so required the
simple mapping for examples in addition to what was needed for the new
certs.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=80434148
The current signatures are not more useful than the async counterparts,
we'd rather not have them until more sepcific requirement comes up.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=80402647
GrpcServer had the same restriction that GrpcClient did: it had a
single enum of all transports that prevents using it in third_party.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=79428440
Our pre-existing tests were slightly different from the integration
tests to be implemented in each language. Change our implementation to
match.
Unfortunately, the standard tests actually trigger Netty flow control
bugs, so we decrease stream sizes to <64K as a workaround.
Because the tests now compare to golden responses, they detected that
the cpp server was sending all a's instead of all 0s. That is now fixed.
We receive errors hitting CPP server for empty_unary and the various
streaming tests, so java_cpp isn't enabled in interop_scenarios.txt.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=79086882
This required making it no longer depend on GrpcClient. Instead, we now
use the builders that make using GrpcClient almost completely obsolete.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=78860648
The TestService proto is temporarily supplied as a generated jar (until the open source protoc compiler supports grpc).
Copies of messages.proto, empty.proto, and message_set.proto are scrubbed and included in the source under integration-testing.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=78711468
Previously we were using English text, but this just adds extra effort for all
the other languages that will need to implement servers. The integration test
case descriptions uses zero-filled bytes.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=78477356
overview:
1) Lots of @SuppressWarnings :)
2) Remove dependencies on StandardCharsets.XXX (which is Java 7)
3) Moved testing/utils/ssl/* to .../stubby/util so that the netty transport doesn't depend directly on the testing module.
-------------
Created by MOE: http://code.google.com/p/moe-java
MOE_MIGRATED_REVID=78460727
Eric Anderson
simonma
yangg
rocking
nathanmittler
lryan
zhangkun