Bring in the testrunner/http server scaffold for better integration testing.

This brings in a copy of llmessage's llsdmessage testing server.  We run
a mocked HTTP service to handle requests and the integration tests run
against it by picking up the LL_TEST_PORT environment variable when running.
Add some checks and output to produce useful info when run in the wrong
environment and when bad status is received.  Later will add a dead port
as well so we can test that rather than use 'localhost:2'.

1 files changed, 36 insertions, 0 deletions

diff --git a/indra/llcorehttp/tests/llcorehttp_test.cpp b/indra/llcorehttp/tests/llcorehttp_test.cpp
index 2d48bca443..f59361ab53 100644
--- a/indra/llcorehttp/tests/llcorehttp_test.cpp
+++ b/indra/llcorehttp/tests/llcorehttp_test.cpp
@@ -27,6 +27,7 @@
 #include "llcorehttp_test.h"
 
 #include <iostream>
+#include <sstream>
 
 // These are not the right way in viewer for some reason:
 // #include <tut/tut.hpp>
@@ -130,3 +131,38 @@ void ssl_locking_callback(int mode, int type, const char * /* file */, int /* li
 }
 
 
+std::string get_base_url()
+{
+	const char * env(getenv("LL_TEST_PORT"));
+
+	if (! env)
+	{
+		std::cerr << "LL_TEST_PORT environment variable missing." << std::endl;
+		std::cerr << "Test expects to run in test_llcorehttp_peer.py script." << std::endl;
+		tut::ensure("LL_TEST_PORT set in environment", NULL != env);
+	}
+
+	int port(atoi(env));
+	std::ostringstream out;
+	out << "http://localhost:" << port << "/";
+	return out.str();
+}
+
+
+void stop_thread(LLCore::HttpRequest * req)
+{
+	if (req)
+	{
+		req->requestStopThread(NULL);
+	
+		int count = 0;
+		int limit = 10;
+		while (count++ < limit && ! HttpService::isStopped())
+		{
+			req->update(1000);
+			usleep(100000);
+		}
+	}
+}
+
+