diff --git a/launch_testing/launch_testing/asserts/assert_output.py b/launch_testing/launch_testing/asserts/assert_output.py
index 09bcc8cc9..5b9192ddb 100644
--- a/launch_testing/launch_testing/asserts/assert_output.py
+++ b/launch_testing/launch_testing/asserts/assert_output.py
@@ -45,7 +45,7 @@ def _match(expected, actual):
         return lambda expected, actual: expected in actual
     elif hasattr(expected_output, 'search'):
         return lambda expected, actual: (
-            expected.match(actual.replace(os.linesep, '\n')) is not None
+            expected.search(actual.replace(os.linesep, '\n')) is not None
         )
     raise ValueError('Unknown format for expected output')
 
diff --git a/launch_testing/launch_testing/launch_test.py b/launch_testing/launch_testing/launch_test.py
index 9ac929474..0557fd8df 100644
--- a/launch_testing/launch_testing/launch_test.py
+++ b/launch_testing/launch_testing/launch_test.py
@@ -34,59 +34,44 @@ def _load_python_file_as_module(test_module_name, python_file_path):
     return loader.load_module()
 
 
-def main():
-
-    logging.basicConfig()
-
-    parser = argparse.ArgumentParser(
-        description='Launch integration testing tool'
+def add_arguments(parser):
+    """Add arguments to the CLI parser."""
+    parser.add_argument('launch_test_file', help='Path to the launch test.')
+    parser.add_argument(
+        '--package-name', action='store', default=None,
+        help='Name of the package the test is in. Useful to aggregate xUnit reports.'
+    )
+    parser.add_argument(
+        '-v', '--verbose', action='store_true', default=False, help='Run with verbose output'
+    )
+    parser.add_argument(
+        '-s', '--show-args', '--show-arguments', action='store_true', default=False,
+        help='Show arguments that may be given to the launch test.'
     )
-
-    parser.add_argument('test_file')
-
-    parser.add_argument('-v', '--verbose',
-                        action='store_true',
-                        default=False,
-                        help='Run with verbose output')
-
-    parser.add_argument('-s', '--show-args', '--show-arguments',
-                        action='store_true',
-                        default=False,
-                        help='Show arguments that may be given to the test file.')
-
     # TODO(hidmic): Provide this option for rostests only.
-    parser.add_argument('-i', '--isolated',
-                        action='store_true',
-                        default=False,
-                        help=('Isolate tests using a custom ROS_DOMAIN_ID.'
-                              'Useful for test parallelization.'))
-
     parser.add_argument(
-        'launch_arguments',
-        nargs='*',
-        help="Arguments to the launch file; '<name>:=<value>' (for duplicates, last one wins)"
+        '-i', '--isolated', action='store_true', default=False,
+        help='Isolate tests using a custom ROS_DOMAIN_ID. Useful for test parallelization.'
     )
-
     parser.add_argument(
-        '--junit-xml',
-        action='store',
-        dest='xmlpath',
-        default=None,
-        help='write junit XML style report to specified path'
+        'launch_arguments', nargs='*',
+        help="Arguments in '<name>:=<value>' format (for duplicates, last one wins)."
     )
-
     parser.add_argument(
-        '--package-name',
-        action='store',
-        default=None,
-        help='a name for the test'
+        '--junit-xml', action='store', dest='xmlpath', default=None,
+        help='Do write xUnit reports to specified path.'
     )
-    args = parser.parse_args()
 
-    if args.verbose:
-        _logger_.setLevel(logging.DEBUG)
-        _logger_.debug('Running with verbose output')
 
+def parse_arguments():
+    parser = argparse.ArgumentParser(
+        description='Launch integration testing tool.'
+    )
+    add_arguments(parser)
+    return parser, parser.parse_args()
+
+
+def run(parser, args, test_runner_cls=LaunchTestRunner):
     if args.isolated:
         domain_id = get_coordinated_domain_id()  # Must copy this to a local to keep it alive
         _logger_.debug('Running with ROS_DOMAIN_ID {}'.format(domain_id))
@@ -94,73 +79,77 @@ def main():
 
     # Load the test file as a module and make sure it has the required
     # components to run it as a launch test
-    _logger_.debug("Loading tests from file '{}'".format(args.test_file))
-    if not os.path.isfile(args.test_file):
+    _logger_.debug("Loading tests from file '{}'".format(args.launch_test_file))
+    if not os.path.isfile(args.launch_test_file):
         # Note to future reader: parser.error also exits as a side effect
-        parser.error("Test file '{}' does not exist".format(args.test_file))
+        parser.error("Test file '{}' does not exist".format(args.launch_test_file))
 
-    args.test_file = os.path.abspath(args.test_file)
-    test_file_basename = os.path.splitext(os.path.basename(args.test_file))[0]
+    args.launch_test_file = os.path.abspath(args.launch_test_file)
+    launch_test_file_basename = os.path.splitext(os.path.basename(args.launch_test_file))[0]
     if not args.package_name:
-        args.package_name = test_file_basename
-    test_module = _load_python_file_as_module(args.package_name, args.test_file)
-
-    _logger_.debug('Checking for generate_test_description')
-    if not hasattr(test_module, 'generate_test_description'):
-        parser.error(
-            "Test file '{}' is missing generate_test_description function".format(args.test_file)
-        )
+        args.package_name = launch_test_file_basename
+    test_module = _load_python_file_as_module(args.package_name, args.launch_test_file)
 
     # This is a list of TestRun objects.  Each run corresponds to one launch.  There may be
     # multiple runs if the launch is parametrized
     test_runs = LoadTestsFromPythonModule(
         test_module, name='{}.{}.launch_tests'.format(
-            args.package_name, test_file_basename
+            args.package_name, launch_test_file_basename
         )
     )
 
     # The runner handles sequcing the launches
-    runner = LaunchTestRunner(
+    runner = test_runner_cls(
         test_runs=test_runs,
         launch_file_arguments=args.launch_arguments,
         debug=args.verbose
     )
 
     _logger_.debug('Validating test configuration')
-    try:
-        runner.validate()
-    except Exception as e:
-        parser.error(e)
+
+    runner.validate()
 
     if args.show_args:
         # TODO pete: Handle the case where different launch descriptions take different args?
         print_arguments_of_launch_description(
             launch_description=test_runs[0].get_launch_description()
         )
-        sys.exit(0)
+        return
 
     _logger_.debug('Running integration test')
-    try:
-        results = runner.run()
-        _logger_.debug('Done running integration test')
-
-        if args.xmlpath:
-            xml_report = unittestResultsToXml(
-                test_results=results, name='{}.{}'.format(
-                    args.package_name, test_file_basename
-                )
+
+    results = runner.run()
+
+    _logger_.debug('Done running integration test')
+
+    if args.xmlpath:
+        xml_report = unittestResultsToXml(
+            test_results=results, name='{}.{}'.format(
+                args.package_name, launch_test_file_basename
             )
-            xml_report.write(args.xmlpath, encoding='utf-8', xml_declaration=True)
+        )
+        xml_report.write(args.xmlpath, encoding='utf-8', xml_declaration=True)
+
+    # There will be one result for every test run (see above where we load the tests)
+    if not all(result.wasSuccessful() for result in results.values()):
+        return 1
+    return 0
+
+
+def main():
+    logging.basicConfig()
 
-        # There will be one result for every test run (see above where we load the tests)
-        for result in results.values():
-            if not result.wasSuccessful():
-                sys.exit(1)
+    parser, args = parse_arguments()
 
+    if args.verbose:
+        _logger_.setLevel(logging.DEBUG)
+        _logger_.debug('Running with verbose output')
+
+    try:
+        sys.exit(run(parser, args))
     except Exception as e:
-        import traceback
-        traceback.print_exc()
         parser.error(e)
+        sys.exit(1)
 
 
 if __name__ == '__main__':
diff --git a/launch_testing/launch_testing/test_runner.py b/launch_testing/launch_testing/test_runner.py
index 98460126d..9b238052a 100644
--- a/launch_testing/launch_testing/test_runner.py
+++ b/launch_testing/launch_testing/test_runner.py
@@ -37,9 +37,11 @@ class _RunnerWorker():
 
     def __init__(self,
                  test_run,
+                 test_run_preamble,
                  launch_file_arguments=[],
                  debug=False):
         self._test_run = test_run
+        self._test_run_preamble = test_run_preamble
         self._launch_service = LaunchService(debug=debug)
         self._processes_launched = threading.Event()  # To signal when all processes started
         self._tests_completed = threading.Event()  # To signal when all the tests have finished
@@ -119,6 +121,7 @@ def run(self):
         # Wrap the test_ld in another launch description so we can bind command line arguments to
         # the test and add our own event handlers for process IO and process exit:
         launch_description = LaunchDescription([
+            *self._test_run_preamble,
             launch.actions.IncludeLaunchDescription(
                 launch.LaunchDescriptionSource(launch_description=test_ld),
                 launch_arguments=parsed_launch_arguments
@@ -212,6 +215,10 @@ def __init__(self,
         self._launch_file_arguments = launch_file_arguments
         self._debug = debug
 
+    def generate_preamble(self):
+        """Generate a launch description preamble for a test to be run with."""
+        return []
+
     def run(self):
         """
         Launch the processes under test and run the tests.
@@ -226,7 +233,11 @@ def run(self):
             if len(self._test_runs) > 1:
                 print('\nStarting test run {}'.format(run))
             try:
-                worker = _RunnerWorker(run, self._launch_file_arguments, self._debug)
+                worker = _RunnerWorker(
+                    run,
+                    self.generate_preamble(),
+                    self._launch_file_arguments,
+                    self._debug)
                 results[run] = worker.run()
             except unittest.case.SkipTest as skip_exception:
                 # If a 'skip' decorator was placed on the generate_test_description function,
diff --git a/launch_testing/test/launch_testing/test_io_handler_and_assertions.py b/launch_testing/test/launch_testing/test_io_handler_and_assertions.py
index 31dcbea8a..c0b33161d 100644
--- a/launch_testing/test/launch_testing/test_io_handler_and_assertions.py
+++ b/launch_testing/test/launch_testing/test_io_handler_and_assertions.py
@@ -13,6 +13,7 @@
 # limitations under the License.
 
 import os
+import re
 import sys
 import unittest
 
@@ -157,6 +158,13 @@ def test_strict_proc_matching_false(self):
             strict_proc_matching=False
         )
 
+    def test_regex_matching(self):
+        assertInStdout(
+            self.proc_output,
+            re.compile(r'Called with arguments \S+'),
+            'terminating_proc-2'
+        )
+
     def test_arguments_disambiguate_processes(self):
         txt = self.EXPECTED_TEXT
         assertInStdout(self.proc_output, txt, 'terminating_proc', '--extra')
diff --git a/launch_testing_ament_cmake/cmake/add_launch_test.cmake b/launch_testing_ament_cmake/cmake/add_launch_test.cmake
index e8277d7e3..5cac7810d 100644
--- a/launch_testing_ament_cmake/cmake/add_launch_test.cmake
+++ b/launch_testing_ament_cmake/cmake/add_launch_test.cmake
@@ -47,77 +47,81 @@
 # ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 # POSSIBILITY OF SUCH DAMAGE.
 
-#
-# Add a launch test
-#
-# :param file: The launch test file containing the test to run
-# :type file: string
-# :param TARGET: The test target name
-# :type TARGET: string
-# :param PYTHON_EXECUTABLE: The python executable to use for the test
-# :type PYTHON_EXECUTABLE: string
-# :param TIMEOUT: The test timeout in seconds
-# :type TIMEOUT: integer
-# :param ARGS: Launch arguments to pass to the launch test
-# :type ARGS: string
-function(add_launch_test file)
 
-  cmake_parse_arguments(_add_launch_test
+macro(parse_launch_test_arguments namespace filename)
+  cmake_parse_arguments(${namespace}
     ""
     "TARGET;TIMEOUT;PYTHON_EXECUTABLE"
     "ARGS"
     ${ARGN})
 
-  if(NOT _add_launch_test_TIMEOUT)
-    set(_add_launch_test_TIMEOUT 60)
+  if(NOT ${namespace}_TIMEOUT)
+    set(${namespace}_TIMEOUT 60)
   endif()
 
-  if(NOT _add_launch_test_PYTHON_EXECUTABLE)
-    set(_add_launch_test_PYTHON_EXECUTABLE "${PYTHON_EXECUTABLE}")
+  if(NOT ${namespace}_PYTHON_EXECUTABLE)
+    set(${namespace}_PYTHON_EXECUTABLE "${PYTHON_EXECUTABLE}")
   endif()
 
-  set(_file_name _file_name-NOTFOUND)
-  if(IS_ABSOLUTE ${file})
-    set(_file_name ${file})
+  set(${namespace}_FILE_NAME NOTFOUND)
+  if(IS_ABSOLUTE ${filename})
+    set(${namespace}_FILE_NAME ${filename})
   else()
-    find_file(_file_name ${file}
+    find_file(${namespace}_FILE_NAME ${filename}
               PATHS ${CMAKE_CURRENT_SOURCE_DIR}
               NO_DEFAULT_PATH
               NO_CMAKE_FIND_ROOT_PATH)
-    if(NOT _file_name)
-      message(FATAL_ERROR "Can't find launch test file \"${file}\"")
+    if(NOT ${namespace}_FILE_NAME)
+      message(FATAL_ERROR "Can't find launch test file \"${filename}\"")
     endif()
   endif()
 
-  if (NOT _add_launch_test_TARGET)
+  if (NOT ${namespace}_TARGET)
     # strip PROJECT_SOURCE_DIR and PROJECT_BINARY_DIR from absolute filename to get unique test name (as rostest does it internally)
-    set(_add_launch_test_TARGET ${_file_name})
-    rostest__strip_prefix(_add_launch_test_TARGET "${PROJECT_SOURCE_DIR}/")
-    rostest__strip_prefix(_add_launch_test_TARGET "${PROJECT_BINARY_DIR}/")
-    string(REPLACE "/" "_" _add_launch_test_TARGET ${_add_launch_test_TARGET})
+    set(${namespace}_TARGET ${${namespace}_FILE_NAME})
+    rostest__strip_prefix(${namespace}_TARGET "${PROJECT_SOURCE_DIR}/")
+    rostest__strip_prefix(${namespace}_TARGET "${PROJECT_BINARY_DIR}/")
+    string(REPLACE "/" "_" ${namespace}_TARGET ${${namespace}_TARGET})
   endif()
 
-  set(result_file "${AMENT_TEST_RESULTS_DIR}/${PROJECT_NAME}/${_add_launch_test_TARGET}.xunit.xml")
+  set(${namespace}_RESULT_FILE "${AMENT_TEST_RESULTS_DIR}/${PROJECT_NAME}/${${namespace}_TARGET}.xunit.xml")
+endmacro()
+
+
+#
+# Add a launch test
+#
+# :param filename: The launch test file containing the test to run
+# :type filename: string
+# :param TARGET: The test target name
+# :type TARGET: string
+# :param PYTHON_EXECUTABLE: The python executable to use for the test
+# :type PYTHON_EXECUTABLE: string
+# :param TIMEOUT: The test timeout in seconds
+# :type TIMEOUT: integer
+# :param ARGS: Launch arguments to pass to the launch test
+# :type ARGS: string
+function(add_launch_test filename)
+  parse_launch_test_arguments(_launch_test ${filename} ${ARGN})
 
   set(cmd
-    "${_add_launch_test_PYTHON_EXECUTABLE}"
+    "${_launch_test_PYTHON_EXECUTABLE}"
     "-m"
     "launch_testing.launch_test"
-    "${_file_name}"
-    "${_add_launch_test_ARGS}"
-    "--junit-xml=${result_file}"
+    "${_launch_test_FILE_NAME}"
+    "${_launch_test_ARGS}"
+    "--junit-xml=${_launch_test_RESULT_FILE}"
     "--package-name=${PROJECT_NAME}"
   )
 
   ament_add_test(
-    "${_add_launch_test_TARGET}"
+    "${_launch_test_TARGET}"
     COMMAND ${cmd}
     OUTPUT_FILE "${CMAKE_BINARY_DIR}/launch_test/CHANGEME.txt"
-    RESULT_FILE "${result_file}"
-    TIMEOUT "${_add_launch_test_TIMEOUT}"
-    ${_add_launch_test_UNPARSED_ARGUMENTS}
+    RESULT_FILE "${_launch_test_RESULT_FILE}"
+    TIMEOUT "${_launch_test_TIMEOUT}"
+    ${_launch_test_UNPARSED_ARGUMENTS}
   )
-
 endfunction()
 
 macro(rostest__strip_prefix var prefix)