add support for specifying the agent queue

This is needed so we can have separate queues for tests that require
special instances. For example, running performance tests require
nothing else to run at the same time.

Signed-off-by: Andreea Florescu <fandree@amazon.com>

.buildkite/autogenerate_pipeline.py

@@ -127,6 +127,11 @@ class BuildkiteStep:
         if timeout:
             self.step_config['timeout_in_minutes'] = timeout
 
+    def _set_agent_queue(self, queue):
+        """Set the agent queue if provided in the json input."""
+        if queue:
+            self.step_config['agents']['queue'] = queue
+
     def _add_docker_config(self, cfg):
         """ Add configuration for docker if given in the json input. """
 
@@ -207,6 +212,7 @@ class BuildkiteStep:
         docker = input.get('docker_plugin')
         conditional = input.get('conditional')
         timeout = input.get('timeout_in_minutes')
+        queue = input.get('queue')
 
         # Mandatory keys.
         assert test_name, "Step is missing test name."
@@ -227,6 +233,7 @@ class BuildkiteStep:
         self._set_conditional(conditional)
         self._add_docker_config(docker)
         self._set_timeout_in_minutes(timeout)
+        self._set_agent_queue(queue)
 
         # Override/add configuration from environment variables.
         self._env_override_agent_tags(test_name)
@@ -238,7 +245,7 @@ class BuildkiteStep:
         # forwarding the key, values without any change.
         # We need to filter for keys that have special meaning and which we
         # don't want to re-add.
-        special_keys = ['conditional', 'docker_plugin', 'platform', 'test_name']
+        special_keys = ['conditional', 'docker_plugin', 'platform', 'test_name', 'queue']
         additional_keys = {k: v for k, v in input.items() if not (k in self.step_config) and
                            not(k in special_keys)}
         if additional_keys: