google-assistant-sdk/pushtotalk: add html output

Bug: 77657440 Bug: 78541315 Change-Id: Ia8aefa2373fc5824addd369775bbcce36c189840
googlesamples · May 3, 2018 · fa6d7d2 · fa6d7d2
1 parent 92bd8ac
commit fa6d7d2
Show file tree

Hide file tree

Showing 3 changed files with 48 additions and 4 deletions.
diff --git a/google-assistant-sdk/googlesamples/assistant/grpc/pushtotalk.py b/google-assistant-sdk/googlesamples/assistant/grpc/pushtotalk.py
@@ -40,18 +40,21 @@
     from . import (
         assistant_helpers,
         audio_helpers,
+        browser_helpers,
         device_helpers
     )
 except (SystemError, ImportError):
     import assistant_helpers
     import audio_helpers
+    import browser_helpers
     import device_helpers
 
 
 ASSISTANT_API_ENDPOINT = 'embeddedassistant.googleapis.com'
 END_OF_UTTERANCE = embedded_assistant_pb2.AssistResponse.END_OF_UTTERANCE
 DIALOG_FOLLOW_ON = embedded_assistant_pb2.DialogStateOut.DIALOG_FOLLOW_ON
 CLOSE_MICROPHONE = embedded_assistant_pb2.DialogStateOut.CLOSE_MICROPHONE
+PLAYING = embedded_assistant_pb2.ScreenOutConfig.PLAYING
 DEFAULT_GRPC_DEADLINE = 60 * 3 + 5
 
 
@@ -70,12 +73,13 @@ class SampleAssistant(object):
     """
 
     def __init__(self, language_code, device_model_id, device_id,
-                 conversation_stream,
+                 conversation_stream, display,
                  channel, deadline_sec, device_handler):
         self.language_code = language_code
         self.device_model_id = device_model_id
         self.device_id = device_id
         self.conversation_stream = conversation_stream
+        self.display = display
 
         # Opaque blob provided in AssistResponse that,
         # when provided in a follow-up AssistRequest,
@@ -142,6 +146,7 @@ def iter_log_assist_requests():
                                       for r in resp.speech_results))
             if len(resp.audio_out.audio_data) > 0:
                 if not self.conversation_stream.playing:
+                    self.conversation_stream.stop_recording()
                     self.conversation_stream.start_playback()
                     logging.info('Playing assistant response.')
                 self.conversation_stream.write(resp.audio_out.audio_data)
@@ -165,6 +170,9 @@ def iter_log_assist_requests():
                 fs = self.device_handler(device_request)
                 if fs:
                     device_actions_futures.extend(fs)
+            if self.display and resp.screen_out.data:
+                system_browser = browser_helpers.system_browser
+                system_browser.display(resp.screen_out.data)
 
         if len(device_actions_futures):
             logging.info('Waiting for device executions to complete.')
@@ -200,6 +208,9 @@ def gen_assist_requests(self):
                 device_model_id=self.device_model_id,
             )
         )
+        if self.display:
+            config.screen_out_config.screen_mode = PLAYING
+
         # The first AssistRequest must contain the AssistConfig
         # and no audio data.
         yield embedded_assistant_pb2.AssistRequest(config=config)
@@ -241,6 +252,9 @@ def gen_assist_requests(self):
               metavar='<language code>',
               default='en-US',
               help='Language code of the Assistant')
+@click.option('--display', is_flag=True, default=False,
+              help='Enable visual display of Assistant '
+                   'rich media responses (for certain queries).')
 @click.option('--verbose', '-v', is_flag=True, default=False,
               help='Verbose logging.')
 @click.option('--input-audio-file', '-i',
@@ -279,7 +293,8 @@ def gen_assist_requests(self):
 @click.option('--once', default=False, is_flag=True,
               help='Force termination after a single conversation.')
 def main(api_endpoint, credentials, project_id,
-         device_model_id, device_id, device_config, lang, verbose,
+         device_model_id, device_id, device_config,
+         lang, display, verbose,
          input_audio_file, output_audio_file,
          audio_sample_rate, audio_sample_width,
          audio_iter_size, audio_block_size, audio_flush_size,
@@ -424,7 +439,7 @@ def blink(speed, number):
             time.sleep(delay)
 
     with SampleAssistant(lang, device_model_id, device_id,
-                         conversation_stream,
+                         conversation_stream, display,
                          grpc_channel, grpc_deadline,
                          device_handler) as assistant:
         # If file arguments are supplied:

diff --git a/google-assistant-sdk/tests/data/grapefruit.riff b/google-assistant-sdk/tests/data/grapefruit.riff
diff --git a/google-assistant-sdk/tests/test_endtoend.py b/google-assistant-sdk/tests/test_endtoend.py
@@ -83,6 +83,35 @@ def test_endtoend_pushtotalk():
     assert os.path.getsize(audio_out_file) > 0
 
 
+def test_endtoend_pushtotalk_htmloutput(device_model, device_instance):
+    temp_dir = tempfile.mkdtemp()
+    audio_out_file = os.path.join(temp_dir, 'out.raw')
+    env = os.environ.copy()
+    env['TMPDIR'] = temp_dir
+    out = subprocess.check_output(['python', '-m',
+                                   'googlesamples.assistant.grpc.pushtotalk',
+                                   '--verbose',
+                                   '--device-model-id', device_model,
+                                   '--device-id', device_instance,
+                                   '-i', 'tests/data/grapefruit.riff',
+                                   '--display',
+                                   '-o', audio_out_file],
+                                  stderr=subprocess.STDOUT, env=env)
+    print(out)
+    assert 'grapefruit' in builtins.str(out).lower()
+    assert os.path.getsize(audio_out_file) > 0
+    files = [os.path.join(path, f)
+             for path, _, fs in os.walk(temp_dir) for f in fs]
+    assert len(files) > 0
+    screen_out = None
+    for f in files:
+        if os.path.basename(f) == 'google-assistant-sdk-screen-out.html':
+            screen_out = f
+    assert screen_out is not None
+    with open(screen_out, 'r') as f:
+        assert 'pamplemousse' in f.read()
+
+
 def test_registration_pushtotalk(device_model):
     temp_dir = tempfile.mkdtemp()
     audio_out_file = os.path.join(temp_dir, 'out.raw')
@@ -136,7 +165,7 @@ def test_endtoend_textinput(device_model, device_instance):
     assert 'pamplemousse' in out
 
 
-def test_endtoend_htmloutput(device_model, device_instance):
+def test_endtoend_textinput_htmloutput(device_model, device_instance):
     temp_dir = tempfile.mkdtemp()
     env = os.environ.copy()
     env['TMPDIR'] = temp_dir