[ATO-1152] Port of Add license hashes to telemetry

changelog/12567.misc.md

@@ -0,0 +1 @@
+Add plugin hook for getting the license hash from Rasa Plus and add the hash to telemetry context.

docs/docs/telemetry/telemetry.mdx

@@ -81,6 +81,8 @@ Specifically, we collect the following information for all telemetry events:
 - General OS level information (operating system, number of CPUs, number of
   GPUs and whether the command is run inside a CI)
 - Current Rasa and Python version
+- Whether the command is run inside a Docker container
+- Hash of the license (if you are using Rasa Pro)
 
 Here is an example report that shows the data reported to Rasa after running
 `rasa train`:
@@ -117,7 +119,9 @@ Here is an example report that shows the data reported to Rasa after running
     "project": "a0a7178e6e5f9e6484c5cfa3ea4497ffc0c96d0ad3f3ad8e9399a1edd88e3cf4",
     "python": "3.7.5",
     "rasa_open_source": "2.0.0",
-    "cpu": 16
+    "cpu": 16,
+    "docker": false,
+    "license_hash": "t1a7170e6e5f9e6484c5cfa3ea4497ffc0c96a0ad3f3ad8e9399adadd88e3cf5"
   }
 }
 ```

rasa/plugin.py

@@ -52,3 +52,8 @@ def configure_commandline(cmdline_arguments: argparse.Namespace) -> Optional[Tex
 @hookspec  # type: ignore[misc]
 def init_telemetry(endpoints_file: Optional[Text]) -> None:
     """Hook specification for initialising plugin telemetry."""
+
+
+@hookspec(firstresult=True)  # type: ignore[misc]
+def get_license_hash() -> Optional[Text]:
+    """Hook specification for getting the license hash."""

rasa/telemetry.py

@@ -26,6 +26,7 @@
     CONFIG_TELEMETRY_ID,
 )
 from rasa.engine.storage.local_model_storage import LocalModelStorage
+from rasa.plugin import plugin_manager
 from rasa.shared.constants import DOCS_URL_TELEMETRY
 from rasa.shared.exceptions import RasaException
 import rasa.shared.utils.io
@@ -490,6 +491,9 @@ def _default_context_fields() -> Dict[Text, Any]:
             "cpu": multiprocessing.cpu_count(),
             "docker": _is_docker(),
         }
+        license_hash = plugin_manager().hook.get_license_hash()
+        if license_hash:
+            TELEMETRY_CONTEXT["license_hash"] = license_hash
 
     # avoid returning the cached dict --> caller could modify the dictionary...
     # usually we would use `lru_cache`, but that doesn't return a dict copy and

tests/test_telemetry.py

@@ -6,7 +6,7 @@
 
 from _pytest.monkeypatch import MonkeyPatch
 import jsonschema
-from unittest.mock import Mock
+from unittest.mock import MagicMock, Mock
 import pytest
 import responses
 
@@ -34,6 +34,15 @@ def patch_global_config_path(tmp_path: Path) -> Generator[None, None, None]:
     rasa.constants.GLOBAL_USER_CONFIG_PATH = default_location
 
 
+@pytest.fixture(autouse=True)
+def patch_telemetry_context() -> Generator[None, None, None]:
+    """Use a telemetry context for each test to avoid tests influencing each other."""
+    defaut_context = telemetry.TELEMETRY_CONTEXT
+    telemetry.TELEMETRY_CONTEXT = None
+    yield
+    telemetry.TELEMETRY_CONTEXT = defaut_context
+
+
 async def test_events_schema(
     monkeypatch: MonkeyPatch, default_agent: Agent, config_path: Text
 ):
@@ -485,3 +494,28 @@ def test_context_contains_os():
     context.pop("os")
 
     assert "os" in telemetry._default_context_fields()
+
+
+def test_context_contains_license_hash(monkeypatch: MonkeyPatch) -> None:
+    mock = MagicMock()
+    mock.return_value.hook.get_license_hash.return_value = "1234567890"
+    monkeypatch.setattr("rasa.telemetry.plugin_manager", mock)
+    context = telemetry._default_context_fields()
+
+    assert "license_hash" in context
+    assert mock.return_value.hook.get_license_hash.called
+    assert context["license_hash"] == "1234567890"
+
+    # make sure it is still there after removing it
+    context.pop("license_hash")
+    assert "license_hash" in telemetry._default_context_fields()
+
+
+def test_context_does_not_contain_license_hash(monkeypatch: MonkeyPatch) -> None:
+    mock = MagicMock()
+    mock.return_value.hook.get_license_hash.return_value = None
+    monkeypatch.setattr("rasa.telemetry.plugin_manager", mock)
+    context = telemetry._default_context_fields()
+
+    assert "license_hash" not in context
+    assert mock.return_value.hook.get_license_hash.called