feat: add params parameter to the multiscan of ggclient

Author: alina-tuholukova-gg

pygitguardian/client.py

@@ -273,6 +273,7 @@ def multi_content_scan(
         self,
         documents: List[Dict[str, str]],
         extra_headers: Optional[Dict[str, str]] = None,
+        ignore_known_secrets: Optional[bool] = None,
     ) -> Union[Detail, MultiScanResult]:
         """
         multi_content_scan handles the /multiscan endpoint of the API.
@@ -284,6 +285,7 @@ def multi_content_scan(
         and, optionally, filename.
             example: [{"document":"example content","filename":"intro.py"}]
         :param extra_headers: additional headers to add to the request
+        :param ignore_known_secrets: indicates whether known secrets should be ignored
         :return: Detail or ScanResult response and status code
         """
         if len(documents) > MULTI_DOCUMENT_LIMIT:
@@ -298,10 +300,16 @@ def multi_content_scan(
         else:
             raise TypeError("each document must be a dict")
 
+        params = (
+            {"ignore_known_secrets": ignore_known_secrets}
+            if ignore_known_secrets
+            else {}
+        )
         resp = self.post(
             endpoint="multiscan",
             data=request_obj,
             extra_headers=extra_headers,
+            params=params,
         )
 
         obj: Union[Detail, MultiScanResult]

pygitguardian/models.py

@@ -205,6 +205,7 @@ class PolicyBreakSchema(BaseSchema):
     break_type = fields.String(data_key="type", required=True)
     policy = fields.String(required=True)
     validity = fields.String(required=False, load_default=None, dump_default=None)
+    known_secret = fields.Boolean(required=False, load_default=False, dump_default=None)
     matches = fields.List(fields.Nested(MatchSchema), required=True)
 
     @post_load
@@ -228,12 +229,14 @@ def __init__(
         policy: str,
         validity: str,
         matches: List[Match],
+        known_secret: bool = False,
         **kwargs: Any,
     ) -> None:
         super().__init__()
         self.break_type = break_type
         self.policy = policy
         self.validity = validity
+        self.known_secret = known_secret
         self.matches = matches
 
     @property

tests/test_client.py

@@ -548,6 +548,52 @@ def test_extra_headers(
     assert expected_headers == kwargs["headers"]
 
 
+@patch("requests.Session.request")
+def test_multiscan_parameters(
+    request_mock: Mock,
+    client: GGClient,
+):
+    """
+    GIVEN a ggclient
+    WHEN calling multi_content_scan with parameters
+    THEN the parameters are passed in the request
+    """
+    mock_response = Mock(spec=Response)
+    mock_response.headers = {"content-type": "application/json"}
+    mock_response.status_code = 200
+    mock_response.json.return_value = [
+        {
+            "policy_break_count": 1,
+            "policies": ["pol"],
+            "policy_breaks": [
+                {
+                    "type": "break",
+                    "policy": "mypol",
+                    "matches": [
+                        {
+                            "match": "hello",
+                            "type": "hello",
+                        }
+                    ],
+                }
+            ],
+        }
+    ]
+
+    request_mock.return_value = mock_response
+
+    params = {"ignore_known_secrets": True}
+
+    client.multi_content_scan(
+        [{"filename": FILENAME, "document": DOCUMENT}],
+        ignore_known_secrets=True,
+    )
+
+    assert request_mock.called
+    # 1 is for kwargs
+    assert request_mock.call_args[1]["params"] == params
+
+
 def test_quota_overview(client: GGClient):
     with my_vcr.use_cassette("quota.yaml"):
         quota_response = client.quota_overview()

tests/test_models.py

@@ -147,3 +147,23 @@ def test_detail_renames_error_field(self):
         """
         detail = Detail.SCHEMA.load({"error": "An error message"})
         assert detail.detail == "An error message"
+
+    @pytest.mark.parametrize("known_secret", [True, False])
+    def test_policy_break_known_secret_field(self, known_secret):
+        """
+        GIVEN the data with policy breaks
+        WHEN loading using the schema
+        THEN known_secret is parsed correctly with the default value set to False
+        """
+        data = {
+            "type": "hello",
+            "policy": "hello",
+            "validity": "hey",
+            "matches": [{"match": "hello", "type": "hello"}],
+        }
+        if known_secret:
+            data["known_secret"] = True
+
+        obj = PolicyBreakSchema().load(data)
+
+        assert obj.known_secret is known_secret