🚨 🚨 Source Mixpanel: fix typing (#30025)

Co-authored-by: Anatolii Yatsuk <tolikyatsuk@gmail.com> Co-authored-by: Pedro S. Lopez <pedroslopez@me.com> Co-authored-by: pedroslopez <pedroslopez@users.noreply.github.com> Co-authored-by: Anatolii Yatsuk <35109939+tolik0@users.noreply.github.com>
airbytehq · Oct 4, 2023 · bd0e9c9 · bd0e9c9
1 parent dceca41
commit bd0e9c9
Show file tree

Hide file tree

Showing 9 changed files with 96 additions and 59 deletions.
diff --git a/airbyte-integrations/connectors/source-mixpanel/Dockerfile b/airbyte-integrations/connectors/source-mixpanel/Dockerfile
@@ -13,5 +13,5 @@ ENV AIRBYTE_ENTRYPOINT "python /airbyte/integration_code/main.py"
 ENTRYPOINT ["python", "/airbyte/integration_code/main.py"]
 
 
-LABEL io.airbyte.version=0.1.41
+LABEL io.airbyte.version=1.0.0
 LABEL io.airbyte.name=airbyte/source-mixpanel
diff --git a/airbyte-integrations/connectors/source-mixpanel/acceptance-test-config.yml b/airbyte-integrations/connectors/source-mixpanel/acceptance-test-config.yml
@@ -1,6 +1,7 @@
 # See [Connector Acceptance Tests](https://docs.airbyte.com/connector-development/testing-connectors/connector-acceptance-tests-reference)
 # for more information about how to configure these tests
 connector_image: airbyte/source-mixpanel:dev
+# custom configuration is used for tests to speed up testing and avoid hitting rate limits
 custom_environment_variables:
   REQS_PER_HOUR_LIMIT: 0
   AVAILABLE_TESTING_RANGE_DAYS: 10

diff --git a/...grations/connectors/source-mixpanel/integration_tests/configured_catalog_incremental.json b/...grations/connectors/source-mixpanel/integration_tests/configured_catalog_incremental.json
@@ -21,12 +21,12 @@
         "supported_sync_modes": ["full_refresh", "incremental"],
         "source_defined_cursor": true,
         "default_cursor_field": ["time"],
-        "source_defined_primary_key": [["distinct_id"]]
+        "source_defined_primary_key": [["distinct_id"], ["event"], ["time"]]
       },
       "sync_mode": "incremental",
       "destination_sync_mode": "append",
       "cursor_field": ["time"],
-      "primary_key": [["distinct_id"]]
+      "primary_key": [["distinct_id"], ["event"], ["time"]]
     },
     {
       "stream": {

diff --git a/airbyte-integrations/connectors/source-mixpanel/metadata.yaml b/airbyte-integrations/connectors/source-mixpanel/metadata.yaml
@@ -6,7 +6,7 @@ data:
   connectorSubtype: api
   connectorType: source
   definitionId: 12928b32-bf0a-4f1e-964f-07e12e37153a
-  dockerImageTag: 0.1.41
+  dockerImageTag: 1.0.0
   dockerRepository: airbyte/source-mixpanel
   githubIssueLabel: source-mixpanel
   icon: mixpanel.svg
@@ -18,6 +18,11 @@ data:
     oss:
       enabled: true
   releaseStage: generally_available
+  releases:
+    breakingChanges:
+      1.0.0:
+        message: In this release, the datetime field of stream engage has had its type changed from date-time to string due to inconsistent data from Mixpanel. Additionally, the primary key for stream export has been fixed to uniquely identify records. Users will need to refresh the source schema and reset affected streams after upgrading.
+        upgradeDeadline: "2023-10-31"
   suggestedStreams:
     streams:
       - export

diff --git a/airbyte-integrations/connectors/source-mixpanel/source_mixpanel/streams/engage.py b/airbyte-integrations/connectors/source-mixpanel/source_mixpanel/streams/engage.py
@@ -184,7 +184,8 @@ def get_json_schema(self) -> Mapping[str, Any]:
         types = {
             "boolean": {"type": ["null", "boolean"]},
             "number": {"type": ["null", "number"], "multipleOf": 1e-20},
-            "datetime": {"type": ["null", "string"], "format": "date-time"},
+            # no format specified as values can be "2021-12-16T00:00:00", "1638298874", "15/08/53895"
+            "datetime": {"type": ["null", "string"]},
             "object": {"type": ["null", "object"], "additionalProperties": True},
             "list": {"type": ["null", "array"], "required": False, "items": {}},
             "string": {"type": ["null", "string"]},

diff --git a/airbyte-integrations/connectors/source-mixpanel/source_mixpanel/streams/export.py b/airbyte-integrations/connectors/source-mixpanel/source_mixpanel/streams/export.py
@@ -75,7 +75,7 @@ class Export(DateSlicesMixin, IncrementalMixpanelStream):
      3 queries per second and 60 queries per hour.
     """
 
-    primary_key: str = "distinct_id"
+    primary_key: str = ["distinct_id", "event", "time"]
     cursor_field: str = "time"
 
     transformer = TypeTransformer(TransformConfig.DefaultSchemaNormalization)

diff --git a/airbyte-integrations/connectors/source-mixpanel/unit_tests/test_streams.py b/airbyte-integrations/connectors/source-mixpanel/unit_tests/test_streams.py
@@ -279,23 +279,47 @@ def engage_schema_response():
         200,
         {
             "results": {
-                "$browser": {"count": 124, "type": "string"},
-                "$browser_version": {"count": 124, "type": "string"},
                 "$created": {"count": 124, "type": "string"},
+                "$is_active": {"count": 412, "type": "boolean"},
+                "$CreatedDateTimestamp": {"count": 300, "type": "number"},
+                "$CreatedDate": {"count": 11, "type": "datetime"},
+                "$properties": {"count": 2, "type": "object"},
+                "$tags": {"count": 131, "type": "list"},
             }
         },
     )
 
 
 def test_engage_schema(requests_mock, engage_schema_response, config):
-
-    stream = EngageSchema(authenticator=MagicMock(), **config)
-    requests_mock.register_uri("GET", get_url_to_mock(stream), engage_schema_response)
-
-    records = stream.read_records(sync_mode=SyncMode.full_refresh)
-
-    records_length = sum(1 for _ in records)
-    assert records_length == 3
+    stream = Engage(authenticator=MagicMock(), **config)
+    requests_mock.register_uri("GET", get_url_to_mock(EngageSchema(authenticator=MagicMock(), **config)), engage_schema_response)
+    assert stream.get_json_schema() == {
+        "$schema": "http://json-schema.org/draft-07/schema#",
+        "additionalProperties": True,
+        "properties": {
+            "CreatedDate": {"type": ["null", "string"]},
+            "CreatedDateTimestamp": {"multipleOf": 1e-20, "type": ["null", "number"]},
+            "browser": {"type": ["null", "string"]},
+            "browser_version": {"type": ["null", "string"]},
+            "city": {"type": ["null", "string"]},
+            "country_code": {"type": ["null", "string"]},
+            "created": {"type": ["null", "string"]},
+            "distinct_id": {"type": ["null", "string"]},
+            "email": {"type": ["null", "string"]},
+            "first_name": {"type": ["null", "string"]},
+            "id": {"type": ["null", "string"]},
+            "is_active": {"type": ["null", "boolean"]},
+            "last_name": {"type": ["null", "string"]},
+            "last_seen": {"format": "date-time", "type": ["null", "string"]},
+            "name": {"type": ["null", "string"]},
+            "properties": {"additionalProperties": True, "type": ["null", "object"]},
+            "region": {"type": ["null", "string"]},
+            "tags": {"items": {}, "required": False, "type": ["null", "array"]},
+            "timezone": {"type": ["null", "string"]},
+            "unblocked": {"type": ["null", "string"]},
+        },
+        "type": "object",
+    }
 
 
 def test_update_engage_schema(requests_mock, config):

diff --git a/docs/integrations/sources/mixpanel-migrations.md b/docs/integrations/sources/mixpanel-migrations.md
@@ -0,0 +1,5 @@
+# Mixpanel Migration Guide
+
+## Upgrading to 1.0.0
+
+In this release, the datetime field of stream engage has had its type changed from date-time to string due to inconsistent data from Mixpanel. Additionally, the primary key for stream export has been fixed to uniquely identify records. Users will need to refresh the source schema and reset affected streams after upgrading.