SIENTIAPDE-1231

Enhance type hinting and code structure in orchestrator activities

- Added type hints for better code clarity in `formatters.py`, `mongo_db.py`, `temporal_manager.py`, and `email_builder.py`.
- Updated `.gitignore` to include `coverage.xml` for improved coverage reporting.
- Improved readability by restructuring variable declarations and method signatures.

orchestrator/activities/formatters.py

@@ -1,3 +1,5 @@
+from collections.abc import Hashable
+
 from temporalio import activity, workflow
 
 with workflow.unsafe.imports_passed_through():
@@ -87,7 +89,10 @@ class Formatters(BaseActivity):
 
         pipelines = input_data['pipelines']
 
-        schedule_config = {self.scouter_namespace: {}, self.laborious_namespace: {}}
+        schedule_config: dict[str, dict[str, Any]] = {
+            self.scouter_namespace: {},
+            self.laborious_namespace: {},
+        }
 
         for pipeline in pipelines:
             if pipeline['workflow_type'] == 'scouter':
@@ -155,7 +160,7 @@ class Formatters(BaseActivity):
         number_of_slots = len(active_ingestors) if active_ingestors else 1
         tags_per_slot = ceil(number_of_tags / number_of_slots)
 
-        slot_config = {}
+        slot_config: dict[str, Any] = {}
         last_index = 0
 
         for i in range(1, number_of_slots):
@@ -212,7 +217,7 @@ class Formatters(BaseActivity):
 
         schedule_config = input_data['schedule_config']
 
-        config = {}
+        config: dict[str, dict[str, str]] = {}
         for schedule in schedule_config:
             namespace = schedule['namespace']
             schedule_name = schedule['schedule_name']
@@ -290,9 +295,18 @@ class Formatters(BaseActivity):
         current_schedule_config = input_data['current_schedule_config']
         schedule_config = input_data['schedule_config']
 
-        to_update = {self.scouter_namespace: {}, self.laborious_namespace: {}}
-        to_create = {self.scouter_namespace: {}, self.laborious_namespace: {}}
-        to_delete = {self.scouter_namespace: [], self.laborious_namespace: []}
+        to_update: dict[str, dict[str, Any]] = {
+            self.scouter_namespace: {},
+            self.laborious_namespace: {},
+        }
+        to_create: dict[str, dict[str, Any]] = {
+            self.scouter_namespace: {},
+            self.laborious_namespace: {},
+        }
+        to_delete: dict[str, list[str]] = {
+            self.scouter_namespace: [],
+            self.laborious_namespace: [],
+        }
 
         for namespace, schedules in schedule_config.items():
             current_schedules = current_schedule_config.get(namespace, {})
@@ -352,7 +366,11 @@ class Formatters(BaseActivity):
         return output
 
     def send_success_report(
-        self, metadata: dict[str, Any], message: str, notification_id: str, attachment: str = None
+        self,
+        metadata: dict[str, Any],
+        message: str,
+        notification_id: str,
+        attachment: Any | None = None,
     ) -> None:
         """
         Sends a success notification report.
@@ -393,7 +411,9 @@ class Formatters(BaseActivity):
             attachment_content=attachment,
         )
 
-    def parse_report_schedule(self, input_data: dict[str, Any]) -> tuple[list[str], dict[str, Any]]:
+    def parse_report_schedule(
+        self, input_data: list[dict[str, Any]]
+    ) -> tuple[list[str], dict[str, Any]]:
         """
         Parses the report schedule data to extract success and error information.
 
@@ -424,7 +444,7 @@ class Formatters(BaseActivity):
 
         return success_keys, error_keys
 
-    def parse_report(self, input_data: dict[str, Any]) -> tuple[list[str], list[str]]:
+    def parse_report(self, input_data: dict[str, dict[str, Any]]) -> tuple[list[str], list[str]]:
         """
         Parses the report data to extract success and error keys.
 
@@ -587,7 +607,7 @@ class Formatters(BaseActivity):
                 )
 
     @activity.defn(name='format_log_report')
-    async def format_log_report(self, input_data: dict[str, Any]) -> dict[str, Any]:
+    async def format_log_report(self, input_data: dict[str, Any]) -> dict[Hashable, Any]:
         """
         Formats the receiver_groups status to a dataframe to be stored in the database.
 
@@ -636,7 +656,9 @@ class Formatters(BaseActivity):
                     if group_name not in data[key]['groups']:
                         data[key]['groups'].append(group_name)
 
-        return DataFrame(list(data.values())).to_dict()
+        data_values: DataFrame = DataFrame(list(data.values()))
+
+        return data_values.to_dict()
 
     @activity.defn(name='filter_notification_reports')
     async def filter_notification_reports(self, input_data: dict[str, Any]) -> dict[str, Any]:

orchestrator/activities/mongo_db.py

@@ -70,7 +70,9 @@ class MongoDB(BaseActivity):
         self.connection_string = connection_string
         self.database_name = database_name
 
-        self.client = MongoClient(self.connection_string, serverSelectionTimeoutMS=5000)
+        self.client: MongoClient = MongoClient(
+            self.connection_string, serverSelectionTimeoutMS=5000
+        )
         self.client.server_info()  # Trigger an exception if connection fails
 
         self.database = self.client[self.database_name]

orchestrator/activities/temporal_manager.py

@@ -52,7 +52,7 @@ class TemporalManager(BaseActivity):
         self.temporal_host = host
         self.scouter_namespace = scouter_namespace
         self.laborious_namespace = laborious_namespace
-        self.temporal_clients = {}
+        self.temporal_clients: dict[str, Client] = {}
 
         self.model_id_id_key = SearchAttributeKey.for_keyword('model_id')
         self.model_name_id_key = SearchAttributeKey.for_keyword('model_name')
@@ -84,7 +84,7 @@ class TemporalManager(BaseActivity):
         }
 
     @activity.defn(name='normalize_schedules')
-    async def normalize_schedules(self, input_data: dict[str, Any]) -> dict[str, Any]:
+    async def normalize_schedules(self, input_data: dict[str, Any]):
         """
         Normalize schedules. Removes schedules with no update time in mongo db collection "orchestrated_schedules".
         input_data:

orchestrator/utils/email_builder.py

@@ -1,3 +1,5 @@
+from typing import Any
+
 from jinja2 import Template
 from sientia_do.observability.logger import Logger
 
@@ -75,12 +77,12 @@ class EmailBuilder:
             else '',
         }
 
-    def build_email(self, report_data: list[dict], mail_type: str) -> str:
+    def build_email(self, report_data: list[dict[str, Any]], mail_type: str) -> str:
         """
         Builds the email HTML by organizing report data by notification level and model.
 
         Args:
-            report_data (list[dict]): List of notification reports, each containing:
+            report_data (List[Dict[str, Any]]): List of notification reports, each containing:
                 - level (str): Notification level (ERROR, WARNING, INFO)
                 - model_name (str): Name of the model
                 - Additional notification details
@@ -88,7 +90,7 @@ class EmailBuilder:
         Returns:
             str: Complete HTML email content ready for sending.
         """
-        general_events = {}
+        general_events: dict[str, dict[str, Any]] = {}
 
         for report in report_data:
             level = report['level']
@@ -100,13 +102,17 @@ class EmailBuilder:
                     'models': {},
                 }
 
-            if model_name not in general_events[level]['models']:
-                general_events[level]['models'][model_name] = {
+            # Type assertion to help the type checker understand the structure
+            level_data = general_events[level]
+            models_dict = level_data['models']
+
+            if model_name not in models_dict:
+                models_dict[model_name] = {
                     'model_name': model_name,
                     'events': [],
                 }
 
-            general_events[level]['models'][model_name]['events'].append(report)
+            models_dict[model_name]['events'].append(report)
 
         for _type, content in general_events.items():
             content['models'] = list(content['models'].values())