skip null fields in telemetry request

Signed-off-by: Sai Shree Pradhan <saishree.pradhan@databricks.com>

Author: saishreeeee

src/databricks/sql/telemetry/models/event.py

@@ -9,7 +9,7 @@
     ExecutionResultFormat,
 )
 from typing import Optional
-from databricks.sql.telemetry.utils import EnumEncoder
+from databricks.sql.telemetry.utils import to_json_compact
 
 
 @dataclass
@@ -26,7 +26,7 @@ class HostDetails:
     port: int
 
     def to_json(self):
-        return json.dumps(asdict(self))
+        return to_json_compact(self)
 
 
 @dataclass
@@ -52,7 +52,7 @@ class DriverConnectionParameters:
     socket_timeout: Optional[int] = None
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -88,7 +88,7 @@ class DriverSystemConfiguration:
     locale_name: Optional[str] = None
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -106,7 +106,7 @@ class DriverVolumeOperation:
     volume_path: str
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -124,7 +124,7 @@ class DriverErrorInfo:
     stack_trace: str
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -146,7 +146,7 @@ class SqlExecutionEvent:
     retry_count: int
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -179,4 +179,4 @@ class TelemetryEvent:
     operation_latency_ms: Optional[int] = None
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)

src/databricks/sql/telemetry/models/frontend_logs.py

@@ -1,7 +1,7 @@
 import json
 from dataclasses import dataclass, asdict
 from databricks.sql.telemetry.models.event import TelemetryEvent
-from databricks.sql.telemetry.utils import EnumEncoder
+from databricks.sql.telemetry.utils import to_json_compact
 from typing import Optional
 
 
@@ -20,7 +20,7 @@ class TelemetryClientContext:
     user_agent: str
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -36,7 +36,7 @@ class FrontendLogContext:
     client_context: TelemetryClientContext
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -52,7 +52,7 @@ class FrontendLogEntry:
     sql_driver_log: TelemetryEvent
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)
 
 
 @dataclass
@@ -75,4 +75,4 @@ class TelemetryFrontendLog:
     workspace_id: Optional[int] = None
 
     def to_json(self):
-        return json.dumps(asdict(self), cls=EnumEncoder)
+        return to_json_compact(self)

src/databricks/sql/telemetry/utils.py

@@ -1,5 +1,6 @@
 import json
 from enum import Enum
+from dataclasses import asdict
 
 
 class EnumEncoder(json.JSONEncoder):
@@ -13,3 +14,23 @@ def default(self, obj):
         if isinstance(obj, Enum):
             return obj.value
         return super().default(obj)
+
+
+def filter_none_values(data):
+    """
+    Recursively remove None values from dictionaries.
+    This reduces telemetry payload size by excluding null fields.
+    """
+    if isinstance(data, dict):
+        return {k: filter_none_values(v) for k, v in data.items() if v is not None}
+    else:
+        return data
+
+
+def to_json_compact(dataclass_obj):
+    """
+    Convert a dataclass to JSON string, excluding None values.
+    """
+    data_dict = asdict(dataclass_obj)
+    filtered_dict = filter_none_values(data_dict)
+    return json.dumps(filtered_dict, cls=EnumEncoder)