vectorwave 0.2.2__tar.gz → 0.2.3__tar.gz

{vectorwave-0.2.2/src/vectorwave.egg-info → vectorwave-0.2.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vectorwave
-Version: 0.2.2
+Version: 0.2.3
 Summary: VectorWave: Seamless Auto-Vectorization Framework
 Author-email: junyeonggim <junyeonggim5@gmail.com>
 License-Expression: MIT

{vectorwave-0.2.2 → vectorwave-0.2.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "vectorwave"
-version = "0.2.02"
+version = "0.2.03"
 authors = [
     { name = "junyeonggim", email = "junyeonggim5@gmail.com" },
 ]

vectorwave-0.2.3/src/tests/utils/test_replay_source.py

@@ -0,0 +1,140 @@
+import pytest
+import sys
+from unittest.mock import MagicMock, patch
+from vectorwave import vectorize, search_executions
+from vectorwave.utils.replayer import VectorWaveReplayer
+from vectorwave.models.db_config import WeaviateSettings
+
+# 1. Dummy function for testing
+@vectorize(unique_id="test_replay_tagging")
+def calc_add(a: int, b: int):
+    """Simple addition function for testing replay tags."""
+    return a + b
+
+@pytest.fixture
+def mock_replay_deps(monkeypatch):
+    """Mocks dependencies for Replay Source testing."""
+    # Mock Settings
+    mock_settings = WeaviateSettings(
+        EXECUTION_COLLECTION_NAME="VectorWaveExecutions",
+        GOLDEN_COLLECTION_NAME="VectorWaveGoldenDataset"
+    )
+    mock_get_settings = MagicMock(return_value=mock_settings)
+
+    # Mock Batch Manager (to verify logs)
+    mock_batch = MagicMock()
+    mock_batch.add_object = MagicMock()
+    mock_get_batch = MagicMock(return_value=mock_batch)
+
+    # Mock DB Client (for Replayer initialization)
+    mock_client = MagicMock()
+    mock_collection = MagicMock()
+    mock_client.collections.get.return_value = mock_collection
+    mock_get_client = MagicMock(return_value=mock_client)
+
+    # --- [수정] Patch dependencies where they are USED, not just defined ---
+
+    # 1. Patch for Replayer (vectorwave/utils/replayer.py)
+    monkeypatch.setattr("vectorwave.utils.replayer.get_cached_client", mock_get_client)
+    monkeypatch.setattr("vectorwave.utils.replayer.get_weaviate_settings", mock_get_settings)
+
+    # 2. Patch for Database (vectorwave/database/db.py) - 안전장치
+    monkeypatch.setattr("vectorwave.database.db.get_cached_client", mock_get_client)
+
+    # 3. Patch for Config (vectorwave/models/db_config.py) - 안전장치
+    monkeypatch.setattr("vectorwave.models.db_config.get_weaviate_settings", mock_get_settings)
+
+    # 4. Patch batch manager for Decorator & Tracer
+    monkeypatch.setattr("vectorwave.core.decorator.get_batch_manager", mock_get_batch)
+    monkeypatch.setattr("vectorwave.monitoring.tracer.get_batch_manager", mock_get_batch)
+
+    # 5. Patch search_executions in THIS test file
+    mock_search = MagicMock()
+    monkeypatch.setattr(sys.modules[__name__], "search_executions", mock_search)
+
+    return {
+        "batch": mock_batch,
+        "search": mock_search,
+        "client": mock_client
+    }
+
+def test_execution_source_tagging(mock_replay_deps):
+    """
+    Verification Scenario:
+    1. Realtime Execution -> Check if saved to DB as 'REALTIME'
+    2. Replay Execution -> Check if saved to DB as 'REPLAY'
+    """
+    mock_batch = mock_replay_deps["batch"]
+
+    # ---------------------------------------------------------
+    # Case 1: Realtime Execution
+    # ---------------------------------------------------------
+    print("\n🚀 1. Running Realtime Execution...")
+    calc_add(10, 20)
+
+    # Check if log was sent to BatchManager
+    assert mock_batch.add_object.called
+
+    # Get the latest call args
+    args, kwargs = mock_batch.add_object.call_args
+    props = kwargs['properties']
+
+    print(f"   [Log Check] Source: {props.get('exec_source')}")
+
+    # Verify exec_source is 'REALTIME'
+    assert props.get("exec_source") == "REALTIME"
+    assert props.get("status") == "SUCCESS"
+
+    # ---------------------------------------------------------
+    # Case 2: Replay Execution
+    # ---------------------------------------------------------
+    print("\n🔄 2. Running Replay...")
+
+    # Prepare dummy log for Replayer to fetch
+    dummy_log_obj = MagicMock()
+    dummy_log_obj.uuid = "old-uuid-123"
+    dummy_log_obj.properties = {
+        "function_name": "calc_add",
+        "a": 10,
+        "b": 20,
+        "return_value": "30",
+        "exec_source": "REALTIME"
+    }
+
+    # Replayer uses client.collections.get().query.fetch_objects()
+    mock_collection = mock_replay_deps["client"].collections.get.return_value
+    mock_collection.query.fetch_objects.return_value.objects = [dummy_log_obj]
+
+    # Also need to mock import_module since Replayer imports the function dynamically
+    with patch("vectorwave.utils.replayer.importlib.import_module") as mock_import:
+        mock_module = MagicMock()
+        # Set the function on the mock module
+        setattr(mock_module, "calc_add", calc_add)
+        mock_import.return_value = mock_module
+
+        replayer = VectorWaveReplayer()
+
+        # Reset batch mock to capture only replay log
+        mock_batch.add_object.reset_mock()
+
+        # Execute Replay
+        result = replayer.replay(
+            function_full_name="src.tests.utils.test_replay_source.calc_add",
+            limit=1
+        )
+
+    assert result['total'] > 0, "Replay should have executed at least 1 test."
+
+    # Check if new log was sent to BatchManager
+    assert mock_batch.add_object.called
+
+    # Get the latest call args (from Replay execution)
+    args, kwargs = mock_batch.add_object.call_args
+    replay_props = kwargs['properties']
+
+    print(f"   [Log Check] Source: {replay_props.get('exec_source')}")
+
+    # Verify exec_source is 'REPLAY'
+    assert replay_props.get("exec_source") == "REPLAY"
+
+    print("\n✅ All tests passed! Execution Source tagging is working.")

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave/core/decorator.py

@@ -11,6 +11,7 @@ from ..monitoring.tracer import trace_root, trace_span
 from ..utils.function_cache import function_cache_manager
 from ..utils.return_caching_utils import _check_and_return_cached_result
 from ..vectorizer.factory import get_vectorizer
+from ..utils.context import execution_source_context
 
 logger = logging.getLogger(__name__)
 
@@ -53,7 +54,7 @@ def vectorize(search_description: Optional[str] = None,
         func_uuid = generate_uuid5(func_identifier)
 
         # Prepare attributes to capture
-        final_attributes = ['function_uuid', 'team', 'priority', 'run_id']
+        final_attributes = ['function_uuid', 'team', 'priority', 'run_id', 'exec_source']
         if attributes_to_capture:
             for attr in attributes_to_capture:
                 if attr not in final_attributes:
@@ -145,7 +146,7 @@ def vectorize(search_description: Optional[str] = None,
             async def inner_wrapper(*args, **kwargs):
                 # Remove injected tags from kwargs before calling original func
                 clean_kwargs = {k: v for k, v in kwargs.items() if
-                                k not in valid_execution_tags and k != 'function_uuid'}
+                                k not in valid_execution_tags and k != 'function_uuid' and k != 'exec_source'}
                 return await func(*args, **clean_kwargs)
 
             @wraps(func)
@@ -157,6 +158,7 @@ def vectorize(search_description: Optional[str] = None,
                 full_kwargs = kwargs.copy()
                 full_kwargs.update(valid_execution_tags)
                 full_kwargs['function_uuid'] = func_uuid
+                full_kwargs['exec_source'] = execution_source_context.get()
                 return await inner_wrapper(*args, **full_kwargs)
 
             return outer_wrapper
@@ -167,7 +169,7 @@ def vectorize(search_description: Optional[str] = None,
             @wraps(func)
             def inner_wrapper(*args, **kwargs):
                 clean_kwargs = {k: v for k, v in kwargs.items() if
-                                k not in valid_execution_tags and k != 'function_uuid'}
+                                k not in valid_execution_tags and k != 'function_uuid' and k != 'exec_source'}
                 return func(*args, **clean_kwargs)
 
             @wraps(func)
@@ -179,8 +181,9 @@ def vectorize(search_description: Optional[str] = None,
                 full_kwargs = kwargs.copy()
                 full_kwargs.update(valid_execution_tags)
                 full_kwargs['function_uuid'] = func_uuid
+                full_kwargs['exec_source'] = execution_source_context.get()
                 return inner_wrapper(*args, **full_kwargs)
 
             return outer_wrapper
 
-    return decorator
+    return decorator

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave/database/db.py

@@ -282,6 +282,11 @@ def create_execution_schema(client: weaviate.WeaviateClient, settings: WeaviateS
             name="return_value",
             data_type=wvc.DataType.TEXT,
             description="The JSON-serialized or str() representation of the function's return value"
+        ),
+        wvc.Property(
+            name="exec_source",
+            data_type=wvc.DataType.TEXT,
+            description="Source of execution: 'REALTIME' (User traffic) or 'REPLAY' (Regression Test)"
         )
     ]
 

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave/monitoring/tracer.py

@@ -15,6 +15,7 @@ from ..models.db_config import get_weaviate_settings, WeaviateSettings
 from .alert.factory import get_alerter
 from ..vectorizer.factory import get_vectorizer
 from ..database.db_search import check_semantic_drift
+from ..utils.context import execution_source_context
 
 # Create module-level logger
 logger = logging.getLogger(__name__)
@@ -137,7 +138,8 @@ def _create_span_properties(
         "status": status,
         "error_message": error_msg,
         "error_code": error_code,
-        "return_value": return_value_to_log
+        "return_value": return_value_to_log,
+        "exec_source": execution_source_context.get()
     }
 
     if tracer.settings.global_custom_values:

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave/search/execution_search.py

@@ -121,3 +121,34 @@ def find_by_trace_id(trace_id: str, limit: int = 100) -> List[Dict[str, Any]]:
         sort_ascending=True,  # Sort chronologically
         limit=limit
     )
+
+
+def find_replay_executions(
+        limit: int = 20,
+        status: Optional[str] = None,
+        function_name: Optional[str] = None
+) -> List[Dict[str, Any]]:
+    """
+    Searches specifically for logs generated during a Replay session.
+
+    Args:
+        limit: Max number of logs to retrieve.
+        status: Optional filter for 'SUCCESS' or 'ERROR'.
+        function_name: Optional filter for specific function name.
+    """
+    logger.info(f"\n--- Searching for REPLAY Executions (Status: {status}, Func: {function_name}) ---")
+
+    filters = {"exec_source": "REPLAY"}
+
+    if status:
+        filters["status"] = status
+
+    if function_name:
+        filters["function_name"] = function_name
+
+    return find_executions(
+        filters=filters,
+        sort_by="timestamp_utc",
+        sort_ascending=False,
+        limit=limit
+    )

vectorwave-0.2.3/src/vectorwave/utils/context.py

@@ -0,0 +1,4 @@
+from contextvars import ContextVar
+
+#default is realtime context
+execution_source_context: ContextVar[str] = ContextVar("execution_source", default="REALTIME")

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave/utils/replayer.py

@@ -13,6 +13,7 @@ import weaviate.classes.query as wvc_query
 from ..database.db import get_cached_client
 from ..models.db_config import get_weaviate_settings
 from ..monitoring.tracer import _mask_and_serialize
+from .context import execution_source_context
 
 logger = logging.getLogger(__name__)
 
@@ -78,7 +79,10 @@ class VectorWaveReplayer:
             # [FIX] Extract only valid arguments for the target function
             inputs = self._extract_inputs(raw_inputs, target_func)
 
+            token = None
             try:
+                token = execution_source_context.set("REPLAY")
+
                 # 3. Function Re-execution
                 if is_async_func:
                     actual_output = asyncio.run(target_func(**inputs))
@@ -127,6 +131,10 @@ class VectorWaveReplayer:
                     "traceback": traceback.format_exc()
                 })
 
+            finally:
+                if token:
+                    execution_source_context.reset(token)
+
         logger.info(f"Replay Finished. Passed: {results['passed']}, Failed: {results['failed']}")
         return results
 

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave/utils/replayer_semantic.py

@@ -6,6 +6,8 @@ import math
 import asyncio
 import inspect
 from typing import Any, Dict, List, Optional
+
+from .context import execution_source_context
 from ..core.llm.factory import get_llm_client
 
 import weaviate.classes.query as wvc_query
@@ -88,12 +90,16 @@ class SemanticReplayer(VectorWaveReplayer):
             inputs = self._extract_inputs(raw_inputs, target_func)
 
             try:
+                token = execution_source_context.set("REPLAY")
                 # 3. Function Re-execution
                 if is_async_func:
                     actual_output = asyncio.run(target_func(**inputs))
                 else:
                     actual_output = target_func(**inputs)
 
+                if token:
+                    execution_source_context.reset(token)
+
                 # 4. Result Validation (Semantic Comparison)
                 is_match, match_reason = self._compare_results_semantic(
                     expected_output,

{vectorwave-0.2.2 → vectorwave-0.2.3/src/vectorwave.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vectorwave
-Version: 0.2.2
+Version: 0.2.3
 Summary: VectorWave: Seamless Auto-Vectorization Framework
 Author-email: junyeonggim <junyeonggim5@gmail.com>
 License-Expression: MIT

{vectorwave-0.2.2 → vectorwave-0.2.3}/src/vectorwave.egg-info/SOURCES.txt

@@ -30,6 +30,7 @@ src/tests/search/__init__.py
 src/tests/search/test_execution_search.py
 src/tests/utils/__init__.py
 src/tests/utils/test_function_cahe.py
+src/tests/utils/test_replay_source.py
 src/tests/utils/test_replayer.py
 src/tests/utils/test_return_caching.py
 src/tests/utils/test_status.py
@@ -74,6 +75,7 @@ src/vectorwave/search/execution_search.py
 src/vectorwave/search/extended_search.py
 src/vectorwave/search/rag_search.py
 src/vectorwave/utils/__init__.py
+src/vectorwave/utils/context.py
 src/vectorwave/utils/function_cache.py
 src/vectorwave/utils/healer.py
 src/vectorwave/utils/replayer.py