vectorwave 0.2.9__tar.gz → 0.3.0__tar.gz

{vectorwave-0.2.9 → vectorwave-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vectorwave
-Version: 0.2.9
+Version: 0.3.0
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
@@ -26,7 +26,10 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
 Project-URL: Repository, https://github.com/cozymori/vectorwave
 
-Need more information? Visit [here](https://cozymori.github.io/vectorwave-docs/)
+
+![VectorWave Logo](./docs_kr/images/vectorwave.png)
+
+Need more information? Visit [here](https://www.cozymori.net/vectorwave)
 
 # VectorWave
 ![Star](https://badgen.net/github/stars/cozymori/vectorwave)

{vectorwave-0.2.9 → vectorwave-0.3.0}/Readme.md

@@ -1,4 +1,7 @@
-Need more information? Visit [here](https://cozymori.github.io/vectorwave-docs/)
+
+![VectorWave Logo](./docs_kr/images/vectorwave.png)
+
+Need more information? Visit [here](https://www.cozymori.net/vectorwave)
 
 # VectorWave
 ![Star](https://badgen.net/github/stars/cozymori/vectorwave)

{vectorwave-0.2.9 → vectorwave-0.3.0}/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "maturin"
 
 [project]
 name = "vectorwave"
-version = "0.2.09"
+version = "0.3.0"
 authors = [
     { name = "junyeonggim", email = "junyeonggim5@gmail.com" },
 ]

{vectorwave-0.2.9 → vectorwave-0.3.0}/src/vectorwave/__init__.py

@@ -1,5 +1,5 @@
 from .core.decorator import vectorize
-from .database.db import initialize_database
+from .database.db import initialize_database, update_database_schema
 from .database.db_search import search_functions, search_executions, search_errors_by_message, search_functions_hybrid
 from .monitoring.tracer import trace_span
 from .search.rag_search import search_and_answer, analyze_trace_log
@@ -25,5 +25,6 @@ __all__ = [
     'VectorWaveReplayer',
     'SemanticReplayer',
     'VectorWaveDatasetManager',
-    'VectorWaveAutoInjector'
+    'VectorWaveAutoInjector',
+    'update_database_schema'
 ]

{vectorwave-0.2.9 → vectorwave-0.3.0}/src/vectorwave/batch/batch.py

@@ -27,11 +27,18 @@ class WeaviateBatchManager:
     Uses High-Performance Rust Core if available, otherwise falls back to Python.
     """
 
-    def __init__(self):
+    def __init__(self, host: Optional[str] = None, port: Optional[int] = None,
+                 grpc_port: Optional[int] = None, api_key: Optional[str] = None):
         self._initialized = False
         self.settings: WeaviateSettings = get_weaviate_settings()
         self.client: Optional[weaviate.WeaviateClient] = None
 
+        # Store dynamic connection params
+        self._host = host
+        self._port = port
+        self._grpc_port = grpc_port
+        self._api_key = api_key
+
         # Batch Configuration
         self.batch_threshold = self.settings.BATCH_THRESHOLD
         self.flush_interval = self.settings.FLUSH_INTERVAL_SECONDS
@@ -60,7 +67,13 @@ class WeaviateBatchManager:
     def _connect_client(self):
         """Attempts to connect to Weaviate."""
         try:
-            self.client = get_weaviate_client(self.settings)
+            if self._host is not None:
+                self.client = get_weaviate_client(
+                    host=self._host, port=self._port,
+                    grpc_port=self._grpc_port, api_key=self._api_key
+                )
+            else:
+                self.client = get_weaviate_client(self.settings)
             if self.client:
                 self._initialized = True
         except Exception as e:
@@ -171,6 +184,11 @@ class WeaviateBatchManager:
             except:
                 pass
 
-@lru_cache(None)
-def get_batch_manager() -> WeaviateBatchManager:
-    return WeaviateBatchManager()
+@lru_cache()
+def get_batch_manager(
+        host: Optional[str] = None,
+        port: Optional[int] = None,
+        grpc_port: Optional[int] = None,
+        api_key: Optional[str] = None
+) -> WeaviateBatchManager:
+    return WeaviateBatchManager(host=host, port=port, grpc_port=grpc_port, api_key=api_key)

{vectorwave-0.2.9 → vectorwave-0.3.0}/src/vectorwave/core/auto_injector.py

@@ -23,7 +23,7 @@ def create_smart_wrapper(original_func, root_wrapper):
         # Check for existing tracer at runtime
         tracer = current_tracer_var.get()
 
-        if tracer:
+        if tracer is not None:
             # [Case A] Parent trace exists -> Act as Child Span
             # (Wrap original function with trace_span and execute)
             return trace_span(original_func, capture_return_value=True)(*args, **kwargs)

{vectorwave-0.2.9 → vectorwave-0.3.0}/src/vectorwave/core/decorator.py

@@ -13,7 +13,7 @@ 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
-from ..utils.path_utils import get_repo_root_and_relative_path # New import
+from ..utils.path_utils import get_repo_root_and_relative_path
 
 logger = logging.getLogger(__name__)
 
@@ -31,6 +31,7 @@ def vectorize(search_description: Optional[str] = None,
               capture_inputs: bool = False,
               semantic_cache_filters: Optional[Dict[str, Any]] = None,
               semantic_cache_scope: Optional[List[str]] = None,
+              enable_alert: bool = True,
               **execution_tags):
     """
     VectorWave Decorator with Auto-Generation support.
@@ -65,7 +66,7 @@ def vectorize(search_description: Optional[str] = None,
                 if attr not in final_attributes:
                     final_attributes.append(attr)
 
-        if capture_inputs:
+        if capture_inputs or replay:
             try:
                 sig = inspect.signature(func)
                 for param_name in sig.parameters:
@@ -74,15 +75,6 @@ def vectorize(search_description: Optional[str] = None,
             except Exception as e:
                 logger.warning(f"Failed to inspect signature for auto-capture in '{function_name}': {e}")
 
-        if replay:
-            try:
-                sig = inspect.signature(func)
-                for param_name in sig.parameters:
-                    if param_name not in ('self', 'cls') and param_name not in final_attributes:
-                        final_attributes.append(param_name)
-            except Exception as e:
-                logger.warning(f"Failed to inspect signature for replay auto-capture in '{function_name}': {e}")
-
         # Extract Execution Tags
         valid_execution_tags = {}
         settings = get_weaviate_settings()
@@ -110,7 +102,7 @@ def vectorize(search_description: Optional[str] = None,
                 if relative_file_path:
                     file_path = relative_file_path
                 else:
-                    file_path = abs_file_path # Fallback to absolute path if not in a git repo
+                    file_path = abs_file_path
                     logger.warning(f"Function '{function_name}' is not in a Git repository. "
                                    f"PR creation might fail for absolute path: {file_path}")
             except Exception as e:
@@ -147,7 +139,7 @@ def vectorize(search_description: Optional[str] = None,
                     vectorizer = get_vectorizer()
                     vector_to_add = None
 
-                    if vectorizer and search_description:
+                    if vectorizer is not None and search_description:
                         try:
                             vector_to_add = vectorizer.embed(search_description)
                         except Exception as e:
@@ -185,9 +177,32 @@ def vectorize(search_description: Optional[str] = None,
 
         # --- Wrapper Logic ---
 
+        def _try_cache(args, kwargs):
+            """Check semantic cache. Returns cached result or None."""
+            if not semantic_cache:
+                return None
+            filters = resolve_semantic_filters(args, kwargs)
+            return _check_and_return_cached_result(
+                func, args, kwargs, function_name, cache_threshold,
+                is_async_func, filters=filters
+            )
+
+        def _build_full_kwargs(kwargs):
+            """Build kwargs with execution tags and metadata."""
+            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 full_kwargs
+
         if is_async_func:
             @trace_root()
-            @trace_span(attributes_to_capture=final_attributes, capture_return_value=capture_return_value)
+            @trace_span(
+                attributes_to_capture=final_attributes,
+                capture_return_value=capture_return_value,
+                force_sync=semantic_cache,
+                enable_alert=enable_alert
+            )
             @wraps(func)
             async def inner_wrapper(*args, **kwargs):
                 clean_kwargs = {k: v for k, v in kwargs.items() if
@@ -196,23 +211,22 @@ def vectorize(search_description: Optional[str] = None,
 
             @wraps(func)
             async def outer_wrapper(*args, **kwargs):
-                if semantic_cache:
-                    filters = resolve_semantic_filters(args, kwargs)
-                    cached = _check_and_return_cached_result(func, args, kwargs, function_name, cache_threshold, True, filters=filters)
-                    if cached is not None: return cached
-
-                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)
+                cached = _try_cache(args, kwargs)
+                if cached is not None:
+                    return cached
+                return await inner_wrapper(*args, **_build_full_kwargs(kwargs))
 
             outer_wrapper._is_vectorized = True
             return outer_wrapper
 
         else:  # Sync wrapper
             @trace_root()
-            @trace_span(attributes_to_capture=final_attributes, capture_return_value=capture_return_value)
+            @trace_span(
+                attributes_to_capture=final_attributes,
+                capture_return_value=capture_return_value,
+                force_sync=semantic_cache,
+                enable_alert=enable_alert
+            )
             @wraps(func)
             def inner_wrapper(*args, **kwargs):
                 clean_kwargs = {k: v for k, v in kwargs.items() if
@@ -221,16 +235,10 @@ def vectorize(search_description: Optional[str] = None,
 
             @wraps(func)
             def outer_wrapper(*args, **kwargs):
-                if semantic_cache:
-                    filters = resolve_semantic_filters(args, kwargs)
-                    cached = _check_and_return_cached_result(func, args, kwargs, function_name, cache_threshold, False, filters=filters)
-                    if cached is not None: return cached
-
-                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)
+                cached = _try_cache(args, kwargs)
+                if cached is not None:
+                    return cached
+                return inner_wrapper(*args, **_build_full_kwargs(kwargs))
 
             outer_wrapper._is_vectorized = True
             return outer_wrapper

{vectorwave-0.2.9 → vectorwave-0.3.0}/src/vectorwave/core/generator.py

@@ -21,7 +21,7 @@ def generate_metadata_via_llm(source_code: str, func_name: str) -> Optional[Dict
     """Call LLM to generate description and narrative from source code."""
     settings = get_weaviate_settings()
     client = get_llm_client()
-    if not client:
+    if client is None:
         return None
 
     prompt = f"""
@@ -121,7 +121,7 @@ def generate_and_register_metadata():
 
         # 4. Vectorize the Description
         vector_to_add = None
-        if vectorizer and final_desc:
+        if vectorizer is not None and final_desc:
             try:
                 vector_to_add = vectorizer.embed(final_desc)
             except Exception as e:

vectorwave-0.3.0/src/vectorwave/core/initializer.py

@@ -0,0 +1,37 @@
+import threading
+import logging
+from ..models.db_config import get_weaviate_settings
+from ..utils.scheduler import start_scheduler
+
+logger = logging.getLogger(__name__)
+
+_HEALER_STARTED = False
+
+def initialize_vectorwave():
+    """
+    Initialize vectorwave system.
+    Refer to config automatically run Auto Healer System.
+    """
+    global _HEALER_STARTED
+    try:
+        settings = get_weaviate_settings()
+    except Exception as e:
+        logger.warning(f"⚠️ Failed to load settings during initialization: {e}")
+        return
+
+    if _HEALER_STARTED:
+        return
+
+    if settings.ENABLE_AUTO_HEALER:
+        logger.info("🔧 AutoHealer is enabled. Starting background scheduler...")
+
+        healer_thread = threading.Thread(
+            target=start_scheduler,
+            kwargs={'interval_minutes': settings.HEALER_CHECK_INTERVAL_MINUTES},
+            daemon=True,
+            name="VectorWave-AutoHealer"
+        )
+        healer_thread.start()
+        _HEALER_STARTED = True
+    else:
+        logger.debug("🔧 AutoHealer is disabled (ENABLE_AUTO_HEALER=False).")

vectorwave-0.3.0/src/vectorwave/core/llm/factory.py

@@ -0,0 +1,10 @@
+from functools import lru_cache
+
+from .base import BaseLLMClient
+from .openai_client import VectorWaveOpenAIClient
+
+
+@lru_cache()
+def get_llm_client() -> BaseLLMClient:
+    """Returns the singleton LLM client instance."""
+    return VectorWaveOpenAIClient()

{vectorwave-0.2.9 → vectorwave-0.3.0}/src/vectorwave/database/dataset.py

@@ -37,7 +37,7 @@ class VectorWaveDatasetManager:
                 include_vector=True
             )
 
-            if not log_obj:
+            if log_obj is None:
                 logger.error(f"Log UUID '{log_uuid}' not found.")
                 return False
 
@@ -55,9 +55,17 @@ class VectorWaveDatasetManager:
             }
 
             # 3. Save (reuse original vector)
+            vector = (log_obj.vector or {}).get("default")
+            if vector is None:
+                logger.error(
+                    f"Log '{log_uuid}' has no stored vector. "
+                    "The source function must have capture_return_value=True for vector storage."
+                )
+                return False
+
             self.golden_col.data.insert(
                 properties=golden_props,
-                vector=log_obj.vector["default"],  # Copy vector
+                vector=vector,
                 uuid=generate_uuid5(log_uuid)  # Regenerate to avoid UUID collision, or maintain relation with original
             )
             logger.info(f"✅ Registered log {log_uuid} as Golden Data.")