alita-sdk 0.3.376__tar.gz → 0.3.378__tar.gz

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.376
+Version: 0.3.378
 Summary: SDK for building langchain agents using resources from Alita
 Author-email: Artem Rozumenko <artyom.rozumenko@gmail.com>, Mikalai Biazruchka <mikalai_biazruchka@epam.com>, Roman Mitusov <roman_mitusov@epam.com>, Ivan Krakhmaliuk <lifedj27@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/runtime/langchain/langraph_agent.py

@@ -556,11 +556,11 @@ def create_graph(
             elif node_type == 'code':
                 from ..tools.sandbox import create_sandbox_tool
                 sandbox_tool = create_sandbox_tool(stateful=False, allow_net=True)
-                code = node.get('code', "return 'Code block is empty'")
+                code_data = node.get('code', {'type': 'fixed', 'value': "return 'Code block is empty'"})
                 lg_builder.add_node(node_id, FunctionTool(
                     tool=sandbox_tool, name=node['id'], return_type='dict',
                     output_variables=node.get('output', []),
-                    input_mapping={'code': {'type': 'fixed', 'value': code}},
+                    input_mapping={'code': code_data},
                     input_variables=node.get('input', ['messages']),
                     structured_output=node.get('structured_output', False),
                     alita_client=kwargs.get('alita_client', None)

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/runtime/langchain/utils.py

@@ -177,7 +177,11 @@ def propagate_the_input_mapping(input_mapping: dict[str, dict], input_variables:
             var_dict = create_params(input_variables, source)
 
         if value['type'] == 'fstring':
-            input_data[key] = value['value'].format(**var_dict)
+            try:
+                input_data[key] = value['value'].format(**var_dict)
+            except KeyError as e:
+                logger.error(f"KeyError in fstring formatting for key '{key}'. Attempt to find proper data in state.\n{e}")
+                input_data[key] = value['value'].format(**state)
         elif value['type'] == 'fixed':
             input_data[key] = value['value']
         else:

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/runtime/tools/function.py

@@ -31,13 +31,10 @@ class FunctionTool(BaseTool):
         """Prepare input for PyodideSandboxTool by injecting state into the code block."""
         # add state into the code block here since it might be changed during the execution of the code
         state_copy = deepcopy(state)
-        # pickle state
-        import pickle
 
         del state_copy['messages']  # remove messages to avoid issues with pickling without langchain-core
-        serialized_state = pickle.dumps(state_copy)
         # inject state into the code block as alita_state variable
-        pyodide_predata = f"""import pickle\nalita_state = pickle.loads({serialized_state})\n"""
+        pyodide_predata = f"alita_state = {state_copy}"
         # add classes related to sandbox client
         # read the content of alita_sdk/runtime/cliens/sandbox_client.py
         try:
@@ -64,6 +61,10 @@ class FunctionTool(BaseTool):
 
         if self.output_variables:
             for var in self.output_variables:
+                if var == "messages":
+                    tool_result_converted.update(
+                        {"messages": [{"role": "assistant", "content": dumps(tool_result)}]})
+                    continue
                 if isinstance(tool_result, dict) and var in tool_result:
                     tool_result_converted[var] = tool_result[var]
                 else:

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/runtime/tools/vectorstore.py

@@ -215,7 +215,7 @@ class VectorStoreWrapper(BaseToolApiWrapper):
         """List all collections in the vectorstore.
         Returns a list of collection names, or if no collections exist,
         returns a dict with an empty list and a message."""
-        raw = self.vector_adapter.list_indexes(self)
+        raw = self.vector_adapter.list_collections(self)
         # Normalize raw result to a list of names
         if not raw:
             # No collections found

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/runtime/tools/vectorstore_base.py

@@ -208,10 +208,10 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
             logger.error(f"Error during similarity search: {str(e)}")
             raise ToolException(f"Search failed: {str(e)}")
 
-    def list_indexes(self) -> List[str]:
+    def list_collections(self) -> List[str]:
         """List all collections in the vectorstore."""
 
-        collections = self.vector_adapter.list_indexes(self)
+        collections = self.vector_adapter.list_collections(self)
         if not collections:
             return "No indexed collections"
         return collections

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/runtime/utils/utils.py

@@ -14,6 +14,7 @@ class IndexerKeywords(Enum):
     INDEX_META_TYPE = 'index_meta'
     INDEX_META_IN_PROGRESS = 'in_progress'
     INDEX_META_COMPLETED = 'completed'
+    INDEX_META_FAILED = 'failed'
 
 # This pattern matches characters that are NOT alphanumeric, underscores, or hyphens
 clean_string_pattern = re.compile(r'[^a-zA-Z0-9_.-]')

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/tools/base_indexer_toolkit.py

@@ -152,39 +152,43 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
 
     def index_data(self, **kwargs):
         index_name = kwargs.get("index_name")
-        progress_step = kwargs.get("progress_step")
         clean_index = kwargs.get("clean_index")
         chunking_tool = kwargs.get("chunking_tool")
         chunking_config = kwargs.get("chunking_config")
+        result = {"count": 0}
         #
-        if clean_index:
-            self._clean_index(index_name)
-        #
-        self.index_meta_init(index_name, kwargs)
-        #
-        self._log_tool_event(f"Indexing data into collection with suffix '{index_name}'. It can take some time...")
-        self._log_tool_event(f"Loading the documents to index...{kwargs}")
-        documents = self._base_loader(**kwargs)
-        documents = list(documents) # consume/exhaust generator to count items
-        documents_count = len(documents)
-        documents = (doc for doc in documents)
-        self._log_tool_event(f"Base documents were pre-loaded. "
-                             f"Search for possible document duplicates and remove them from the indexing list...")
-        documents = self._reduce_duplicates(documents, index_name)
-        self._log_tool_event(f"Duplicates were removed. "
-                             f"Processing documents to collect dependencies and prepare them for indexing...")
-        result = self._save_index_generator(documents, documents_count, chunking_tool, chunking_config, index_name=index_name, progress_step=progress_step)
-        #
-        self.index_meta_update(index_name, IndexerKeywords.INDEX_META_COMPLETED.value, result)
-        #
-        return {"status": "ok", "message": f"successfully indexed {result} documents"}
+        try:
+            if clean_index:
+                self._clean_index(index_name)
+            #
+            self.index_meta_init(index_name, kwargs)
+            #
+            self._log_tool_event(f"Indexing data into collection with suffix '{index_name}'. It can take some time...")
+            self._log_tool_event(f"Loading the documents to index...{kwargs}")
+            documents = self._base_loader(**kwargs)
+            documents = list(documents) # consume/exhaust generator to count items
+            documents_count = len(documents)
+            documents = (doc for doc in documents)
+            self._log_tool_event(f"Base documents were pre-loaded. "
+                                 f"Search for possible document duplicates and remove them from the indexing list...")
+            documents = self._reduce_duplicates(documents, index_name)
+            self._log_tool_event(f"Duplicates were removed. "
+                                 f"Processing documents to collect dependencies and prepare them for indexing...")
+            self._save_index_generator(documents, documents_count, chunking_tool, chunking_config, index_name=index_name, result=result)
+            #
+            self.index_meta_update(index_name, IndexerKeywords.INDEX_META_COMPLETED.value, result["count"])
+            #
+            return {"status": "ok", "message": f"successfully indexed {result["count"]} documents"}
+        except Exception as e:
+            self.index_meta_update(index_name, IndexerKeywords.INDEX_META_FAILED.value, result["count"])
+            raise e
+            
 
-    def _save_index_generator(self, base_documents: Generator[Document, None, None], base_total: int, chunking_tool, chunking_config, index_name: Optional[str] = None, progress_step: int = 20):
+    def _save_index_generator(self, base_documents: Generator[Document, None, None], base_total: int, chunking_tool, chunking_config, result, index_name: Optional[str] = None):
         self._log_tool_event(f"Base documents are ready for indexing. {base_total} base documents in total to index.")
         from ..runtime.langchain.interfaces.llm_processor import add_documents
         #
         base_doc_counter = 0
-        total_counter = 0
         pg_vector_add_docs_chunk = []
         for base_doc in base_documents:
             base_doc_counter += 1
@@ -232,10 +236,9 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
             msg = f"Indexed base document #{base_doc_counter} out of {base_total} (with {dependent_docs_counter} dependencies)."
             logger.debug(msg)
             self._log_tool_event(msg)
-            total_counter += dependent_docs_counter
+            result["count"] += dependent_docs_counter
         if pg_vector_add_docs_chunk:
             add_documents(vectorstore=self.vectorstore, documents=pg_vector_add_docs_chunk)
-        return total_counter
 
     def _apply_loaders_chunkers(self, documents: Generator[Document, None, None], chunking_tool: str=None, chunking_config=None) -> Generator[Document, None, None]:
         from ..tools.chunkers import __all__ as chunkers
@@ -343,7 +346,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
         """Cleans the indexed data in the collection."""
         super()._clean_collection(index_name=index_name)
         return (f"Collection '{index_name}' has been removed from the vector store.\n"
-                f"Available collections: {self.list_indexes()}") if index_name \
+                f"Available collections: {self.list_collections()}") if index_name \
             else "All collections have been removed from the vector store." 
 
     def _build_collection_filter(self, filter: dict | str, index_name: str = "") -> dict:
@@ -385,7 +388,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
         """ Searches indexed documents in the vector store."""
         # build filter on top of index_name
 
-        available_collections = super().list_indexes()
+        available_collections = super().list_collections()
         if index_name and index_name not in available_collections:
             return f"Collection '{index_name}' not found. Available collections: {available_collections}"
 
@@ -547,10 +550,10 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
                 "args_schema": RemoveIndexParams
             },
             {
-                "name": "list_indexes",
-                "mode": "list_indexes",
-                "ref": self.list_indexes,
-                "description": self.list_indexes.__doc__,
+                "name": "list_collections",
+                "mode": "list_collections",
+                "ref": self.list_collections,
+                "description": self.list_collections.__doc__,
                 "args_schema": create_model("ListCollectionsParams")  # No parameters
             },
         ]

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/tools/elitea_base.py

@@ -17,7 +17,7 @@ from ..runtime.utils.utils import IndexerKeywords
 
 logger = logging.getLogger(__name__)
 
-INDEX_TOOL_NAMES = ['index_data', 'remove_index', 'list_indexes', 'search_index', 'stepback_search_index',
+INDEX_TOOL_NAMES = ['index_data', 'remove_index', 'list_collections', 'search_index', 'stepback_search_index',
                             'stepback_summary_index']
 
 LoaderSchema = create_model(
@@ -403,9 +403,9 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
         """Cleans the indexed data in the collection."""
         self._init_vector_store()._clean_collection(index_name=index_name)
         return (f"Collection '{index_name}' has been removed from the vector store.\n"
-                f"Available collections: {self.list_indexes()}")
+                f"Available collections: {self.list_collections()}")
 
-    def list_indexes(self):
+    def list_collections(self):
         """Lists all collections in the vector store."""
         vectorstore_wrapper = self._init_vector_store()
         return vectorstore_wrapper.list_collections()
@@ -537,10 +537,10 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
                 "args_schema": RemoveIndexParams
             },
             {
-                "name": "list_indexes",
-                "mode": "list_indexes",
-                "ref": self.list_indexes,
-                "description": self.list_indexes.__doc__,
+                "name": "list_collections",
+                "mode": "list_collections",
+                "ref": self.list_collections,
+                "description": self.list_collections.__doc__,
                 "args_schema": create_model("ListCollectionsParams")  # No parameters
             },
 

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk/tools/vector_adapters/VectorStoreAdapter.py

@@ -307,7 +307,7 @@ class ChromaAdapter(VectorStoreAdapter):
 
     def list_collections(self, vectorstore_wrapper) -> str:
         vector_client = vectorstore_wrapper.vectorstore._client
-        return ','.join([collection.name for collection in vector_client.list_indexes()])
+        return ','.join([collection.name for collection in vector_client.list_collections()])
 
     def remove_collection(self, vectorstore_wrapper, collection_name: str):
         vectorstore_wrapper.vectorstore.delete_collection()

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/alita_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.376
+Version: 0.3.378
 Summary: SDK for building langchain agents using resources from Alita
 Author-email: Artem Rozumenko <artyom.rozumenko@gmail.com>, Mikalai Biazruchka <mikalai_biazruchka@epam.com>, Roman Mitusov <roman_mitusov@epam.com>, Ivan Krakhmaliuk <lifedj27@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.376 → alita_sdk-0.3.378}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "alita_sdk"
-version = "0.3.376"
+version = "0.3.378"
 description = "SDK for building langchain agents using resources from Alita"
 readme = "README.md"
 requires-python = ">=3.10"