alita-sdk 0.3.227__tar.gz → 0.3.228__tar.gz

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.227
+Version: 0.3.228
 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 <lifedjik@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/runtime/toolkits/tools.py

@@ -75,7 +75,7 @@ def get_tools(tools_list: list, alita_client, llm, memory_store: BaseStore = Non
                 llm=tool['settings'].get('llm'),
                 # indexer settings
                 connection_string=tool['settings'].get('connection_string', None),
-                collection_name=f"{tool.get('toolkit_name')}_{str(tool['id'])}",
+                collection_name=tool.get('toolkit_name'),
                 embedding_model=tool['settings'].get('embedding_model', None),
                 embedding_model_params=tool['settings'].get('embedding_model_params', None),
                 vectorstore_type="PGVector"

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/tools/ado/__init__.py

@@ -15,13 +15,13 @@ def get_tools(tool_type, tool):
         "limit": tool['settings'].get('limit', 5),
         "toolkit_name": tool.get('toolkit_name', ''),
         # indexer settings
-        "llm":tool['settings'].get('llm', None),
-        "connection_string":tool['settings'].get('connection_string', None),
-        "collection_name":str(tool['id']),
-        "doctype":'doc',
-        "embedding_model":"HuggingFaceEmbeddings",
-        "embedding_model_params":{"model_name": "sentence-transformers/all-MiniLM-L6-v2"},
-        "vectorstore_type":"PGVector"
+        "llm": tool['settings'].get('llm', None),
+        "connection_string": tool['settings'].get('connection_string', None),
+        "collection_name": tool['toolkit_name'],
+        "doctype": 'doc',
+        "embedding_model": "HuggingFaceEmbeddings",
+        "embedding_model_params": {"model_name": "sentence-transformers/all-MiniLM-L6-v2"},
+        "vectorstore_type": "PGVector"
     }
     if tool_type == 'ado_plans':
         return AzureDevOpsPlansToolkit().get_toolkit(**config_dict).get_tools()

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/tools/ado/repos/__init__.py

@@ -23,7 +23,7 @@ def _get_toolkit(tool) -> BaseToolkit:
         active_branch=tool['settings'].get('active_branch', ""),
         toolkit_name=tool['settings'].get('toolkit_name', ""),
         connection_string=tool['settings'].get('connection_string', None),
-        collection_name=str(tool['id']),
+        collection_name=tool['toolkit_name'],
         doctype='code',
         embedding_model="HuggingFaceEmbeddings",
         embedding_model_params={"model_name": "sentence-transformers/all-MiniLM-L6-v2"},

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/tools/ado/wiki/ado_wrapper.py

@@ -1,3 +1,4 @@
+import hashlib
 import logging
 from typing import Any, Dict, Generator, Optional
 
@@ -229,21 +230,26 @@ class AzureDevOpsApiWrapper(BaseVectorStoreToolApiWrapper):
             logger.error(f"Unable to modify wiki page: {str(e)}")
             return ToolException(f"Unable to modify wiki page: {str(e)}")
 
-    def _base_loader(self, wiki_identifier: str, **kwargs) -> Generator[Document, None, None]:
+    def _base_loader(self, wiki_identifier: str, title_contains: Optional[str] = None, **kwargs) -> Generator[Document, None, None]:
         pages = self._client.get_pages_batch(pages_batch_request={}, project=self.project, wiki_identifier=wiki_identifier)
         #
         for page in pages:
             content = self._client.get_page_by_id(project=self.project, wiki_identifier=wiki_identifier, id=page.id, include_content=True).page.content
-            yield Document(page_content=content, metadata={
-                'id': page.id, 
-                'path': page.path,
-                'updated_on': ''
-            })
+            content_hash = hashlib.sha256(content.encode("utf-8")).hexdigest()
+            title = page.path.rsplit("/", 1)[-1]
+            if not title_contains or (title_contains and title_contains.lower() in title.lower()):
+                yield Document(page_content=content, metadata={
+                    'id': str(page.id),
+                    'path': page.path,
+                    'title': title,
+                    'updated_on': content_hash
+                })
 
     def _index_tool_params(self):
         """Return the parameters for indexing data."""
         return {
-            "wiki_identifier": (str, Field(description="Wiki identifier to index, e.g., 'ABCProject.wiki'"))
+            "wiki_identifier": (str, Field(description="Wiki identifier to index, e.g., 'ABCProject.wiki'")),
+            'title_contains': (Optional[str], Field(default=None, description="Optional filter to include only pages with titles containing exact this string"))
         }
 
     @extend_with_vector_tools

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/tools/confluence/api_wrapper.py

@@ -214,8 +214,11 @@ class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
                 "`atlassian` package not found, please run "
                 "`pip install atlassian-python-api`"
             )
-
+        if not values.get('base_url'):
+            raise ValueError("Base URL is required for Confluence API Wrapper.")
         url = values['base_url']
+        if not (values.get('token') or (values.get('api_key') and values.get('username'))):
+            raise ValueError("Either 'token' or both 'api_key' and 'username' must be provided for authentication.")
         api_key = values.get('api_key')
         username = values.get('username')
         token = values.get('token')
@@ -844,8 +847,12 @@ class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
             yield document
 
     def _process_document(self, document: Document) -> Generator[Document, None, None]:
-        for attachment in self.get_page_attachments(document.metadata.get('id')):
-            yield Document(page_content=attachment.get('content', ''), metadata=attachment.get('metadata', {}))
+        attachments = self.get_page_attachments(document.metadata.get('id'))
+        if isinstance(attachments, str):
+            logger.info(f" {document.metadata.get('id')}: {attachments}")
+            return
+        for attachment in attachments:
+            yield Document(page_content=attachment.get('content', '') or attachment.get('llm_analysis', ''), metadata=attachment.get('metadata', {}))
 
     def _download_image(self, image_url):
         """
@@ -1577,7 +1584,7 @@ class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
         """Return the parameters for indexing data."""
         return {
             "content_format": (Literal['view', 'storage', 'export_view', 'editor', 'anonymous'],
-                            Field(description="The format of the content to be retrieved.")),
+                            Field(description="The format of the content to be retrieved.", default='view')),
             "page_ids": (Optional[List[str]], Field(description="List of page IDs to retrieve.", default=None)),
             "label": (Optional[str], Field(description="Label to filter pages.", default=None)),
             "cql": (Optional[str], Field(description="CQL query to filter pages.", default=None)),

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/tools/elitea_base.py

@@ -374,7 +374,7 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
     def list_collections(self):
         """Lists all collections in the vector store."""
         vectorstore_wrapper = self._init_vector_store()
-        return self._adapter.list_collections(vectorstore_wrapper)
+        return self._adapter.list_collections(vectorstore_wrapper, self.collection_name or "")
 
     def search_index(self,
                      query: str,

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk/tools/vector_adapters/VectorStoreAdapter.py

@@ -35,15 +35,16 @@ class PGVectorAdapter(VectorStoreAdapter):
             "connection_string": connection_string
         }
 
-    def list_collections(self, vectorstore_wrapper) -> str:
+    def list_collections(self, vectorstore_wrapper, collection_name) -> str:
         from sqlalchemy import text
         from sqlalchemy.orm import Session
 
         with Session(vectorstore_wrapper.vectorstore.session_maker.bind) as session:
-            get_collections = text("""
-                SELECT table_schema
-                FROM information_schema.columns
-                WHERE udt_name = 'vector';
+            get_collections = text(f"""
+                 SELECT table_schema
+                 FROM information_schema.columns
+                 WHERE udt_name = 'vector'
+                   AND table_schema LIKE '%{collection_name}%';
             """)
             result = session.execute(get_collections)
             docs = result.fetchall()

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/alita_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.227
+Version: 0.3.228
 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 <lifedjik@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.227 → alita_sdk-0.3.228}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "alita_sdk"
-version = "0.3.227"
+version = "0.3.228"
 description = "SDK for building langchain agents using resources from Alita"
 readme = "README.md"
 requires-python = ">=3.10"