alita-sdk 0.3.226__tar.gz → 0.3.228__tar.gz

{alita_sdk-0.3.226 → alita_sdk-0.3.228}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.226
+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-0.3.226 → 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-0.3.226 → 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-0.3.226 → 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-0.3.226 → 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-0.3.226 → alita_sdk-0.3.228}/alita_sdk/tools/bitbucket/api_wrapper.py

@@ -23,7 +23,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
     """Wrapper for Bitbucket API."""
 
     _bitbucket: Any = PrivateAttr()
-    active_branch: Any = PrivateAttr()
+    _active_branch: Any = PrivateAttr()
     url: str = ''
     project: str = ''
     """The key of the project this repo belongs to"""
@@ -78,12 +78,12 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
             project=values['project'],
             repository=values['repository']
         )
-        cls.active_branch = values.get('branch')
+        cls._active_branch = values.get('branch')
         return values
 
     def set_active_branch(self, branch: str) -> None:
         """Set the active branch for the bot."""
-        self.active_branch = branch
+        self._active_branch = branch
         return f"Active branch set to `{branch}`"
 
     def list_branches_in_repo(self) -> List[str]:
@@ -93,15 +93,15 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
     def create_branch(self, branch_name: str) -> None:
         """Create a new branch in the repository."""
         try:
-            self._bitbucket.create_branch(branch_name, self.active_branch)
+            self._bitbucket.create_branch(branch_name, self._active_branch)
         except Exception as e:
             if "not permitted to access this resource" in str(e):
                 return f"Please, verify you token/password: {str}"
             if "already exists" in str(e):
-                self.active_branch = branch_name
+                self._active_branch = branch_name
                 return f"Branch {branch_name} already exists. set it as active"
             return f"Unable to create branch due to error:\n{e}"
-        self.active_branch = branch_name
+        self._active_branch = branch_name
         return f"Branch {branch_name} created successfully and set as active"
 
     def create_pull_request(self, pr_json_data: str) -> str:
@@ -161,7 +161,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
             return result if isinstance(result, ToolException) else f"File has been updated: {file_path}."
         except Exception as e:
             return ToolException(f"File was not updated due to error: {str(e)}")
-    
+
     def get_pull_requests_commits(self, pr_id: str) -> List[Dict[str, Any]]:
         """
         Get commits from a pull request
@@ -175,7 +175,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
             return result
         except Exception as e:
             return ToolException(f"Can't get commits from pull request `{pr_id}` due to error:\n{str(e)}")
-        
+
     def get_pull_requests(self) -> List[Dict[str, Any]]:
         """
         Get pull requests from the repository
@@ -183,7 +183,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
             List[Dict[str, Any]]: List of pull requests in the repository
         """
         return self._bitbucket.get_pull_requests()
-    
+
     def get_pull_request(self, pr_id: str) -> Dict[str, Any]:
         """
         Get details of a pull request
@@ -196,7 +196,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
             return self._bitbucket.get_pull_request(pr_id=pr_id)
         except Exception as e:
             return ToolException(f"Can't get pull request `{pr_id}` due to error:\n{str(e)}")
-        
+
     def get_pull_requests_changes(self, pr_id: str) -> Dict[str, Any]:
         """
         Get changes of a pull request
@@ -209,7 +209,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
             return self._bitbucket.get_pull_requests_changes(pr_id=pr_id)
         except Exception as e:
             return ToolException(f"Can't get changes from pull request `{pr_id}` due to error:\n{str(e)}")
-        
+
     def add_pull_request_comment(self, pr_id: str, content, inline=None) -> str:
         """
         Add a comment to a pull request. Supports multiple content types and inline comments.
@@ -234,7 +234,7 @@ class BitbucketAPIWrapper(BaseCodeToolApiWrapper):
         Returns:
             str: List of the files
         """
-        return str(self._bitbucket.get_files_list(file_path=path if path else '', branch=branch if branch else self.active_branch))
+        return str(self._bitbucket.get_files_list(file_path=path if path else '', branch=branch if branch else self._active_branch))
 
     # TODO: review this method, it may not work as expected
     # def _file_commit_hash(self, file_path: str, branch: str):

{alita_sdk-0.3.226 → 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-0.3.226 → 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,
@@ -555,7 +555,7 @@ class BaseCodeToolApiWrapper(BaseVectorStoreToolApiWrapper):
         """
         from .chunkers.code.codeparser import parse_code_files_for_db
 
-        _files = self.__handle_get_files("", branch or self.active_branch)
+        _files = self.__handle_get_files("", branch or self.active_branch or self._active_branch)
 
         logger.info(f"Files in branch: {_files}")
 
@@ -573,8 +573,8 @@ class BaseCodeToolApiWrapper(BaseVectorStoreToolApiWrapper):
             for file in _files:
                 if is_whitelisted(file) and not is_blacklisted(file):
                     yield {"file_name": file,
-                           "file_content": self._read_file(file, branch=branch or self.active_branch),
-                           "commit_hash": self._file_commit_hash(file, branch=branch or self.active_branch)}
+                           "file_content": self._read_file(file, branch=branch or self.active_branch or self._active_branch),
+                           "commit_hash": self._file_commit_hash(file, branch=branch or self.active_branch or self._active_branch)}
 
         return parse_code_files_for_db(file_content_generator())
     

{alita_sdk-0.3.226 → alita_sdk-0.3.228}/alita_sdk/tools/gitlab/api_wrapper.py

@@ -104,7 +104,7 @@ class GitLabAPIWrapper(BaseCodeToolApiWrapper):
     branch: Optional[str] = 'main'
     _git: Any = PrivateAttr()
     _repo_instance: Any = PrivateAttr()
-    active_branch: Any = PrivateAttr()
+    _active_branch: Any = PrivateAttr()
 
     llm: Optional[Any] = None
     # Alita instance
@@ -138,11 +138,11 @@ class GitLabAPIWrapper(BaseCodeToolApiWrapper):
         g.auth()
         cls._repo_instance = g.projects.get(values.get('repository'))
         cls._git = g
-        cls.active_branch = values.get('branch')
+        cls._active_branch = values.get('branch')
         return values
 
     def set_active_branch(self, branch_name: str) -> str:
-        self.active_branch = branch_name
+        self._active_branch = branch_name
         self._repo_instance.default_branch = branch_name
         return f"Active branch set to {branch_name}"
 
@@ -172,19 +172,19 @@ class GitLabAPIWrapper(BaseCodeToolApiWrapper):
             return f"Failed to list branches: {str(e)}"
 
     def list_files(self, path: str = None, recursive: bool = True, branch: str = None) -> List[str]:
-        branch = branch if branch else self.active_branch
+        branch = branch if branch else self._active_branch
         files = self._get_all_files(path, recursive, branch)
         paths = [file['path'] for file in files if file['type'] == 'blob']
         return paths
 
     def list_folders(self, path: str = None, recursive: bool = True, branch: str = None) -> List[str]:
-        branch = branch if branch else self.active_branch
+        branch = branch if branch else self._active_branch
         files = self._get_all_files(path, recursive, branch)
         paths = [file['path'] for file in files if file['type'] == 'tree']
         return paths
 
     def _get_all_files(self, path: str = None, recursive: bool = True, branch: str = None):
-        branch = branch if branch else self.active_branch
+        branch = branch if branch else self._active_branch
         return self._repo_instance.repository_tree(path=path, ref=branch, recursive=recursive, all=True)
 
     # overrided for indexer
@@ -210,15 +210,15 @@ class GitLabAPIWrapper(BaseCodeToolApiWrapper):
             self._repo_instance.branches.create(
                 {
                     'branch': branch_name,
-                    'ref': self.active_branch,
+                    'ref': self._active_branch,
                 }
             )
         except Exception as e:
             if "Branch already exists" in str(e):
-                self.active_branch = branch_name
+                self._active_branch = branch_name
                 return f"Branch {branch_name} already exists. set it as active"
             return f"Unable to create branch due to error:\n{e}"
-        self.active_branch = branch_name
+        self._active_branch = branch_name
         return f"Branch {branch_name} created successfully and set as active"
 
     def parse_issues(self, issues: List[Any]) -> List[dict]:

{alita_sdk-0.3.226 → 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-0.3.226 → alita_sdk-0.3.228}/alita_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.226
+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-0.3.226 → alita_sdk-0.3.228}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "alita_sdk"
-version = "0.3.226"
+version = "0.3.228"
 description = "SDK for building langchain agents using resources from Alita"
 readme = "README.md"
 requires-python = ">=3.10"