o |q iI$@sdZddlmZddlZddlmZddlmZmZm Z m Z m Z ddl Z ddl mZddlmZddlmZdd lmZGd d d eeZdS) z7Loader that loads data from Sharepoint Document Library) annotationsN)Path)AnyDictIteratorListOptional) BaseLoader)Document)Field)O365BaseLoaderc@seZdZUdZedZded< dZded< dZded < dZ ded < d Z d ed < e ddZ ded< d Zd ed< ed$ddZd%ddZd&ddZd'dd Zd(d"d#ZdS))SharePointLoaderzLoad from `SharePoint`..strdocument_library_idNz Optional[str] folder_pathzOptional[List[str]] object_ids folder_idFzOptional[bool] load_authz .credentialszo365_token.txtr token_pathload_extended_metadatareturn List[str]cCsddgS)zcReturn required scopes. Returns: List[str]: A list of required scopes. Z sharepointbasic)selfrrv/home/app/PaddleOCR-VL/.venv_paddleocr/lib/python3.10/site-packages/langchain_community/document_loaders/sharepoint.py_scopes%szSharePointLoader._scopesIterator[Document]c csz ddlm}m}Wn tytdw||j}t||s/t d|jd|j r| |j }t||sFt d|j d| |D]G}t |jd}|jdur_||}|jdurq||}|d |ji|j|D]} |jdur|| jd <|jdur| j|| VqwqK|jr||j}t||st d|j d| |D]G}t |jd}|jdur||}|jdur||}|d |ji|j|D]} |jdur|| jd <|jdur| j|| Vqq|jrH|||jD]E}t |jd}|jdur||}|jdur"||}|j|D]} |jdur5|| jd <|jdurA| j|| Vq(q|j s|js|js|}t||sbt d | |D][}t |jd}|jdur|||}|jdur||}|j|D]-} | j|j|jdur|| jd <|jdur| j|| jd |ji| Vqqgd Sd Sd Sd S) z Load documents lazily. Use this when working at a large scale. Yields: Document: A document object representing the parsed blob. r)DriveFolderzAO365 package not found, please install it with `pip install o365`zThere isn't a Drive with id .zThere isn't a folder with path idTZsource_full_urlauthorized_identitieszUnable to fetch root folderN)Z O365.driverr ImportErrorZ_authZstorageZ get_driver isinstance ValueErrorrZget_item_by_pathZ_load_from_folderrmetadatagetrr"rget_extended_metadataupdateZweb_urlZ _blob_parserZ lazy_parserZget_itemrZ_load_from_object_idsZget_root_folder) rrrdriveZ target_folderZblobfile_idZauth_identitiesZextended_metadataZ parsed_blobZ blob_partrrr lazy_load-s                                          zSharePointLoader.lazy_loadr+rc Cs|}|d}d|jd|d}dd|i}tjd||d}|}g}|d D]-} | d rZ| d d pK| d d pK| d d } | rZ| d} | rZ|| q-|S)a Retrieve the access identities (user/group emails) for a given file. Args: file_id (str): The ID of the file. Returns: List: A list of group names (email addresses) that have access to the file. access_token(https://graph.microsoft.com/v1.0/drives//items/z /permissions AuthorizationBearer GETheadersvalueZ grantedToV2ZsiteUserusergroupemail)_fetch_access_tokenr'rrequestsrequestjsonappend) rr+datar-urlr4responseZ access_listZ group_namesZ access_dataZ site_datar8rrrr"s2    z&SharePointLoader.authorized_identitiesrcCsDt|jdd }|}Wdn1swYt|}|S)z| Fetch the access token from the token file. Returns: The access token as a dictionary. zutf-8)encodingN)openrreadr<loads)rfsr>rrrr9s   z$SharePointLoader._fetch_access_tokenrc Cs|}|d}d|jd|d}dd|i}tjd||d}|}|d d |d id id d|diddddd|ddd}|S)a Retrieve extended metadata for a file in SharePoint. As of today, following fields are supported in the extended metadata: - size: size of the source file. - owner: display name of the owner of the source file. - full_path: pretty human readable path of the source file. Args: file_id (str): The ID of the file. Returns: dict: A dictionary containing the extended metadata of the file, including size, owner, and full path. r-r.r/z,?$select=size,createdBy,parentReference,namer0r1r2r3sizerZ createdByr6Z displayNameZparentReferencepath:/name)rGowner full_path)r9r'rr:r;r<split) rr+r>r-r?r4r@r&Zstaged_metadatarrrr(s2      z&SharePointLoader.get_extended_metadata)rr)rr)r+rrr)rr)r+rrr)__name__ __module__ __qualname____doc__r r__annotations__rrrrrhomerrpropertyrr,r"r9r(rrrrr s,         T " r )rT __future__rr<pathlibrtypingrrrrrr:Zlangchain_core.document_loadersr Zlangchain_core.documentsr Zpydanticr Z.langchain_community.document_loaders.base_o365r r rrrrs