o Zhy4@stdZddlmZddlZddlZddlZddlZddlZddlZddl m Z ddl m Z ddl m Z mZddlmZmZmZmZmZmZmZmZddlmZmZmZmZmZdd lmZm Z dd l!m"Z"m#Z#dd l$m%Z%dd l&m'Z'dd l(m)Z)ddl*m+Z+erddl,m-Z-ddl.m/Z/m0Z0e1e2Z3dZ4GdddeZ5GdddeZ6d!ddZ7d"ddZ8Gdd d e#eZ9dS)#1Base class for all loaders that uses O365 Package) annotationsN)abstractmethod)datetime)PathPurePath) TYPE_CHECKINGAnyDictIterableListOptionalSequenceUnion) BaseModelFieldFilePath PrivateAttr SecretStr) BaseSettingsSettingsConfigDict)BaseBlobParser BaseLoader)FileSystemBlobLoader)Blob)MimeTypeBasedParser) get_parser)Account)DriveFolderiPc@sFeZdZUedddZded<edddZded<ed d d d d ZdS) _O365Settings.ZO365_CLIENT_ID)aliasstr client_idZO365_CLIENT_SECRETr client_secretFz.envignore)case_sensitiveZenv_fileZ env_prefixextraN) __name__ __module__ __qualname__rr#__annotations__r$rZ model_configr-r-e/var/www/html/lang_env/lib/python3.10/site-packages/langchain_community/document_loaders/base_o365.pyr *s  r c@s&eZdZUeddZded<dS)_O365TokenStorage .credentialszo365_token.txtr token_pathN)r)r*r+rhomer1r,r-r-r-r.r/3s r/ file_types Sequence[str]returnDict[str, str]cCs@i}|D]}td|\}}|r|||<qtd||S)2Fetch the mime types for the specified file types.zfile.zUnknown mimetype of extension ) mimetypes guess_type ValueError)r3mime_types_mappingext mime_type_r-r-r.fetch_mime_types7s r? mime_typescCs>i}|D]}t|}|r|||dd<qtd||S)r7NzUnknown mimetype )r8guess_extensionr:)r@r;r=r<r-r-r.fetch_extensionsCs rCcseZdZUdZeedZded< dZded< e Z ded < dZ ded < d Z d ed < iZ ded< eZded<eZded<eZded<d.fdd Zed/ddZeed0dd Zd1d$d%Zd2d)d*Zd3d,d-ZZS)4O365BaseLoaderr)default_factoryr settingsFboolauth_with_tokenzUnion[int, str] chunk_size recursiveNzOptional[datetime]modified_sincezOptional[Dict[str, Any]]handlersr _blob_parserr4 _file_typesr6 _mime_typeskwargsr r5Nonec s tjdi|jrdtj}zt|_tt|_fddj D}Wn-t yZzt |_ttj_j}Wnt yWt d|dwYnwt |dd_ dStd_ tj t sytdtj t tj j_dS) Ncsi|] \}}j||qSr-rO).0 extensionhandlerselfr-r. s z+O365BaseLoader.__init__..z=`handlers` keys must be either file extensions or mimetypes. zj could not be interpreted as either. File extensions and mimetypes cannot mix. Use either one or the other)rLZfallback_parserdefaultzLget_parser("default) was supposed to return MimeTypeBasedParser.It returned r-)super__init__rLlistkeysr?rOsetrNitemsr:rCrrMr isinstance TypeErrortype)rWrPZ handler_keysZ mime_handlers __class__rVr.r[sD        zO365BaseLoader.__init__cCs|jS)zBReturn a dict of supported file types to corresponding mime types.rRrVr-r-r._fetch_mime_typessz O365BaseLoader._fetch_mime_types List[str]cCsdS)zReturn required scopes.Nr-rVr-r-r._scopesszO365BaseLoader._scopesfolderrIterable[Blob]c cs|j}|}i}t}tjtj|dd|D]U}|jrq|j t | vrq|j r3|j |j krq|j}td|jrJ|jjdtj|j}|j||jd||j t|jt|j t|jt|j|jt|jd||j<qt|d}|D]"} t | jt!st"d| jr|#t| jji} | j$%| | Vq{Wd n1swY|j&r|'D] } |(| Ed Hqd Sd S) aLazily load all files from a specified folder of the configured MIME type. Args: folder: The Folder instance from which the files are to be loaded. This Folder instance should represent a directory in a file system where the files are stored. Yields: An iterator that yields Blob instances, which are binary representations of the files loaded from the folder. T)exist_ok"Doc.aspx\?sourcedoc=.*file=([^&]+)/Zto_pathrIsourcer=createdmodified created_by modified_by descriptionidpath#Expected blob path to be a PurePathN))reZ get_itemstempfileTemporaryDirectoryosmakedirsrwdirnameis_filer=r\valuesrKrqweb_urlresearch_parenturllibparsequotenamedownloadrIr"rprrrsrt object_idr yield_blobsr`rNotImplementedErrorgetmetadataupdaterJZget_child_folders_load_from_folder) rWrhfile_mime_typesr_ metadata_dicttemp_dirfileroloaderblobfile_metadata_Z subfolderr-r-r.rs^        % z O365BaseLoader._load_from_folderdriver object_idsc csP|j}i}t}|D]\}||}|s#td|d|dq |jri|jt| vri|j }t d|j rF|j j dtj|j}|j||jd||j|j|jt|jt|j|jt|jd||j<q t|d} | D]"} t| jtst d | jr|!t| jji} | j"#| | VqsWd d S1swYd S) aLazily load files specified by their object_ids from a drive. Load files into the system as binary large objects (Blobs) and return Iterable. Args: drive: The Drive instance from which the files are to be loaded. This Drive instance should represent a cloud storage service or similar storage system where the files are stored. object_ids: A list of object_id strings. Each object_id represents a unique identifier for a file in the drive. Yields: An iterator that yields Blob instances, which are binary representations of the files loaded from the drive using the specified object_ids. z!There isn't a file withobject_id z in drive .rkrlrmrnrvrxN)$reryrzZget_itemloggingwarningr~r=r\rrrrrrrrrrrIrprqr"rrrsrtrrrr`rwrrrrr) rWrrrrrrrrorrrr-r-r._load_from_object_idss`       "z$O365BaseLoader._load_from_object_idsrcCsz ddlm}m}Wn tytdw|jr?t}|j}||j|jd}|d |j j |j j f|j |dddi}|S|tdd }|d |j j |j j f|j |dddi}||S) znAuthenticates the OneDrive API client Returns: The authenticated Account object. r)rFileSystemTokenBackendzAO365 package not found, please install it with `pip install o365`)r1Ztoken_filename) credentialsZscopes token_backendZraise_http_errorsFr0)r1Nr-)O365rr ImportErrorrHr/r1parentrrFr#r$Zget_secret_valuergrr2Z authenticate)rWrrZ token_storager1raccountr-r-r._auth$sJ     zO365BaseLoader._auth)rPr r5rQ)r5r6)r5rf)rhrr5ri)rrrrfr5ri)r5r)r)r*r+__doc__rr rFr,rH CHUNK_SIZErIrJrKrLrrMrNrOr[propertyrerrgrrr __classcell__r-r-rcr.rDOs4      &   8=rD)r3r4r5r6)r@r4r5r6):r __future__rrr8r{rryrabcrrpathlibrrtypingrr r r r r rrZpydanticrrrrrZpydantic_settingsrrZ)langchain_community.document_loaders.baserrZ=langchain_community.document_loaders.blob_loaders.file_systemrZ8langchain_community.document_loaders.blob_loaders.schemarZ4langchain_community.document_loaders.parsers.genericrZ5langchain_community.document_loaders.parsers.registryrrrZ O365.driverr getLoggerr)loggerrr r/r?rCrDr-r-r-r.s:   (