a bgy4@sdZddlmZddlZddlZddlZddlZddlZddlZddl m Z ddl m Z ddl m Z mZddlmZmZmZmZmZmZmZmZddlmZmZmZmZmZdd lmZm Z dd l!m"Z"m#Z#dd l$m%Z%dd l&m'Z'dd l(m)Z)ddl*m+Z+erddl,m-Z-ddl.m/Z/m0Z0e1e2Z3dZ4GdddeZ5GdddeZ6dddddZ7dddddZ8Gddde#eZ9dS) 1Base class for all loaders that uses O365 Package) annotationsN)abstractmethod)datetime)PathPurePath) TYPE_CHECKINGAnyDictIterableListOptionalSequenceUnion) BaseModelFieldFilePath PrivateAttr SecretStr) BaseSettingsSettingsConfigDict)BaseBlobParser BaseLoader)FileSystemBlobLoader)Blob)MimeTypeBasedParser) get_parser)Account)DriveFolderiPc@sFeZdZUedddZded<edddZded<ed d d d d ZdS) _O365Settings.ZO365_CLIENT_ID)aliasstr client_idZO365_CLIENT_SECRETr client_secretFz.envignore)case_sensitiveZenv_fileZ env_prefixextraN) __name__ __module__ __qualname__rr#__annotations__r$rZ model_configr-r-|/var/www/html/cobodadashboardai.evdpl.com/venv/lib/python3.9/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/ Sequence[str]Dict[str, str]) file_typesreturncCsBi}|D]4}td|\}}|r.|||<qtd|q|S)2Fetch the mime types for the specified file types.zfile.zUnknown mimetype of extension ) mimetypes guess_type ValueError)r5mime_types_mappingext mime_type_r-r-r.fetch_mime_types7s r?) mime_typesr6cCs@i}|D]2}t|}|r,|||dd<qtd|q|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<dddfdd ZeddddZeedddd Zd!d"d#d$d%Zd&dd"d'd(d)Zd*dd+d,ZZS)-O365BaseLoaderr)default_factoryr settingsFboolauth_with_tokenzUnion[int, str] chunk_size recursiveNzOptional[datetime]modified_sincezOptional[Dict[str, Any]]handlersr _blob_parserr3 _file_typesr4 _mime_typesr None)kwargsr6c stjfi|jrtj}z4t|_tt|_fddj D}Wn^t yz(t |_ttj_j}Wn"t yt d|dYn0Yn0t |dd_ n@td_ tj t stdtj t tj j_dS)Ncsi|]\}}j||qSr-rO).0 extensionhandlerselfr-r. sz+O365BaseLoader.__init__..z=`handlers` keys must be either file extensions or mimetypes. zj could not be interpreted as either. File extensions and mimetypes cannot mix. Use either one or the other)rLZfallback_parserdefaultzLget_parser("default) was supposed to return MimeTypeBasedParser.It returned )super__init__rLlistkeysr?rOsetrNitemsr:rCrrMr isinstance TypeErrortype)rWrQZ handler_keysZ mime_handlers __class__rVr.r[s>          zO365BaseLoader.__init__)r6cCs|jS)zBReturn a dict of supported file types to corresponding mime types.rRrVr-r-r._fetch_mime_typessz O365BaseLoader._fetch_mime_typesz List[str]cCsdS)zReturn required scopes.Nr-rVr-r-r._scopesszO365BaseLoader._scopesrzIterable[Blob])folderr6c cs|j}|}i}t2}tjtj|dd|D]}|jr8|j t | vr8|j rf|j |j kr8|j}td|jr|jjdtj|j}|j||jd||j t|jt|j t|jt|j|jt|jd||j<q8t|d}|D]H} t | jt!st"d| jr8|#t| jji} | j$%| | VqWd n1sV0Y|j&r|'D]} |(| Ed Hqpd 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_itemstempfileTemporaryDirectoryosmakedirsrudirnameis_filer=r\valuesrKroweb_urlresearch_parenturllibparsequotenamedownloadrIr"rnrprqrr object_idr yield_blobsr`rNotImplementedErrorgetmetadataupdaterJZget_child_folders_load_from_folder) rWrgfile_mime_typesr_ metadata_dicttemp_dirfilermloaderblobfile_metadata_Z subfolderr-r-r.rsT       ( z O365BaseLoader._load_from_folderr)drive object_idsr6c csR|j}i}t(}|D]}||}|sFtd|d|dq|jr|jt| vr|j }t d|j r|j j dtj|j}|j||jd||j|j|jt|jt|j|jt|jd||j<qt|d} | D]F} t| jtst d | jr&|!t| jji} | j"#| | VqWd n1sD0Yd 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 .rirjrkrlrtrvN)$rerwrxZget_itemloggingwarningr|r=r\r}r~rrrrrrrrrIrnror"rprqrrrrrr`rurrrrr) rWrrrrrrrrmrrrr-r-r._load_from_object_idssX       z$O365BaseLoader._load_from_object_idsrcCszddlm}m}Wnty.tdYn0|jrt}|j}||j|jd}|f|j j |j j f|j |dddi}nF|tdd }|f|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)r1)O365rr ImportErrorrHr/r1parentrrFr#r$Zget_secret_valuerfrr2Z authenticate)rWrrZ token_storager1raccountr-r-r._auth$sF       zO365BaseLoader._auth)r)r*r+__doc__rr rFr,rH CHUNK_SIZErIrJrKrLrrMrNrOr[propertyrerrfrrr __classcell__r-r-rcr.rDOs&      &8=rD):r __future__rrr8ryrrwrabcrrpathlibrrtypingrr 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.s8   (