Ig/dZddlmZddlZddlZddlZddlZddlmZddl m Z m Z ddl m Z mZmZmZmZmZmZmZddlmZmZmZmZmZddlmZmZdd lmZm Z dd l!m"Z"dd l#m$Z$dd l%m&Z&dd l'm(Z(e rddl)m*Z*ddl+m,Z,m-Z-ej\e/Z0dZ1GddeZ2GddeZ3ddZ4ddZ5Gdde eZ6y)1Base class for all loaders that uses O365 Package) annotationsN)abstractmethod)PathPurePath) TYPE_CHECKINGAnyDictIterableListOptionalSequenceUnion) BaseModelFieldFilePath PrivateAttr SecretStr) BaseSettingsSettingsConfigDict)BaseBlobParser BaseLoader)FileSystemBlobLoader)Blob)MimeTypeBasedParser) get_parser)Account)DriveFolderiPcbeZdZUeddZded<eddZded<ed d d d Zy) _O365Settings.O365_CLIENT_ID)aliasstr client_idO365_CLIENT_SECRETr client_secretFz.envignore)case_sensitiveenv_file env_prefixextraN) __name__ __module__ __qualname__rr%__annotations__r'r model_configk/var/www/html/answerous/venv/lib/python3.12/site-packages/langchain_community/document_loaders/base_o365.pyr!r!'s93&67Is7$S0DEM9E%v"HLr4r!cFeZdZUejdz dz Zded<y)_O365TokenStorage .credentialszo365_token.txtr token_pathN)r.r/r0rhomer9r1r3r4r5r7r70s$499;7:JJJJr4r7cxi}|D]2}tjd|\}}|r|||<&td||S)2Fetch the mime types for the specified file types.zfile.zUnknown mimetype of extention ) mimetypes guess_type ValueError) file_typesmime_types_mappingext mime_type_s r5fetch_mime_typesrE4s\E ++eC5M: 1 &/ s #=cUCD D E r4cri}|D]/}tj|}|r |||dd<#td||S)r<NzUnknown mimetype )r=guess_extensionr?) mime_typesrArCrBs r5fetch_extensionsrJ@sV> '' 2 *3 s12w '0 <= = > r4ceZdZUdZeeZded< dZded< e Z ded < dZ ded < iZ d ed < e Zd ed<e Zded<e Zded<dfd ZeddZeeddZddZ ddZddZxZS)O365BaseLoaderr)default_factoryr!settingsFboolauth_with_tokenzUnion[int, str] chunk_size recursivezOptional[Dict[str, Any]]handlersr _blob_parser Sequence[str] _file_typesDict[str, str] _mime_typesc ~t|di||jrt|jj } t ||_tt||_|jjDcic]\}}|j ||}}}t|d|_ ytd|_ t|jts!t!dt#|jtt|jjj |_ycc}}w#t$ro t||_tt|j j |_|j}n#t$rtd|dwxYwYwxYw)Nz=`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)rSfallback_parserdefaultzLget_parser("default) was supposed to return MimeTypeBasedParser.It returned r3)super__init__rSlistkeysrErXsetrVitemsr?rJrrTr isinstance TypeErrortype)selfkwargs handler_keys extensionhandler mime_handlers __class__s r5r]zO365BaseLoader.__init__ys "6" == 2 2 45L #3L#A #'L(9#: /3mm.A.A.C!* 7$$Y/8! !$!4&!D !+9 5D d//1DE##'(9(9#:";= 0T5F5F5O5O5T5T5V0WXD 9!  '7 'ED$'+C0@0@0E0E0G,H'ID$$(MMM!$X'.)66" s>AED>!E>E F<A FF<F55F<;F<c|jS)zBReturn a dict of supported file types to corresponding mime types.)rXres r5_fetch_mime_typesz O365BaseLoader._fetch_mime_typessr4cy)zReturn required scopes.Nr3rms r5_scopeszO365BaseLoader._scopessr4c #K|j}|j}i}tj5}t j tj j|d|D]}|js|jt|jvs6|j||j|j|jt|j t|j"t|j$t|j&|j(t|j*d||j,<t/|}|j1D]}t3|j t4s t7d|j rJ|j9t|j j,i} |j:j=| | ddd|j>r/|jAD]} |jC| Ed{yy#1swYExYw7w)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_okto_pathrQsourcerCcreatedmodified created_by modified_by descriptionidpath#Expected blob path to be a PurePathN)"rn get_itemstempfileTemporaryDirectoryosmakedirsr~dirnameis_filerCr^valuesdownloadrQweb_urlr$rwrxryrzr{ object_idnamer yield_blobsrbrNotImplementedErrorgetmetadataupdaterRget_child_folders_load_from_folder) refolderfile_mime_typesra metadata_dicttemp_dirfileloaderblobfile_metadata_ subfolders r5rz O365BaseLoader._load_from_folders00  "35  ( ( * h KK1D A <<~~o.D.D.F)GG h4?? S&*ll)-'*4<<'8(+DMM(:*-doo*>+.t/?/?+@+/+;+;"%dnn"5 4 dii0 *x8F**, !$))X6-.STT99%2%6%6s499>>7JB%ONMM((8  % 2 >>#557 = 11)<<< = 3  6=s63I AH;<%H;"EH;7;I 2I3I ;II c #K|j}i}tj5}|D]}|j|}|st j d|d|d4|j sA|jt|jvsg|j||j|j|j|j|jt|j t|j"|j$t|j&d||j(< t+|}|j-D]} t/| j0t2s t5d| j0rJ|j7t| j0j(i} | j8j;| |  dddy#1swYyxYww) aLazily load files specified by their object_ids from a drive. Load files into the system as binary large objects (Blobs) and return Iterable. Args: drive: The Drive instance from which the files are to be loaded. This Drive instance should represent a cloud storage service or similar storage system where the files are stored. object_ids: A list of object_id strings. Each object_id represents a unique identifier for a file in the drive. Yields: An iterator that yields Blob instances, which are binary representations of the files loaded from the drive using the specified object_ids. z!There isn't a file withobject_id z in drive .rsrur}rN)rnrrget_itemloggingwarningrrCr^rrrQrrwrxr$ryrzr{rrrrrbr~rrrrr) redrive object_idsrrrrrrrrs r5_load_from_object_idsz$O365BaseLoader._load_from_object_idss$0035  ( ( * h'  ~~i0OO%%.Kz%C<<~~o.D.D.F)GG h4?? S&*ll)-'+||(, *-doo*>+.t/?/?+@+/+;+;"%dnn"5 4 dii0 ,*x8F**, !$))X6-.STT99%2%6%6s499>>7JB%ONMM((8  1   s*#G(AG)%GEG G(G%!G(ch ddlm}m}|jrt }|j }||j|j}|d |jj|jjjf|j|dddi}|S|tjdz }|d |jj|jjjf|j|dddi}|j!|S#t$r tdwxYw) znAuthenticates the OneDrive API client Returns: The authenticated Account object. r)rFileSystemTokenBackendzAO365 package not found, please install it with `pip install o365`)r9token_filename) credentialsscopes token_backendraise_http_errorsFr8)r9r3)O365rr ImportErrorrPr7r9parentrrNr%r'get_secret_valuerprr: authenticate)rerr token_storager9raccounts r5_authzO365BaseLoader._auth s?   <   -/M&11J2%,,Z__MMM++MM//@@B||+ '.G0399;7MMM++MM//@@B||+ '.G  "E S  s DD1)rfr returnNone)rrW)r List[str])rrrIterable[Blob])rrrrrr)rr)r.r/r0__doc__rr!rNr1rP CHUNK_SIZErQrRrSrrTrVrXr]propertyrnrrprrr __classcell__)rks@r5rLrLLs;#MBHmB0!OT!I",J,VIt8)+H&+:$/=L.0!,K."--K/$YL  &&*=X22(12 2h*r4rL)r@rUrrW)rIrUrrW)7r __future__rrr=rrabcrpathlibrrtypingrr r r r r rrpydanticrrrrrpydantic_settingsrr)langchain_community.document_loaders.baserr=langchain_community.document_loaders.blob_loaders.file_systemr8langchain_community.document_loaders.blob_loaders.schemar4langchain_community.document_loaders.parsers.genericr5langchain_community.document_loaders.parsers.registryrrr O365.driverr getLoggerr.loggerrr!r7rErJrLr3r4r5rs7" "VVV?PJTL(   8 $  LK K  gZgr4