hhdZddlZddlZddlmZddlmZddlmZm Z m Z ddl m Z ddl Z ddlmZddlmZdd lmZmZGd d eeZGd d eZdS)zLoads word documents.N)ABC)Path)AnyListUnion)urlparse)Document) BaseLoader)UnstructuredFileLoadervalidate_unstructured_versioncneZdZdZdeeeffdZd dZde e fdZ e dede fd ZdS) Docx2txtLoaderzLoad `DOCX` file using `docx2txt` and chunks at character level. Defaults to check for local file, but if the file is a web path, it will download it to a temporary file, and use that, then clean up the temporary file after completion file_pathct||_|j|_d|jvr)tj|j|_tj|js||jrtj |j}|j dkrtd|j z|j|_ tj|_|j|j|jj|_dStj|jstd|jzdS)zInitialize with file path.~z3Check the url of your file; returned status code %sz'File path %s is not a valid file or urlN)strroriginal_file_pathospath expanduserisfile _is_valid_urlrequestsget status_code ValueErrorweb_pathtempfileNamedTemporaryFile temp_filewritecontentname)selfrrs o/var/www/FlaskApp/flask-venv/lib/python3.11/site-packages/langchain_community/document_loaders/word_document.py__init__zDocx2txtLoader.__init__s-Y"&. $. W//??DNw~~dn-- Y$2D2DT^2T2T Y T^,,A}## Im$ !NDM%8::DN N  + + +!^0DNNN// YFWXX X Y YreturnNc\t|dr|jdSdS)Nr!)hasattrr!close)r%s r'__del__zDocx2txtLoader.__del__3s8 4 % % # N " " " " " # #r)clddl}t||jd|jigS)zLoad given path as single page.rNsource) page_contentmetadata)docx2txtr processrr)r%r3s r'loadzDocx2txtLoader.load7sG %--dn=="D$;<     r)urlcpt|}t|jot|jS)zCheck if the url is valid.)rboolnetlocscheme)r6parseds r'rzDocx2txtLoader._is_valid_urlBs.#FM"":tFM':'::r))r*N)__name__ __module__ __qualname____doc__rrrr(r.rr r5 staticmethodr8rr)r'rrs Y%T "2YYYY0####  d8n     ;3;4;;;\;;;r)rcPeZdZdZ d deeefdedeffd Zde fdZ xZ S) UnstructuredWordDocumentLoadera_Load `Microsoft Word` file using `Unstructured`. Works with both .docx and .doc files. You can run the loader in one of two modes: "single" and "elements". If you use "single" mode, the document will be returned as a single langchain Document object. If you use "elements" mode, the unstructured library will split the document into elements such as Title and NarrativeText. You can pass in additional unstructured kwargs after mode to apply different unstructured settings. Examples -------- from langchain_community.document_loaders import UnstructuredWordDocumentLoader loader = UnstructuredWordDocumentLoader( "example.docx", mode="elements", strategy="fast", ) docs = loader.load() References ---------- https://unstructured-io.github.io/unstructured/bricks.html#partition-docx singlermodeunstructured_kwargsc ^t|}tjd||d|dS)a& Args: file_path: The path to the Word file to load. mode: The mode to use when loading the file. Can be one of "single", "multi", or "all". Default is "single". **unstructured_kwargs: Any kwargs to pass to the unstructured. )rrENrA)rsuperr()r%rrErF __class__s r'r(z'UnstructuredWordDocumentLoader.__init__bs; NN O94OO;NOOOOOr)r*crddlm}m} ddl}||j|jk}nJ#t $r=tj t|j\}}|dk}YnwxYw|rtd|rddl m }|dd|ji|jSddlm}|dd|ji|jS) Nr)FileTypedetect_filetypez.docz0.4.11) partition_docfilename)partition_docxrA) unstructured.file_utils.filetyperKrLmagicrDOC ImportErrorrrsplitextrr unstructured.partition.docrMrFunstructured.partition.docxrO) r%rKrLrQis_doc_ extensionrMrOs r' _get_elementsz,UnstructuredWordDocumentLoader._get_elementsssNNNNNNNN  ) LLL$_T^44 DFF ) ) )7++C,?,?@@LAy&(FFF )  4 )( 3 3 3  W @ @ @ @ @ @ =UU$.UDVV4>VT=UVV Vs(AA/.A/)rD) r<r=r>r?rrrrr(rrZ __classcell__)rIs@r'rCrCIs6PPd#PP # PPPPPP"WtWWWWWWWWr)rC)r?rrabcrpathlibrtypingrrr urllib.parserrlangchain_core.documentsr )langchain_community.document_loaders.baser 1langchain_community.document_loaders.unstructuredr r rrCrAr)r'rcs< ##########!!!!!!------@@@@@@ 2;2;2;2;2;Z2;2;2;jBWBWBWBWBW%;BWBWBWBWBWr)