jaUddlZddlmZddlmZddlmZmZmZm Z m Z m Z m Z ddl mZddlmZddlmZmZmZmZddlmZdd lmZmZdd lmZmZmZmZdd l m!Z!dd l"m#Z#d Z$e%e&d< d"de e%de e ee'efde e%de e%dee e%f dZ( d#de e%ee'fde e%dee e%fdZ)eeej*edddddddddddgdf de e%de e ee'efde e%de+de e%de e%de+de e%de e%de e%de e e%d e+de efd!Z,dS)$N)BytesIO)SpooledTemporaryFile)IOBinaryIOIteratorListOptionalUnioncast)etree)add_chunking_strategy)ElementElementMetadataTextprocess_metadata) read_txt_file)FileTypeadd_metadata_with_filetype) exactly_oneget_last_modified_date get_last_modified_date_from_filespooled_to_bytes_io_if_needed)apply_lang_metadata)element_from_textxmlDETECTION_ORIGINfilenamefiletextxml_pathreturnc t||||rt||S|rctttt tt ttf|}t||Stt tt|d}t||S)zGGet leaf elements from the XML tree defined in filename, file, or text.rrr)r zutf-8)encoding) r_get_leaf_elementsr rbytesrr rrrstr)rrrr fbs d/lsinfo/ai/hellotax_ai/base_platform/venv/lib/python3.11/site-packages/unstructured/partition/xml.pyget_leaf_elementsr+s48888 8!(X>>>> 8  uI )U8%99:DAA     "!h7777 E$sD//G<<< = =!!h7777c#Kg}tj|d}|;t|\}}tj|}d||D}|D]\}}|dkr|||dkr=|j"|jr |jV||rJ|d0| |r|d0dS)zz%_get_leaf_elements..Bs&IIBUBKIIIIIIr,r.r/) r iterparsenextXPathappendrstripclear getparentpop)rr element_stackelement_iterator_element compiled_pathevents r*r%r%4s6 Mt4DEEE*++ 7 H-- II--2H2HIII*   w G     ) ) ) E>>|'GL,>,>,@,@'l""" MMOOO b 1 ; ; = = E       b 1 ; ; = = E   r,FTauto xml_keep_tagsmetadata_filenameinclude_metadatar$metadata_last_modifiedchunking_strategy languagesdetect_language_per_elementc  t|||g} d}|rt|}n|rt|}|r"t|p||p|}t|_nt}|ry|rt ||\}}nO|rItttttf|}t ||\}}n|r|}t||g} nWt||||}|D]A}|r=t|}tj||_| |Bt't)| | | } | S) aPartitions an XML document into its document elements. Parameters ---------- filename A string defining the target filename path. file A file-like object using "rb" mode --> open(filename, "rb"). text The text of the XML file. xml_keep_tags If True, will retain the XML tags in the output. Otherwise it will simply extract the text from within the tags. xml_path The xml_path to use for extracting the text. Only used if xml_keep_tags=False. encoding The encoding method used to decode the text input. If None, utf-8 will be used. include_metadata Determines whether or not metadata is included in the metadata attribute on the elements in the output. metadata_last_modified The day of the last modification. languages User defined value for `metadata.languages` if provided. Otherwise language is detected using naive Bayesian filter via `langdetect`. Multiple languages indicates text could be in either language. Additional Parameters: detect_language_per_element Detect language per element instead of at the document level. r#N)r last_modified)rr$)rr$)rmetadata)rrrr )elementsrKrL)rrrrrdetection_originrrr r rrrr+rcopydeepcopyrOr:listr)rrrrFr rGrHr$rIrJrKrLkwargsrPlast_modification_daterOrAraw_textr( leaf_elements leaf_elementrBs r* partition_xmlrZRs`48888 H!H!7!A!A H!A$!G!G%"&2(0J4J   %5!!"$$)  'HMMMKAxx  -U8%99:DAAA(QBBBKAxx  H h 2 2 2  *    * ) )L )+L99#'=#:#: ((((C   H Or,)NNNN)N)-rRiortempfilertypingrrrrr r r lxmlr unstructured.chunkingr unstructured.documents.elementsrrrr unstructured.file_utils.encodingr unstructured.file_utils.filetyperrunstructured.partition.commonrrrrunstructured.partition.langrunstructured.partition.textrrr'__annotations__r&r+r%XMLboolrZr2r,r*ris ))))))FFFFFFFFFFFFFFFFFF777777 ;:::::QQQQQQQQ <;;;;;999999##=A" 88sm8 5E$889 :8 3-8sm 8 hsm 88882#  RY  sm hsm    <HL))"=A"'+!",0'+&,X(-ccsmc 5E$889 :c 3-c c sm c  } ccsmc%SMc }cS "c"&c ']ccc*)cccr,