:Qg@ XUddlmZddlZddlmZddlmZmZmZm Z ddl m Z ddl m Z ddlmZmZmZddlmZdd lmZdd lmZmZdd lmZmZdd lmZd Zded<eej@e ddddddd ddZ! ddZ" ddZ#y)) annotationsN)BytesIO)IOAnyIteratorcast)etree)add_chunking_strategy)ElementElementMetadataText) read_txt_file)FileType) exactly_onespooled_to_bytes_io_if_needed)apply_metadataget_last_modified_date)element_from_textxmlstrDETECTION_ORIGINF)filetextencoding xml_keep_tagsxml_pathc t|||g}t||r t|nd}t|_|rE|rt ||d} n"|rt t ||d} n|J|} t| |g}|St||||} | D];} | st| } tj|| _ |j| =|S) apPartitions 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. encoding The encoding method used to decode the text input. If None, utf-8 will be used. 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. filenamerrN)r last_modified)rr)rr)rmetadata)rrrr)rr rrdetection_originrrr get_leaf_elementsrcopydeepcopyr"append) rrrrrrkwargselementsr"raw_text leaf_elements leaf_elementelements W/var/www/html/answerous/venv/lib/python3.12/site-packages/unstructured/partition/xml.py partition_xmlr/s<48 HX)?)I[_H!1H $hJ1MH $*G*MX`abcdH# ##Hh:; O*  * )L+L9#'==#: (  ) Oct||||r t||S|rtt||Stt t t |d}t||S)zGGet leaf elements from the XML tree defined in filename, file, or text.r)r)rrzutf-8)r)r_get_leaf_elementsrrbytesrr)rrrrbs r.r$r$Ys[48!(X>> !'DT'JU]^^ E$sD/G< =!!h77r0c#Kg}tj|dd}|2t|\}}tj|}d||D}|D]\}}|dk(r|j ||dk(rD|j (|j j r|j |j|sh|dj||j|s|dj'yw) zz%_get_leaf_elements..usIBUBKIsr6r7) r iterparsenextXPathr'rstripclear getparentpop)rr element_stackelement_iterator_r- compiled_pathevents r.r2r2gs +-Mt4DW\]*+ 7 H- I-2HI*  w G    ) E>||'GLL,>,>,@ll" MMO b 1 ; ; = E     b 1 ; ; = E  sB8C9;C9C9"C96C9)N)r str | NonerIO[bytes] | NonerrLrrLrboolrrLr(rreturnz list[Element]) rrLrrMrrLrrLrOIterator[str | None])rzstr | IO[bytes]rrLrOrP)$ __future__rr%iortypingrrrrlxmlr unstructured.chunkingr unstructured.documents.elementsr r r unstructured.file_utils.encodingrunstructured.file_utils.modelr$unstructured.partition.common.commonrr&unstructured.partition.common.metadatarrunstructured.partition.textrr__annotations__XMLr/r$r2r;r0r.r^s" **7JJ:2Z9# ="== =  =  =  =====@ 8 8 0 88B 8NX 8 8    r0