Fg%3:SrSSKrSSKrSSKJr SSKJrJrJrJ r J r J r SSK r SSK rSSKrSSKJr SSKJrJrJrJrJr SSKJr SSKJrJr SS KJrJr SS K J!r! SS K"J#r#J$r$ SS K%J&r& SS K'J(r(J)r)J*r* S-S\S\(S\+S\+S\ \S\,S\ \\,S\+S\-S\,S\+S\ \+S\.S\.S\.S\,S\+S\+S \,S!\S"\+S#\+S$\+S%\S&S42S'jjr/S.S(\)S\+S)\ \\,S\,S*\.S\+S\ \S&\+4S+jjr0S/S(\)S\+S)\ \\,S\,S*\.S\ \S&\ \4S,jjr1g)0zGFunctions that can be used for the most common use-cases for pdf2zh.sixN)StringIO)AnyBinaryIO ContainerIteratorOptionalcast)Document) HOCRConverter HTMLConverterPDFPageAggregator TextConverter XMLConverter) ImageWriter)LAParamsLTPage) PDFDevice TagExtractor) PDFValueError)PDFPageInterpreterPDFResourceManager)PDFPage)AnyIO FileOrName open_filenameinfoutfp output_typecodeclaparamsmaxpagespagespasswordscalerotation layoutmode output_dir strip_controldebugdisable_caching page_countvfontvcharthreaddoc_enlang_inlang_outservicekwargsreturnc U (a2[R"5R[R5 SnU (a [ U 5n[ U(+S9nSn0nUS:wa.U[ R:Xa[ RRnUS:Xa[UUUUUUUUUUUUS9 nOoUS:Xa[UUUUUU S9nOZUS:Xa[UUUUU UUS9nODUS :Xa[UUUUU S 9nO0US :Xa[U[[U5US 9nOS U3n[!U5eUce0n[#UUU5nU(a [%U5n OUn [&R&"[(R*"UUUUU(+S9U SS9GHn!UU!R,R/5n"[0R2"U"R4[0R65R9U"R:U"R<S5SS2SS2SSS24n#UR?U#[AU"R:S- 5S-[BRDRG5(aSOSS9Sn$[0RH"U"R:U"R<45n%U%RJun&n'/SQn([MU$RN5GHun)n*U$RP[AU*RR5U(;dM/U*RTRW5un+n,n-n.[0RX"[AU+S- 5SU'S- 5[0RX"[AU&U.- S- 5SU&S- 5[0RX"[AU-S-5SU'S- 5[0RX"[AU&U,- S-5SU&S- 54un+n,n-n.U)S-U%U,U.2U+U-24'GM [MU$RN5Hun)n*U$RP[AU*RR5U(;dM.U*RTRW5un+n,n-n.[0RX"[AU+S- 5SU'S- 5[0RX"[AU&U.- S- 5SU&S- 5[0RX"[AU-S-5SU'S- 5[0RX"[AU&U,- S-5SU&S- 54un+n,n-n.SU%U,U.2U+U-24'GM U%UU!R,'U!RZU -S-U!l-UR]5U!l/URaU!R^S5 URcU!R^S5 UU!R,ReU!R^5 URgU!5 GM URi5 U$)aParses text from inf-file and writes to outfp file-like object. Takes loads of optional arguments but the defaults are somewhat sane. Beware laparams: Including an empty LAParams is not the same as passing None! :param inf: a file-like object to read PDF structure from, such as a file handler (using the builtin `open()` function) or a `BytesIO`. :param outfp: a file-like object to write the text to. :param output_type: May be 'text', 'xml', 'html', 'hocr', 'tag'. Only 'text' works properly. :param codec: Text decoding codec :param laparams: An LAParams object from pdf2zh.layout. Default is None but may not layout correctly. :param maxpages: How many pages to stop parsing after :param page_numbers: zero-indexed page numbers to operate on. :param password: For encrypted PDFs, the password to decrypt. :param scale: Scale factor :param rotation: Rotation factor :param layoutmode: Default is 'normal', see pdf2zh.converter.HTMLConverter :param output_dir: If given, creates an ImageWriter for extracted images. :param strip_control: Does what it says on the tin :param debug: Output more logging data :param disable_caching: Does what it says on the tin :param other: :return: nothing, acting as it does on two streams. Use StringIO to get strings. Ncachingtext) rr imagewriterr,r-r.layoutr0r1r2xml)rr r9 stripcontrolhtml)rr$r&r r9hocr)rr r<tag)rz1Output type can be text, html, xml or tag but is r!r#r7r)totalposition zcuda:0cpu)imgszdevice)abandonfiguretableisolate_formulaformula_captionihz<<>>)5logging getLoggersetLevelDEBUGrrsysstdoutbufferrrr r rr rrrlentqdmr get_pagespageno get_pixmapnp fromstringsamplesuint8reshapeheightwidthpredictinttorchcuda is_availableonesshape enumerateboxesnamesclsxyxysqueezecliprotate get_new_xref page_xref update_object update_stream set_contents process_pageclose)/rrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/modelr0r1r2r3r9rsrcmgrrHr:msg obj_patch interpreter total_pagespagepiximage page_layoutboxhwvclsidx0y0x1y1s/ b/Users/admin/workspace/ai/PDFMathTranslate/myenv/lib/python3.13/site-packages/pdf2zh/high_level.pyextract_text_to_fprsRp $$W]]3K!*-  _)<=G"&F Ff#**!4 !!f  #      #&      !#     &    gtHe'"#3G()9B %%    D  $ $T *&&(FL ' & &sB(A2B B( B%!B()r8utf-8NrNg?rnormalNFFFrrrrNNrrr)rNrTrN)rNrTN)2__doc__rQrUiortypingrrrrrr rfnumpyr]rYpymupdfr pdf2zh.converterr r r rr pdf2zh.imager pdf2zh.layoutrrpdf2zh.pdfdevicerrpdf2zh.pdfexceptionsrpdf2zh.pdfinterprrpdf2zh.pdfpager pdf2zh.utilsrrrstrrefloatboolrrrrPrrsM EE  %*4.C"99 #'&* $! 1j j jj  j x j  j IcN #jj jjj jj jj !j" #j$ %j& 'j( )j,-j./j01j23j4 5j^-1#'((((((9S>*(( ((  ((  ((x (( ((Z-1#' %%%9S>*% %  % x %f%rP