o hg#@sdZddlZddlZddlmZddlmZmZmZm Z m Z m Z ddl m Z mZmZmZmZddlmZddlmZmZddlmZmZdd lmZdd lmZmZdd lm Z dd l!m"Z"m#Z#m$Z$            d-dede"de%de%de ede&de ee&de%de'de&de%de e%de(d e(d!e(d"ed#df"d$d%Z)    &  d.d'e#de%de ee&de&d(e(de%de ed#e%fd)d*Z*    & d/d'e#de%de ee&de&d(e(de ed#e efd+d,Z+dS)0zIFunctions that can be used for the most common use-cases for pdfminer.sixN)StringIO)AnyBinaryIO ContainerIteratorOptionalcast) HOCRConverter HTMLConverterPDFPageAggregator TextConverter XMLConverter) ImageWriter)LAParamsLTPage) PDFDevice TagExtractor) PDFValueError)PDFPageInterpreterPDFResourceManager)PDFPage)AnyIO FileOrName open_filenametextutf-8?normalFinfoutfp output_typecodeclaparamsmaxpages page_numberspasswordscalerotation layoutmode output_dir strip_controldebugdisable_cachingkwargsreturnc KsL| r ttjd}| rt| }t| d}d}|dkr'|tjkr'tjj}|dkr5t |||||d}nE|dkrDt |||||| d}n6|dkrTt ||||| ||d}n&|d krbt ||||| d }n|d krqt |tt||d }n d |}t||dusJt||}tj||||| dD]}|j| d|_||q|dS)akParses 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 pdfminer.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 pdfminer.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. Ncachingr)r"r# imagewriterxml)r"r#r2 stripcontrolhtml)r"r'r)r#r2hocr)r"r#r4tag)r"z1Output type can be text, html, xml or tag but is r$r&r1ih)logging getLoggersetLevelDEBUGrrsysstdoutbufferr r r r rrrrrr get_pagesrotate process_pageclose)rr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r2rsrcmgrdevicemsg interpreterpagerIL/var/www/html/govbot/env/lib/python3.10/site-packages/pdfminer/high_level.pyextract_text_to_fpsx/         rKTpdf_filer1c Cs|durt}t|dP}t<}tt|}t|d} t| |||d} t| | } tj |||||dD]} | | q2| WdWdS1sOwYWddS1s_wYdS)awParse and return the text contained in a PDF file. :param pdf_file: Either a file path or a file-like object for the PDF file to be worked on. :param password: For encrypted PDFs, the password to decrypt. :param page_numbers: List of zero-indexed page numbers to extract. :param maxpages: The maximum number of pages to parse :param caching: If resources should be cached :param codec: Text decoding codec :param laparams: An LAParams object from pdfminer.layout. If None, uses some default settings that often work well. :return: a string containing all of the text extracted. Nrbr0)r"r#r8) rrrrrrr rrr@rBgetvalue) rLr&r%r$r1r"r#fp output_stringrDrErGrHrIrIrJ extract_texts"    RrQc cs|durt}t|d7}tt|}t|d}t||d}t||} tj|||||dD]} | | | } | Vq-WddS1sGwYdS)aExtract and yield LTPage objects :param pdf_file: Either a file path or a file-like object for the PDF file to be worked on. :param password: For encrypted PDFs, the password to decrypt. :param page_numbers: List of zero-indexed page numbers to extract. :param maxpages: The maximum number of pages to parse :param caching: If resources should be cached :param laparams: An LAParams object from pdfminer.layout. If None, uses some default settings that often work well. :return: LTPage objects NrMr0)r#r8) rrrrrr rrr@rB get_result) rLr&r%r$r1r#rOresource_managerrErGrHlayoutrIrIrJ extract_pagess(      "rU) rrNrNrrrrNFFF)rNrTrN)rNrTN),__doc__r9r=iortypingrrrrrrpdfminer.converterr r r r r pdfminer.imagerpdfminer.layoutrrpdfminer.pdfdevicerrpdfminer.pdfexceptionsrpdfminer.pdfinterprrpdfminer.pdfpagerpdfminer.utilsrrrstrintfloatboolrKrQrUrIrIrIrJs            |  -