indZddlZddlmZddlmZddlmZddlm Z ddlm Z ddl m Z dd l mZmZd Zy) zY PassportEye::Util: PDF processing utilities. Author: Konstantin Tretyakov License: MIT N) PDFParser) PDFDocument)PDFPage)PDFResourceManager)PDFPageInterpreter)PDFPageAggregator)LTFigureLTImagect|}t|}t}t|}t ||}t j |}|D]}|j||j}|D]d} t| ts| D]L} t| tsd} | jj} | 4| jdsF| cccSfy#| jj} YAxYw)a Reads a given PDF file and scans for the first valid embedded JPEG image. Returns either None (if none found) or a string of data for the image. There is no 100% guarantee for this code, yet it seems to work fine with most scanner-produced images around. More testing might be needed though. Note that in principle there is no serious problem extracting PNGs or other image types from PDFs, however at the moment I do not have enough test data to try this, and the one I have seems to be unsuitable for PDFMiner. :param fstream: Readable binary stream of the PDF :return: binary stream, containing the whole contents of the JPEG image or None if extraction failed. Ns)rrrrrr create_pages process_pageresult isinstancer r streamget_data get_rawdata startswith) fstreamparserdocumentrsrcmgrdevice interpreterpagespagelayoutelimstimdatas y/var/www/ITT-ServiceAPI-Services/ITT-Document-Scan/mrzscanner/mrzenv/lib/python3.12/site-packages/passporteye/util/pdf.pyextract_first_jpeg_in_pdfr"sw F6"H "G w 'F$Wf5K   *E*  & *B"h' *B!"g.!=%'YY%7%7%9F"-&2C2CDW2X#)M * **"  =%'YY%:%:%r-s) (,$110-&r,