o =ji{@sNddlZddlZddlZddlmZddlZddlmZdedefddZ dS)N) UploadFile)Imagefilereturnc s|IdH}|dIdH|jdkrzWtj|dd}d}|D] }||d7}q!t|dkrftd |j d d}|D]$}|j t d d d }| d }t t|}t|}||d7}qA||WSty} ztd|j d| WYd} ~ dSd} ~ ww|jdkrztt|}ddd|jDWSty} ztd|j d| WYd} ~ dSd} ~ ww|jdkr|dSz|dWSYdS)z Extracts text from an uploaded file. Supports PDF, DOCX, and plain text. Uses PyMuPDF (fitz) for reliable PDF extraction and Tesseract for OCR if needed. Nrzapplication/pdfpdf)streamfiletype 2z"Normal text extraction failed for z, attempting OCR...)matrixpngzError reading PDF z with OCR fallback: zGapplication/vnd.openxmlformats-officedocument.wordprocessingml.documentcSsg|]}|jqS)text).0 paragraphrr:/var/www/html/Resume-Parser/resume-parser-inhouse/utils.py -sz*extract_text_from_file..zError reading DOCX z: z text/plainzutf-8)readseek content_typefitzopenget_textlenstripprintfilename get_pixmapMatrixtobytesrioBytesIO pytesseractimage_to_stringclose ExceptiondocxDocumentjoin paragraphsdecode) rcontentdocrpagepiximg_dataimg page_texterrrextract_text_from_filesP       r5) r"r(rfastapirr$PILrstrr5rrrrs