python-pdf2txt/pdf_to_word.py

import logging
from pdf2image import convert_from_path
import pytesseract
from PIL import Image
import tempfile
from docx import Document
from docx.enum.text import WD_ALIGN_PARAGRAPH
import os

# Configure logging
logging.basicConfig(level=logging.DEBUG, filename='/app/logs/pdf_to_word.log', filemode='w',
                    format='%(name)s - %(levelname)s - %(message)s')

def pdf_to_word(pdf_path, output_dir, lang='eng+fas'):
    pdf_name = os.path.basename(pdf_path).split('.')[0]
    output_path = os.path.join(output_dir, f'{pdf_name}.docx')
    logging.info(f'Starting conversion for {pdf_path}')

    try:
        pages = convert_from_path(pdf_path, 300)
        logging.info(f'Converted PDF to images, number of pages: {len(pages)}')

        document = Document()
        document.add_heading('Document Title', 0)

        for i, page in enumerate(pages):
            with tempfile.NamedTemporaryFile(delete=False, suffix='.jpg') as temp_image:
                page.save(temp_image.name, 'JPEG')
                text = pytesseract.image_to_string(Image.open(temp_image.name), lang=lang, config='--psm 6')
                logging.debug(f'Extracted text from page {i+1}')

                heading = document.add_heading(f'Page {i+1}', level=1)
                heading.alignment = WD_ALIGN_PARAGRAPH.RIGHT
                paragraph = document.add_paragraph(text)
                paragraph.alignment = WD_ALIGN_PARAGRAPH.RIGHT
                
                logging.info("\n------"+f'Page {i+1}'+"------\n"+text+"\n------*------\n")
                os.unlink(temp_image.name)

        document.save(output_path)
        logging.info(f'Document saved to {output_path}')
        return output_path
    except Exception as e:
        logging.error(f'An error occurred: {e}', exc_info=True)
        return None