extract_tables_piMuPDF.py

import sys, pathlib, pymupdf, pymupdf4llm
from tabulate import tabulate
import pandas
fname = sys.argv[1] 


with pymupdf.open(fname) as doc:
    for page in doc:
        tables=page.find_tables()
        for table in tables:
            with open("Extracted_tables_pimupdf","a") as table_file:
                 table_file.write(tabulate(table.extract(), 
                                    headers="firstrow",  # First row as headers
                                    tablefmt="grid",     # Grid format with borders 
                                    stralign="left"))    # Left align text   
                 
            #Extract tables as markdown 
            with open("Extracted_tables_markdown","a") as table_file:
                    table_file.write(table.to_markdown() + "\n *3")
                    
            #Extract tables as html 
            with open("Extracted_tables_markdown.html","a") as table_file:
                    table_file.write(table.to_pandas().to_html()+ "\n *3")
                    
            #Extract tables as html 
            with open("Extracted_tables_markdown.json","a") as table_file:
                    table_file.write(table.to_pandas().to_json()+ "\n *3")