Pdf-forms-to-HTML-using-GenAI

A simple tool with streamlit interface made to convert PDF forms to HTML forms using LLM (Mixtral)

PDF to Interactive Form Converter

This project provides a Streamlit application that converts PDF files into interactive HTML forms using a language model.

Overview

The application allows users to upload a PDF file, extracts the text from the PDF, and then uses a language model to generate an interactive HTML form based on the extracted text. The generated HTML form can be viewed directly in the application and downloaded for further use.

Setup

To run this application, follow these steps:

Clone the Repository

git clone https://github.com/your-username/pdf-to-interactive-form.git
cd pdf-to-interactive-form

Install Dependencies

Make sure you have Python installed, then install the required packages using pip

Set Up Environment Variables

Create a .env file in the root directory of the project and add your API key for the language model:

GROQ_API_KEY=your_groq_api_key

Run the Application

Start the Streamlit application by running:

streamlit run app.py

Code Explanation

Importing Libraries

from langchain_groq import ChatGroq
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain
from dotenv import load_dotenv
import os
import streamlit as st
from pdfminer.high_level import extract_text

langchain_groq: For interacting with the language model.
dotenv: For loading environment variables.
streamlit: For creating the web application.
pdfminer: For extracting text from PDF files.

Loading Environment Variables

load_dotenv()

Loads the environment variables from a .env file.

Function to Extract Text from PDF

def load_pdf(file):
    try:
        pdf_text = extract_text(file)
        return pdf_text
    except Exception as e:
        st.error(f"Error reading PDF file: {e}")
        return None

This function extracts text from the uploaded PDF file using pdfminer.

Function to Generate Interactive HTML Form

def pdf_to_llm(pdf_text):
    answer_prompt = PromptTemplate.from_template(
        "Given the following parsed pdf form, generate a proper interactive HTML form:\n\nParsed PDF Form: {pdf_text}\n\n Your job is to output the whole complete html, only provide output of the converted html nothing more, make sure the whole complete html is provided."
    )
    llm_model = ChatGroq(model="mixtral-8x7b-32768", api_key=os.getenv("GROQ_API_KEY"))
    llm = LLMChain(llm=llm_model, prompt=answer_prompt)

    try:
        st.text("Sending the following prompt to the LLM:")
        st.text(answer_prompt.format(pdf_text=pdf_text))

        response = llm.run({"pdf_text": pdf_text})

        st.text("Received the following response from the LLM:")
        st.text(response)

        return response
    except Exception as e:
        st.error(f"Error generating interactive form: {e}")
        return None

This function generates an interactive HTML form from the extracted PDF text using a language model.

Main Function for Streamlit App

def main():
    st.title("PDF to Interactive Form Converter")

    uploaded_file = st.file_uploader("Upload a PDF file", type=["pdf"])

    if uploaded_file is not None:
        pdf_text = load_pdf(uploaded_file)
        if pdf_text:
            st.subheader("Uploaded PDF file content:")
            st.text(pdf_text)

            interactive_form = pdf_to_llm(pdf_text)
            if interactive_form:
                st.subheader("Interactive HTML Form:")
                st.markdown(interactive_form, unsafe_allow_html=True)

                st.markdown("---")
                st.subheader("Download Interactive Form")
                st.download_button(
                    label="Download Form",
                    data=interactive_form.encode("utf-8"),
                    file_name="interactive_form.html",
                    mime="text/html",
                )

if __name__ == "__main__":
    main()

The main function sets up the Streamlit application, handles file uploads, displays the extracted PDF text, generates the interactive form, and provides an option to download the form.

Requirements

Create a requirements.txt file with the following content:

langchain_groq
python-dotenv
streamlit
pdfminer.six

This file lists all the dependencies required to run the application.

Note: This code contains alot of unneccessary print statements in streamlit this was just for debugging i was lazy and i didnt remove them while uploading to github, u may remove them:). Any constructive feedback regarding this is appreciated :). This is a simpler version currently working on a better version using llama-parser to parse the pdf and using multiple chains for the conversion to get a more better result any reccommendations for better approach is appreciated.

abdullah-w-21/Pdf-forms-to-HTML-using-GenAI