app.py

import io
import os
import streamlit as st
import tempfile

from scripts import analyze_metadata, generate_metadata, ingest, MODEL_NAME


st.title('# DocVerifyRAG')
st.write('## Anomaly detection for BIM document metadata')

st.write('### Enter your file metadata in the following schema:')

user_input = st.text_input(
    label='Filename, Description, Discipline',
    value="", placeholder=str)

if st.button('Submit'):
    try:
        filename, description, discipline = user_input.split(',')

        st.write('## Analyzing with Vectara + together.ai')
        analysis = analyze_metadata(filename, description, discipline)

        st.write(analysis)

        st.write('## Generate metadata?')
        st.write('### Upload the file that corresponds to the submitted metadata')

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

        if uploaded_file is not None:
            try:
                with tempfile.NamedTemporaryFile(delete=False, suffix=os.path.splitext(uploaded_file.name)[1]) as tmp:
                    tmp.write(uploaded_file.read())
                    file_path = tmp.name
                    st.write(f'Created temporary file {file_path}')

                docs = ingest(file_path)
                st.write('## Querying Together.ai API')
                metadata = generate_metadata(docs)
                st.write(f'## Metadata Generated by {MODEL_NAME}')
                st.write(metadata)

                # Clean up the temporary file
                os.remove(file_path)

            except Exception as e:
                st.error(f'Error: {e}')
    except ValueError:
        st.error('Please enter 3 comma separated values')