round1.py

import json
import streamlit as st
import shelve
# Import Google's Generative AI package
import google.generativeai as genai

# Configure the Gemini API with your API key
genai.configure(api_key="AIzaSyCA4__JMC_ZIQ9xQegIj5LOMLhSSrn3pMw")

def generate_questions(company_name, role_applied_for, job_description, interview_type, focus_areas):
    """
    Generates custom interview questions based on the provided details,
    explicitly asking for the output in a list format.
    """
    generation_config = {
        "temperature": 0.9,
        "top_p": 1,
        "top_k": 1,
        "max_output_tokens": 2048,
    }

    safety_settings = [
        {"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_MEDIUM_AND_ABOVE"},
        {"category": "HARM_CATEGORY_HATE_SPEECH", "threshold": "BLOCK_MEDIUM_AND_ABOVE"},
        {"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT", "threshold": "BLOCK_MEDIUM_AND_ABOVE"},
        {"category": "HARM_CATEGORY_DANGEROUS_CONTENT", "threshold": "BLOCK_MEDIUM_AND_ABOVE"},
    ]

    model = genai.GenerativeModel(model_name="gemini-1.0-pro",
                                  generation_config=generation_config,
                                  safety_settings=safety_settings)
    
    # Adjust the prompt to explicitly request a list of questions
    prompt_parts = [
        f"Based on the following details, generate a list of interview questions:",
        f"Company Name: {company_name}",
        f"Role Applied For: {role_applied_for}",
        f"Job Description: {job_description}",
        f"Interview Type: {interview_type}",
        f"Focus Areas: {focus_areas}",
        "Format the questions as a bulleted list."
    ]
    
    # Joining the prompt parts to form the complete prompt
    complete_prompt = "\n".join(prompt_parts)
    
    # Call Gemini to generate content
    response = model.generate_content(complete_prompt)
    return response.text




def parse_response_text(response_text):
    """
    Manually parse the structured response text into a dictionary.
    This function assumes a very specific format and might need adjustments
    based on actual response variations.
    """
    questions_dict = {}
    current_section = ""
    for line in response_text.split("\n"):
        if line.strip().endswith(": {"):
            # This is a section header
            current_section = line.split(":")[0].strip()
            questions_dict[current_section] = {}
        elif line.strip().startswith('"Question'):
            # This is a question within the current section
            question_number, question_text = line.strip().split(":")
            question_text = question_text.strip().strip('"').strip(",")
            questions_dict[current_section][question_number.strip('"')] = question_text
    return questions_dict


def app():
    st.title("Interview Round 1 Preparation")

    # Load saved data from bio.py
    with shelve.open("datastore") as db:
        candidate_bio = db.get("candidate_bio", {})  

    if candidate_bio:
        # Displaying loaded information for verification
        st.write("## Loaded Information for Interview Preparation")
        # Extract necessary details
        company_name = candidate_bio.get("company_name", "Not specified")
        interviewing_department = candidate_bio.get("interviewing_department", "Not specified")
        role_applied_for = candidate_bio.get("role_applied_for", "Not specified")
        job_description = candidate_bio.get("job_description", "Not specified")
        interview_type = candidate_bio.get("interview_type", "Not specified")
        focus_areas = candidate_bio.get("focus_areas", "Not specified")

        # Display the loaded information
        st.write(f"**Company Name:** {company_name}")
        st.write(f"**Interviewing Department:** {interviewing_department}")
        st.write(f"**Role Applied For:** {role_applied_for}")
        st.write(f"**Job Description:** {job_description}")
        st.write(f"**Type of Interview:** {interview_type}")
        st.write(f"**Focus Areas for Preparation:** {focus_areas}")

        # Generate and display questions
        if st.button("Generate Interview Questions"):
            questions_text = generate_questions(company_name, role_applied_for, job_description, interview_type, focus_areas)
            st.write(questions_text)


            st.write("## Generated Interview Questions")
            questions_list = questions_text.split('\n')  # Splitting the generated text into a list of questions
            
            # Initialize a place to store responses
            responses = {}
            for i, question in enumerate(questions_list, start=1):
                st.write(f"**Question {i}:** {question}")
                # Capture the user's response
                response = st.text_input(f"Your answer to Question {i}", key=f"response_{i}")
                responses[f"Question {i}"] = response
            
            # Optional: Display captured responses
            if st.button("Submit Responses"):
                st.write("## Your Responses")
                for question, response in responses.items():
                    st.write(f"{question}: {response}")

         



if __name__ == "__main__":
    app()