import streamlit as st
import pandas as pd
import sqlite3
from transformers import pipeline

# Database Setup
conn = sqlite3.connect("career_coach.db", check_same_thread=False)
c = conn.cursor()
c.execute('''CREATE TABLE IF NOT EXISTS users (name TEXT, skills TEXT, experience TEXT, education TEXT, goals TEXT)''')
conn.commit()

# Hugging Face Setup
interview_model = pipeline("text-generation", model="gpt2")  # You can replace with a more suitable model
resume_model = pipeline("text-classification", model="distilbert-base-uncased")

# Career Profile Class
class CareerProfile:
    def __init__(self, name, skills, experience, education, goals):
        self.name = name
        self.skills = skills
        self.experience = experience
        self.education = education
        self.goals = goals

    def save_to_db(self):
        c.execute("INSERT INTO users VALUES (?, ?, ?, ?, ?)",
                  (self.name, self.skills, self.experience, self.education, self.goals))
        conn.commit()

# Job Recommender Class
class JobRecommender:
    def __init__(self, user_skills):
        self.user_skills = user_skills
        self.jobs_df = pd.read_csv("jobs_data.csv")  # Load job dataset

    def recommend_jobs(self):
        return self.jobs_df[self.jobs_df['Skills'].str.contains(self.user_skills, case=False, na=False)].head(5)

# Interview Coach Class
class InterviewCoach:
    def __init__(self, question):
        self.question = question

    def get_feedback(self, response):
        prompt = f"Evaluate the following response for a job interview question '{self.question}': {response}"
        result = interview_model(prompt, max_length=200, num_return_sequences=1)
        return result[0]['generated_text']

# Resume Enhancer Class
class ResumeEnhancer:
    def __init__(self, resume_text):
        self.resume_text = resume_text

    def analyze_resume(self):
        prompt = f"Analyze this resume and suggest improvements for ATS compatibility: {self.resume_text}"
        result = resume_model(prompt)
        return result[0]['label']  # Assuming the model returns the label for the analysis

# Streamlit UI
st.sidebar.title("Virtual Career Coach")
page = st.sidebar.radio("Navigation", ["Home", "Job Recommendations", "Interview Coach", "Resume Enhancer", "Career Growth"])

if page == "Home":
    st.title("Career Profile Setup")
    name = st.text_input("Name")
    skills = st.text_area("Skills")
    experience = st.text_area("Experience")
    education = st.text_area("Education")
    goals = st.text_area("Career Goals")
    if st.button("Save Profile"):
        profile = CareerProfile(name, skills, experience, education, goals)
        profile.save_to_db()
        st.success("Profile saved successfully!")

elif page == "Job Recommendations":
    st.title("Job Recommendations")
    user_skills = st.text_input("Enter your skills for job matching")
    if st.button("Find Jobs"):
        recommender = JobRecommender(user_skills)
        jobs = recommender.recommend_jobs()
        st.write(jobs)

elif page == "Interview Coach":
    st.title("AI-Powered Interview Coach")
    question = st.text_input("Enter a job interview question")
    response = st.text_area("Your Response")
    if st.button("Get Feedback"):
        coach = InterviewCoach(question)
        feedback = coach.get_feedback(response)
        st.write(feedback)

elif page == "Resume Enhancer":
    st.title("Resume & Cover Letter Enhancer")
    resume_text = st.text_area("Paste your resume text")
    if st.button("Analyze Resume"):
        enhancer = ResumeEnhancer(resume_text)
        feedback = enhancer.analyze_resume()
        st.write(feedback)

elif page == "Career Growth":
    st.title("Career Growth Suggestions")
    st.write("Coming Soon: Personalized Learning Paths, Networking Strategies, and Industry Insights!")