import gradio as gr
import spacy
import requests
from bs4 import BeautifulSoup
import PyPDF2
import subprocess
import sys
import time

# Ensure the language model is downloaded
try:
    nlp = spacy.load("en_core_web_sm")
except OSError:
    subprocess.run([sys.executable, "-m", "spacy", "download", "en_core_web_sm"])
    nlp = spacy.load("en_core_web_sm")

# Extract text from PDF
def extract_text_from_pdf(pdf_file):
    pdf_reader = PyPDF2.PdfReader(pdf_file)
    text = ""
    for page in pdf_reader.pages:
        text += page.extract_text() or ""
    return text

# Extract skills and location from text
def extract_skills_and_location(text):
    doc = nlp(text)
    skills = []
    location = None

    skill_keywords = ['Python', 'Data Analysis', 'Machine Learning', 'SQL', 'Java', 'Project Management']

    for token in doc:
        if token.text in skill_keywords and token.text not in skills:
            skills.append(token.text)
    
    for ent in doc.ents:
        if ent.label_ == 'GPE':  # GPE = Geopolitical Entity (e.g., city, country)
            location = ent.text
            break
    
    return skills, location

# Web scraper function for job listings
def fetch_job_listings(job_title, location, retries=3):
    base_url = f"https://www.examplejobboard.com/jobs?query={job_title}&location={location}"
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/85.0.4183.102 Safari/537.36'
    }

    for attempt in range(retries):
        try:
            response = requests.get(base_url, headers=headers, timeout=10)
            response.raise_for_status()
            break  # Exit loop if request is successful
        except requests.RequestException as e:
            if attempt < retries - 1:  # If not the last attempt
                time.sleep(2)  # Wait before retrying
                continue  # Retry the request
            return f"Failed to retrieve job listings. Error: {str(e)}"

    soup = BeautifulSoup(response.text, 'html.parser')
    job_cards = soup.find_all('div', class_='job-card')[:5]  # Adjust to match the website's structure

    if not job_cards:
        return "No job listings found."

    job_listings = []
    for job in job_cards:
        title_tag = job.find('h2', class_='job-title')
        company_tag = job.find('div', class_='company')
        location_tag = job.find('span', class_='job-location')
        description_tag = job.find('p', class_='job-description')
        date_posted_tag = job.find('time', class_='date-posted')

        job_info = {
            'Title': title_tag.text.strip() if title_tag else 'N/A',
            'Company': company_tag.text.strip() if company_tag else 'N/A',
            'Location': location_tag.text.strip() if location_tag else 'N/A',
            'Description': description_tag.text.strip() if description_tag else 'N/A',
            'Date Posted': date_posted_tag.text.strip() if date_posted_tag else 'N/A',
            'URL': title_tag.a['href'] if title_tag and title_tag.a else ''
        }
        job_listings.append(job_info)

    return job_listings

# Main function to handle resume processing
def process_resume(file):
    resume_text = extract_text_from_pdf(file)
    skills, location = extract_skills_and_location(resume_text)
    
    job_title = skills[0] if skills else "Software Engineer"
    job_listings = fetch_job_listings(job_title, location if location else "India")
    
    html_content = "<h2>Extracted Skills</h2>"
    html_content += f"<p>{', '.join(skills)}</p>" if skills else "<p>No specific skills found.</p>"
    html_content += "<h2>Location</h2>"
    html_content += f"<p>{location}</p>" if location else "<p>Location not identified.</p>"
    html_content += "<h2>Job Listings</h2>"
    
    if isinstance(job_listings, str):
        html_content += f"<p>{job_listings}</p>"
    else:
        for job in job_listings:
            html_content += f"""
            <div style='border:1px solid #ccc; padding:10px; margin-bottom:10px;'>
                <h3><a href='{job['URL']}' target='_blank'>{job['Title']}</a></h3>
                <p><strong>Company:</strong> {job['Company']}</p>
                <p><strong>Location:</strong> {job['Location']}</p>
                <p><strong>Date Posted:</strong> {job['Date Posted']}</p>
                <p>{job['Description']}</p>
            </div>
            """
    
    return html_content

# Create a Gradio interface for deployment
gr_interface = gr.Interface(
    fn=process_resume,
    inputs=gr.File(label="Upload your resume (PDF format)"),
    outputs=gr.HTML(),
    title="Resume-Based Job Recommender",
    description="Upload your resume and get job recommendations based on your skills and location.",
)

# Run the interface in the cloud
if __name__ == "__main__":
    gr_interface.launch()