import os
import gradio as gr
from openai import OpenAI
import PyPDF2
import pandas as pd
import matplotlib.pyplot as plt
import tempfile
import json
import re

# Initialize OpenAI client
client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))

# ---------------------------
# 🧠 Helper Functions
# ---------------------------

def extract_text_from_pdf(file_path):
    """Extract text safely from a PDF"""
    try:
        text = ""
        with open(file_path, "rb") as f:
            reader = PyPDF2.PdfReader(f)
            for page in reader.pages:
                page_text = page.extract_text() or ""
                text += page_text
        return text.strip()
    except Exception as e:
        return f"Error reading PDF: {e}"

def analyze_resume(resume_text, job_desc):
    """Call GPT-4 model to analyze resume"""
    try:
        prompt = f"""
        You are an expert HR AI.
        Compare the following resume to the given job description.
        Rate how well the resume fits the job on a scale of 0–100.
        Then summarize the key strengths and weaknesses.

        Return output strictly in JSON format:
        {{
          "score": <number>,
          "summary": "<text>"
        }}

        JOB DESCRIPTION:
        {job_desc}

        RESUME:
        {resume_text}
        """

        response = client.chat.completions.create(
            model="gpt-4o-mini",
            messages=[{"role": "user", "content": prompt}],
            temperature=0.4,
        )

        reply = response.choices[0].message.content
        match = re.search(r"\{.*\}", reply, re.S)
        if match:
            return json.loads(match.group())
        else:
            return {"score": 50, "summary": reply.strip()}

    except Exception as e:
        return {"score": 0, "summary": f"Error: {e}"}

def process_resumes(files, job_desc):
    """Main pipeline: extract text, analyze, chart results"""
    if not files or not job_desc:
        return "⚠️ Please upload resumes and enter a job description.", None, None

    results = []
    for f in files:
        text = extract_text_from_pdf(f.name)
        result = analyze_resume(text, job_desc)
        results.append({
            "Filename": os.path.basename(f.name),
            "Score": result.get("score", 0),
            "Summary": result.get("summary", "")
        })

    df = pd.DataFrame(results)

    # Generate visualization
    fig, ax = plt.subplots(figsize=(6, 3))
    ax.bar(df["Filename"], df["Score"], color="skyblue")
    ax.set_title("Resume Fit Scores")
    ax.set_ylabel("Score (0–100)")
    plt.xticks(rotation=15, ha="right")
    plt.tight_layout()

    tmp_chart = tempfile.NamedTemporaryFile(delete=False, suffix=".png")
    plt.savefig(tmp_chart.name)
    plt.close(fig)

    return df, tmp_chart.name, "✅ Analysis complete!"

# ---------------------------
# 🎨 Gradio UI
# ---------------------------

with gr.Blocks(theme=gr.themes.Soft(), title="AI Resume Screener Pro") as app:
    gr.Markdown("## 🤖 AI Resume Screener Pro\nUpload resumes and compare them with a job description using GPT-4.")

    with gr.Row():
        job_desc = gr.Textbox(label="Job Description", lines=6, placeholder="Enter the job description here...")
        files = gr.File(label="Upload Resumes", file_count="multiple", type="filepath")

    analyze_btn = gr.Button("🚀 Analyze Resumes", variant="primary")

    output_table = gr.Dataframe(headers=["Filename", "Score", "Summary"], label="Results")
    output_plot = gr.Image(label="Visualization")
    output_status = gr.Textbox(label="Status", interactive=False)

    analyze_btn.click(fn=process_resumes, inputs=[files, job_desc],
                      outputs=[output_table, output_plot, output_status])

    gr.Markdown("---")
    gr.Markdown("Built with ❤️ using OpenAI GPT-4 and Gradio.")

app.launch()