app.py

# IMPORTS
import io
import re
from PIL import Image
import gradio as gr
import pdfplumber
from docx import Document

# Hugging Face OCR model imports
from transformers import TrOCRProcessor, VisionEncoderDecoderModel
import torch

# ---------------- HF OCR SETUP ----------------
processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-stage1")
model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-stage1")

def extract_text_from_image(file_bytes):
    """Extract text from image using HF TrOCR model"""
    image = Image.open(io.BytesIO(file_bytes)).convert("RGB")
    pixel_values = processor(images=image, return_tensors="pt").pixel_values
    generated_ids = model.generate(pixel_values)
    text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
    return text

# ---------------- PDF TEXT ----------------
def extract_text_from_pdf(file_bytes):
    text = ""
    with pdfplumber.open(io.BytesIO(file_bytes)) as pdf:
        for page in pdf.pages:
            text += page.extract_text() or ""
    return text

# ---------------- DOCX TEXT ----------------
def extract_text_from_docx(file_bytes):
    doc = Document(io.BytesIO(file_bytes))
    return "\n".join([p.text for p in doc.paragraphs])

# ---------------- FILE HANDLER ----------------
def extract_text_from_file(file_bytes, file_ext):
    if file_ext == "pdf":
        return extract_text_from_pdf(file_bytes)
    elif file_ext == "docx":
        return extract_text_from_docx(file_bytes)
    elif file_ext in ["jpg", "jpeg", "png"]:
        return extract_text_from_image(file_bytes)
    return ""

# ---------------- RESUME VALIDATION ----------------
def validate_resume(text):
    text_lower = text.lower()
    issues = []

    # Contact Info
    has_email = bool(re.search(r'[\w\.-]+@[\w\.-]+\.\w+', text_lower))
    has_phone = bool(re.search(r'\d{3}[-.\s]?\d{3}[-.\s]?\d{4}|\d{10,11}', text_lower))
    email_keywords = ['@gmail', '@yahoo', '@hotmail', '.com', 'email', 'e-mail']
    has_any_email_keyword = any(k in text_lower for k in email_keywords)
    if not (has_email or has_phone or has_any_email_keyword):
        issues.append("Missing Contact Info")

    # Education
    if not re.search(r'(education|degree|bachelor|master|university|school|college|bs|ms|phd)', text_lower):
        issues.append("Missing Education")

    # Experience
    if not re.search(r'(experience|worked|roles?|employment|projects?|internship|career|manager|designer|assistant|executive|specialist|developer|engineer|analyst|officer|coordinator)', text_lower):
        issues.append("Missing Experience")

    # Skills
    if not re.search(r'(marketing|communication|skills|technologies|tools|competencies|python|excel|sql|java|c\+\+|javascript|html|css|react|node|git|linux|aws|docker|kubernetes|leadership|teamwork)', text_lower):
        issues.append("Missing Skills")

    return issues if issues else ["✅ Resume is Complete!"]

# ---------------- MAIN FUNCTION ----------------
def check_resume(file):
    if file is None:
        return "⚠️ Please upload a file", ""
    try:
        file_ext = file.name.split(".")[-1].lower()

        # Gradio v3+ safe file reading
        with open(file.name, "rb") as f:
            file_bytes = f.read()

        text = extract_text_from_file(file_bytes, file_ext)
        if not text.strip():
            return "⚠️ Could not extract text", ""
        result = "\n".join(validate_resume(text))
        return result, text[:1000]  # show first 1000 chars
    except Exception as e:
        return f"⚠️ Error: {str(e)}", ""

# ---------------- GRADIO INTERFACE ----------------
demo = gr.Interface(
    fn=check_resume,
    inputs=gr.File(label="Upload Resume (PDF, DOCX, JPG, PNG)"),
    outputs=[gr.Textbox(label="Result", lines=6),
             gr.Textbox(label="Extracted Text", lines=6)],
    title="📄 Resume Completeness Checker",
    description="Upload clear resume files for better results."
)

demo.launch(share=True)