startup.py

import os
import urllib.request
import zipfile
import subprocess

def setup():
    os.makedirs("data", exist_ok=True)
    os.makedirs("artifacts", exist_ok=True)

    if not os.path.exists("data/medquad.json"):
        print("Downloading MedQuAD dataset...")
        url = "https://github.com/abachaa/MedQuAD/archive/refs/heads/master.zip"
        urllib.request.urlretrieve(url, "medquad.zip")
        
        print("Extracting...")
        with zipfile.ZipFile("medquad.zip", "r") as z:
            z.extractall(".")
        os.remove("medquad.zip")
        
        # Rename MedQuAD-master to MedQuAD
        if os.path.exists("MedQuAD-master"):
            os.rename("MedQuAD-master", "MedQuAD")
            print("Renamed MedQuAD-master to MedQuAD")
        
        print("Parsing XML files...")
        result = subprocess.run(["python", "parse_dataset.py"], 
                              capture_output=True, text=True)
        print(result.stdout)
        if result.returncode != 0:
            print("Parse error:", result.stderr)
        print("Dataset ready!")
    else:
        print("Dataset already exists, skipping download")

    if not os.path.exists("artifacts/retriever.pkl"):
        print("Building retriever...")
        result = subprocess.run(["python", "retriever.py"],
                              capture_output=True, text=True)
        print(result.stdout)
        if result.returncode != 0:
            print("Retriever error:", result.stderr)
        print("Retriever ready!")
    else:
        print("Retriever already exists, skipping build")

if __name__ == "__main__":
    setup()