seed_data.py

import random
import logging
from datetime import datetime, timedelta
from focusguard.db.database_manager import DatabaseManager
from focusguard.db.user_repo import UserRepo
from focusguard.db.session_repo import SessionRepo
from focusguard.db.app_usage_repo import AppUsageRepo, DistractionRepo
from focusguard.db.score_repo import ScoreRepo, TipRepo
from focusguard.config import SESSION

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("FocusGuard-Seeder")

def generate_dummy_data(num_users=5, sessions_per_user=15):
    db_manager = DatabaseManager()
    user_repo = UserRepo()
    session_repo = SessionRepo()
    usage_repo = AppUsageRepo()
    distraction_repo = DistractionRepo()
    score_repo = ScoreRepo()
    tip_repo = TipRepo()

    # 1. Seed Tips
    tips = [
        ("The Pomodoro Technique: 25m focus, 5m break.", "Method"),
        ("Deep Work: Eliminate all distractions for 90m.", "Method"),
        ("Stay hydrated: Drink a glass of water every hour.", "Health"),
        ("Stretch your back and neck every 30 minutes.", "Health"),
        ("The 2-minute rule: If it takes < 2m, do it now.", "Productivity"),
        ("Use 'Eat the Frog': Do the hardest task first.", "Productivity"),
        ("Binaural beats can help increase concentration.", "Audio"),
        ("A clean workspace leads to a clear mind.", "Environment"),
    ]
    for content, cat in tips:
        tip_repo.add_tip(content, cat)

    # 2. Seed Users
    usernames = ["FocusMaster", "DeepWorkPro", "ZenCoder", "Sushant", "ProductivityNinja"]
    for name in usernames:
        try:
            user_repo.create_user(name)
        except Exception:
            pass # User already exists

    # 3. Generate Sessions and related data
    for name in usernames:
        user = user_repo.get_user_by_username(name)
        if not user: continue

        # Generate sessions over the last 30 days
        for i in range(sessions_per_user):
            # Random date in last 30 days
            days_offset = random.randint(0, 30)
            hours_offset = random.randint(8, 20) # Between 8am and 8pm
            start_dt = datetime.now() - timedelta(days=days_offset)
            start_time = start_dt.replace(hour=hours_offset, minute=random.randint(0, 59)).strftime("%Y-%m-%d %H:%M:%S")

            # Session details
            duration = random.randint(600, 3600) # 10 to 60 mins
            state = "COMPLETED" if random.random() > 0.2 else "ABANDONED"
            score = float(random.randint(40, 100)) if state == "COMPLETED" else None

            session = session_repo.create_session(user.id, start_time, state)
            session_repo.update_session(
                session.id,
                end_time=(datetime.strptime(start_time, "%Y-%m-%d %H:%M:%S") + timedelta(seconds=duration)).strftime("%Y-%m-%d %H:%M:%S"),
                duration=duration,
                state=state,
                score=score
            )

            # Log usage and distractions for this session
            # Focused app vs Distraction app
            focused_app = random.choice(["VS Code", "PyCharm", "Notion", "Obsidian", "Terminal"])
            distraction_app = random.choice(["Chrome", "Discord", "Twitter", "YouTube", "Instagram"])

            # Add focused usage
            usage_repo.log_usage(session.id, focused_app, duration * 0.8, start_time)

            # Add a few distractions
            num_distractions = random.randint(0, 5) if state == "COMPLETED" else random.randint(3, 10)
            for _ in range(num_distractions):
                distraction_repo.log_distraction(session.id, distraction_app, "BLOCKED_APP")
                usage_repo.log_usage(session.id, distraction_app, 60, start_time) # each distraction 1 min

            # Log the score
            if score:
                score_repo.add_score(user.id, session.id, score)

            # Update user points based on session (simplified)
            user_repo.update_points(user.id, 10 if state == "COMPLETED" else -5)

    logger.info(f"Successfully seeded data for {num_users} users and approx {num_users * sessions_per_user} sessions.")

if __name__ == "__main__":
    generate_dummy_data()