backend/seed_sample_data.py

import csv
import os
from datetime import datetime
from bson import ObjectId
from django.core.management.base import BaseCommand
from expense_tracker.utils import MongoDBClient


class Command(BaseCommand):
    help = 'Seed dataset.csv into the shared sample collection in MongoDB Atlas'

    def handle(self, *args, **options):
        from django.conf import settings

        # dataset.csv lives at the project root (one level above backend/)
        # settings.BASE_DIR points to the backend/ directory
        csv_path = os.path.join(settings.BASE_DIR, '..', 'dataset.csv')
        csv_path = os.path.abspath(csv_path)

        if not os.path.exists(csv_path):
            self.stderr.write(self.style.ERROR(f'dataset.csv not found at: {csv_path}'))
            return

        self.stdout.write(f'Reading from: {csv_path}')

        docs = []
        with open(csv_path, newline='', encoding='utf-8') as f:
            reader = csv.DictReader(f)
            for row in reader:
                try:
                    date_obj = datetime.strptime(row['Date'].strip(), '%Y-%m-%d')
                except ValueError:
                    date_obj = datetime.now()

                doc = {
                    '_id': ObjectId(),
                    'title': row['Title'].strip(),
                    'amount': float(row['Amount'].strip()),
                    'category': row['Category'].strip(),
                    'date': date_obj,
                    'type': row['Type'].strip(),        # 'Income' or 'Expense'
                    'source': 'sample',
                    'created_at': datetime.now(),
                }
                docs.append(doc)

        if not docs:
            self.stderr.write(self.style.ERROR('No rows found in CSV.'))
            return

        db = MongoDBClient.get_client()

        # Drop existing sample data and replace with fresh seed
        self.stdout.write('Dropping existing sample collection...')
        db.sample.drop()

        self.stdout.write(f'Inserting {len(docs)} documents into sample collection...')
        db.sample.insert_many(docs)

        # Quick stats
        income_count = sum(1 for d in docs if d['type'].lower() == 'income')
        expense_count = len(docs) - income_count

        self.stdout.write(self.style.SUCCESS(
            f'✅ Done! Seeded {len(docs)} sample transactions '
            f'({income_count} income, {expense_count} expense) '
            f'into the sample collection.'
        ))