Colecciones frente a tablas SQL

Tables vs Collections at a Glance

Fixed Schema: The SQL Way

-- SQL table: schema defined upfront, rigid
CREATE TABLE users (
  id         SERIAL PRIMARY KEY,
  name       VARCHAR(100) NOT NULL,
  email      VARCHAR(200) UNIQUE NOT NULL,
  age        INT,
  created_at TIMESTAMP DEFAULT NOW()
);
-- Every row must have exactly these columns

Flexible Schema: The MongoDB Way

// No schema definition needed - collection created on first insert
db.users.insertOne({ name: 'Alice', email: 'alice@test.com', age: 30 });

// Next insert can have completely different fields
db.users.insertOne({ name: 'Bob', email: 'bob@test.com', company: 'Acme', role: 'admin' });

// Both documents live in the same 'users' collection

Schema vs Schema-Less: The Trade-Off

Normalization vs Denormalization

// SQL normalized: address in separate table
// SELECT u.name, a.city FROM users u JOIN addresses a ON a.user_id = u.id

// MongoDB denormalized: address embedded in user document
{
  _id: ObjectId('...'),
  name: 'Alice',
  address: { city: 'London', zip: 'EC1A' }   // no JOIN needed
}

Creating Collections Explicitly

// Create a capped collection explicitly
db.createCollection('appLogs', {
  capped: true,
  size: 10485760,   // 10 MB maximum size
  max: 50000        // optional: max 50,000 documents
});
// When full, oldest documents are automatically removed

Listing and Dropping Collections

// Useful collection management commands in mongosh
db.getCollectionNames();
// ['users', 'orders', 'products']

db.users.countDocuments({});
// 4823

db.users.stats().storageSize;
// 2097152 (bytes)

// Delete all documents but keep the collection:
db.users.deleteMany({});
// { acknowledged: true, deletedCount: 4823 }

The _id Field and Primary Keys

// Custom _id values
db.products.insertOne({
  _id: 'SKU-HEADPHONES-BLK-42',   // string _id
  name: 'Wireless Headphones Black',
  price: 79.99
});

// Lookup by custom _id is O(log n) via the _id index
db.products.findOne({ _id: 'SKU-HEADPHONES-BLK-42' });

Index Structure Differences

// Index a nested field and an array field
db.users.createIndex({ 'address.city': 1 });
// Now queries on city use an index:
db.users.find({ 'address.city': 'Chicago' });

// Multikey index on array field - indexes each element
db.products.createIndex({ tags: 1 });
db.products.find({ tags: 'electronics' }); // uses multikey index

Transactions: Tables vs Collections

// Single-document atomicity (always available)
// Updating order status and adding a tracking number
db.orders.updateOne(
  { _id: orderId },
  { $set: { status: 'shipped', trackingNumber: 'UPS123456' } }
);
// These two field updates happen atomically - no transaction needed

When to Choose Tables Over Collections

Quick Check

Lesson Recap