Spring Boot and MongoDB?

Question

Accepted Answer

**Spring Data MongoDB** — a Spring module for working with **MongoDB** (NoSQL, document database). **Simple analogy:** MongoDB is like a cabinet with folders. In each folder (collection) are documents (JSON), each with its own set of fields. No fixed structure (schema) is needed. Imagine a filing cabinet where each folder can hold documents of completely different formats — one might be a receipt, another a photo, another a letter. This flexibility is MongoDB's core strength. **Why it matters:** MongoDB is ideal when your data model evolves rapidly, when you have deeply nested documents (like product catalogs with varying attributes), or when you need horizontal scalability through sharding. Spring Data MongoDB makes working with it as seamless as JPA makes working with relational databases. **Connection:** ```yaml # application.yml spring: data: mongodb: uri: mongodb://localhost:27017/mydb auto-index-creation: true ``` **Document:** ```java @Document(collection = "users") @Indexed(compoundIndex = @CompoundIndex(name = "email_status", def = "{'email': 1, 'status': 1}")) public class User { @Id private String id; private String name; private String email; @Field("created_at") private LocalDateTime createdAt; @DBRef // Reference to another document private Role role; @Version private Long version; // Optimistic locking } ``` **Repository:** ```java public interface UserRepository extends MongoRepository { Optional findByEmail(String email); List findByNameContainingIgnoreCase(String name); @Query("{'age': {'$gte': ?0, '$lte': ?1}}") List findByAgeRange(int min, int max); List findByRoleName(String roleName); // Nested document @Aggregation(pipeline = "{ "$group": {"_id": "$status", "count": {"$sum": 1}} }") List countByStatus(); } ``` **MongoTemplate (low-level):** ```java @Service public class UserService { @Autowired private MongoTemplate mongoTemplate; public List findActiveUsers() { Query query = new Query(); query.addCriteria(Criteria.where("status").is("ACTIVE")); query.with(Sort.by(Sort.Direction.DESC, "createdAt")); query.limit(10); return mongoTemplate.find(query, User.class); } public void upsert(User user) { Query query = new Query(Criteria.where("email").is(user.getEmail())); Update update = new Update() .set("name", user.getName()) .set("updatedAt", LocalDateTime.now()); mongoTemplate.upsert(query, update, User.class); } } ``` **Common mistakes:** - **N+1 queries with `@DBRef`** — each lazy-loaded reference triggers a separate query. Use `@DBRef(lazy = false)` or fetch eagerly. - **Missing indexes** — MongoDB collections without indexes scan every document. Add indexes for frequently queried fields. - **Using MongoDB as a relational database** — if you need complex joins and transactions across many tables, use PostgreSQL instead. **When MongoDB is better than PostgreSQL:** - Data structure changes frequently. - Horizontal scalability (sharding) is needed. - Data is nested documents (product catalog with specifications). - Rapid prototyping where schema flexibility speeds up development.

Java: Spring Boot and MongoDB?