@SpringBootApplication — What It Combines

@SpringBootApplication is a single annotation that composes three: @SpringBootConfiguration (this class is a primary bean source enforced as singleton per context), @EnableAutoConfiguration (activate Spring Boot's auto-configuration), and @ComponentScan (scan from this class's package downward).

The scan boundary rule

• The component scan starts from the package containing the @SpringBootApplication class
• Place the main class in the root package (com.example) so every sub-package (com.example.service, com.example.repository) is automatically discovered
• A main class buried inside a sub-package means sibling packages are outside the scan boundary — beans there are invisible

Customizing the scan

• @SpringBootApplication(scanBasePackages = "com.example") extends the boundary, but string-based — renaming the package breaks it silently
• scanBasePackageClasses = {ServiceMarker.class} is refactoring-safe — a compile error tells you immediately when the marker class moves

Excluding auto-configurations

• @SpringBootApplication(exclude = {DataSourceAutoConfiguration.class}) prevents that auto-config from running — useful when a starter pulls in database config you do not want
• Type-safe and visible at the annotation site; prefer this over the string-based property approach

Test discovery

• @SpringBootTest walks up from the test class to find one @SpringBootConfiguration
• Two configurations cause ambiguous configuration errors
• @SpringBootTest(classes = App.class) bypasses auto-discovery when class hierarchies don't align

In an interview, the strong answer is: the three annotations are easy; the test is whether you understand the scan boundary, the singleton constraint, and refactoring-safe scan configuration.