使用Hibernate自动DDL创建数据库表不符合预期问题咨询

Hey there! Let's work through this table generation issue you're hitting with Hibernate and Spring Boot. I took a look at your Content entity snippet, and here are the most common fixes to get your database table matching exactly what you expect:

A lot of databases (like MySQL with the default lower_case_table_names=1 setting) automatically convert table and column names to lowercase. But your entity uses mixed-case names like @Table(name = "tContent") and @Column(name = "Content_ID")—this mismatch can lead to tables named tcontent or columns named content_id instead of what you specified.

To fix this:

• If your database supports case-sensitive naming (like PostgreSQL, or MySQL with lower_case_table_names=0), make sure your entity's annotations match the database's case rules exactly.
• Or force Hibernate to use your exact naming by adding this to your application.properties:

spring.jpa.hibernate.naming.physical-strategy=org.hibernate.boot.model.naming.PhysicalNamingStrategyStandardImpl

This strategy will strictly follow the names you define in @Table and @Column without automatic conversion.

Since Content extends BaseEntity, any annotations or fields in the parent class could be messing with table generation. For example:

• If BaseEntity isn't marked with @MappedSuperclass, Hibernate won't map its fields to the Content table.
• Duplicate @Id annotations or conflicting column definitions in BaseEntity can cause unexpected table structures.

Double-check BaseEntity to ensure:

• It has the @MappedSuperclass annotation (so Hibernate knows to inherit its fields).
• There are no conflicting or duplicate field configurations that override your Content entity settings.

The update mode in ddl-auto only makes incremental changes to existing tables—it won't fix already-generated tables that don't match your entity. If you've already got a misconfigured table in your database, you'll need to reset it:

• Temporarily set spring.jpa.hibernate.ddl-auto=create-drop (this creates tables on startup and drops them on shutdown, perfect for dev environments).
• Or manually delete the existing tContent table, then restart your app to let Hibernate regenerate it correctly.

If you have fields in Content without an @Column annotation, Hibernate will use its default naming strategy (converting camelCase to snake_case, e.g., contentBody becomes content_body). If these default names don't fit your requirements, add @Column to every field to specify the exact name.

Here's a cleaned-up example of your entity (including a properly configured BaseEntity):

@MappedSuperclass
public abstract class BaseEntity {
    @Column(name = "Create_Time")
    private LocalDateTime createTime;

    // Add getters and setters here
}

@Entity
@Table(name = "tContent")
public class Content extends BaseEntity implements Serializable {
    private static final long serialVersionUID = 1L;

    @Id
    @Column(name = "Content_ID")
    @GeneratedValue(strategy = GenerationType.IDENTITY)
    private Long id;

    @Column(name = "Title")
    private String title;

    // Explicitly name every column to avoid defaults
    @Column(name = "Content_Body")
    private String contentBody;

    // Add getters and setters here
}

If you're still stuck, enable Hibernate's SQL logging to see exactly what CREATE TABLE statement it's generating. This will let you spot the difference between what Hibernate is doing and what you want.

Add these lines to your application.properties:

logging.level.org.hibernate.SQL=DEBUG
logging.level.org.hibernate.type.descriptor.sql.BasicBinder=TRACE

When you start your app, check the console for the CREATE TABLE output—you'll immediately see where the mismatch is.

内容的提问来源于stack exchange，提问作者Nikesh Kedlaya