Getting started

1) Add the Maven profile

Add this profile section to your pom.xml:

<profiles>
    <profile>
        <id>generate-resources</id>
        <build>
            <plugins>
                <plugin>
                    <groupId>dev.markozivkovic</groupId>
                    <artifactId>spring-crud-generator</artifactId>
                    <version>1.8.0</version>
                    <executions>
                        <execution>
                            <id>generate-spring-crud</id>
                            <goals>
                                <goal>generate</goal>
                            </goals>
                            <configuration>
                                <inputSpecFile>
                                    ${project.basedir}/src/main/resources/crud-spec.yaml
                                </inputSpecFile>
                                <outputDir>
                                    ${project.basedir}/src/main/java/com/sql/demo/springboot_postgres_json_demo
                                </outputDir>
                                <forceRegeneration>true</forceRegeneration>
                            </configuration>
                        </execution>
                    </executions>
                </plugin>
            </plugins>
        </build>
    </profile>
</profiles>

Parameters

Parameter	Description
inputSpecFile	Path to the YAML/JSON configuration file
outputDir	Directory where generated source code is written
forceRegeneration	Forces regeneration of all non-ignored entities, ignoring state files

2) Create the spec file

Create the file defined by inputSpecFile. The file name is arbitrary. Only the extension is required: .yaml, .yml, or .json.

SQL databases (PostgreSQL, MySQL, MariaDB, MSSQL)

configuration:
  database: postgresql   # or: mysql, mariadb, mssql
  openApi:
    apiSpec: true
    generateResources: true
  errorResponse: simple
  tests:
    unit: true
    dataGenerator: instancio
entities:
  - name: ProductModel
    storageName: product_table    # SQL table name
    description: "Represents a product"
    fields:
      - name: id
        type: Long
        description: "The unique identifier for the product"
        id:
          strategy: IDENTITY
      - name: name
        description: "The name of the product"
        type: String
        column:
          nullable: false
          unique: true
          length: 255

MongoDB

configuration:
  database: mongodb
  openApi:
    apiSpec: true
    generateResources: true
  errorResponse: simple
  tests:
    unit: true
    dataGenerator: instancio
entities:
  - name: ProductModel
    storageName: products         # MongoDB collection name
    description: "Represents a product"
    fields:
      - name: id
        type: String
        id: true                  # MongoDB document id — use id: true instead of id.strategy
        description: "MongoDB document id"
      - name: name
        description: "The name of the product"
        type: String
        validation:
          required: true
          notBlank: true
          maxLength: 255

Full example specs:

• SQL: crud-spec-full.yaml
• MongoDB: mongo-crud-spec-full.yaml

Schema for validation and editor autocomplete: crud-spec.schema.json Security configuration reference (all supported modes and examples): configuration.md#configurationsecurity

Schema-based autocomplete and validation

Schema hints are optional, but recommended.

They are the easiest way to enable validation and autocomplete for spec files with arbitrary names. If your editor is already configured to associate these files with crud-spec.schema.json, you do not need to add the schema hint manually.

For YAML files, add a schema hint comment:

# yaml-language-server: $schema=https://raw.githubusercontent.com/mzivkovicdev/spring-crud-generator/main/docs/schema/crud-spec.schema.json
configuration:
  database: postgresql

For JSON files, use the $schema property:

{
  "$schema": "https://raw.githubusercontent.com/mzivkovicdev/spring-crud-generator/main/docs/schema/crud-spec.schema.json",
  "configuration": {
    "database": "postgresql"
  }
}

Works in editors that support JSON Schema, including YAML editors that use yaml-language-server.

3) Validate spec (dry-run, optional)

Use the validate goal to verify spec correctness before generation:

mvn spring-crud-generator:validate -DinputSpecFile=src/main/resources/crud-spec.yaml

4) Run generation

mvn clean install -Pgenerate-resources -DskipTests

After execution:

• source code is generated into the directory defined by outputDir
• Flyway .sql migration scripts are generated (if enabled, SQL databases)
• Mongock @ChangeUnit Java classes are generated (if enabled, MongoDB)
• Swagger/OpenAPI resources are generated (if enabled)
• Docker and Docker Compose files are generated (if enabled)
• GraphQL resources (schema, resolvers, mappers etc.) are generated (if enabled)

For the validate goal, only inputSpecFile is required.

Notes

Incremental generation is enabled by default:

• Generator state is stored in: .crud-generator/generator-state.json
• Migration state: .crud-generator/migration-state.json (SQL) or .crud-generator/mongock-state.json (MongoDB)
• Setting forceRegeneration=true ignores the generator state
• Tables and collections are never dropped automatically