---

Introduction

Enterprise applications communicate using different message formats.

For example:

• Mobile App → JSON
• Banking Core → Java Objects
• Payment Gateway → XML
• Kafka → Avro
• Legacy Systems → CSV
• External APIs → REST JSON

Since every system speaks a different "language", messages must be converted before processing.

Spring Integration provides Transformers to convert one message format into another without modifying business logic.

A Transformer changes the message payload while preserving the message flow.

---

Transformer Architecture

flowchart LR

InputMessage --> Transformer

Transformer --> OutputMessage

OutputMessage --> ServiceActivator

---

Q1. What is a Transformer?

Answer

A Transformer converts one message into another.

It typically changes

• Payload format
• Payload type
• Payload structure

while preserving message headers unless explicitly modified.

Benefits

• Loose coupling
• Reusable conversion logic
• Clean integration flows
• Independent business services

---

Q2. Why do we need Transformers?

Without Transformers

Business services must understand every external message format.

Example

JSON

↓

Business Logic

↓

XML

↓

CSV

This mixes conversion logic with business logic.

With Transformers

flowchart LR

JSON --> Transformer

Transformer --> JavaObject

JavaObject --> BusinessService

Business services process only domain objects.

---

Q3. What does a Transformer modify?

A Transformer can modify

• Payload
• Headers (optional)

It should not perform business logic.

Example

JSON

↓

Customer Object

Headers such as

• Correlation ID
• Timestamp
• Priority

are normally preserved.

---

Q4. How do you create a Transformer?

Example

@Bean

IntegrationFlow

customerFlow(){

    return flow -> flow

    .transform(

    CustomerTransformer::convert);

}

Custom transformer

public Customer

convert(

String json){

}

---

Q5. What built-in Transformers are available?

Spring Integration provides many built-in transformers.

Examples

• Object → JSON
• JSON → Object
• Object → String
• XML → Object
• Object → XML
• Byte Array → String
• String → Byte Array

Transformation Flow

flowchart TD

JSON --> Transformer

Transformer --> JavaObject

JavaObject --> XML

XML --> CSV

---

Q6. What is Header Enrichment?

Header Enricher adds or updates message headers.

Example

.enrichHeaders(

h -> h

.header(

"source",

"MOBILE"))

Result

Payload

Payment Request

Headers

source=MOBILE

priority=HIGH

Header enrichment avoids modifying payloads for metadata.

---

Q7. Transformer vs Filter

Transformer	Filter
Converts message	Accepts or rejects
Output is modified	Output may not exist
Always forwards	May discard message
Data conversion	Validation

Use

• Transformer → Convert data
• Filter → Validate data

---

Q8. Transformer vs Service Activator

Transformer	Service Activator
Data conversion	Business processing
Stateless	Business logic
Reusable	Domain-specific
Changes payload	Performs operations

Transformers prepare messages.

Service Activators execute business functionality.

---

Q9. What are common Transformer use cases?

Typical enterprise scenarios

• JSON → Java Object
• Java Object → XML
• CSV → Entity
• Entity → DTO
• XML → REST Request
• Kafka Event → Domain Object
• File → Business Object

Transformers isolate protocol-specific conversions.

---

Q10. Transformer Best Practices

Keep Transformers Stateless

Avoid shared mutable state.

---

Perform Only Data Conversion

Do not embed business rules.

---

Preserve Headers

Unless intentionally changing metadata.

---

Reuse Common Transformers

Avoid duplicate conversion logic.

---

Validate Before Transforming

Use Filters before Transformers when possible.

---

Banking Example

flowchart TD

RESTRequest --> JSONTransformer

JSONTransformer --> PaymentObject

PaymentObject --> HeaderEnricher

HeaderEnricher --> PaymentProcessor

PaymentProcessor --> XMLTransformer

XMLTransformer --> CoreBanking

Each transformation is isolated and reusable.

---

Common Interview Questions

• What is a Transformer?
• Why use Transformers?
• What does a Transformer modify?
• How do you create a custom Transformer?
• Built-in Transformers?
• What is Header Enrichment?
• Transformer vs Filter?
• Transformer vs Service Activator?
• Common Transformer use cases?
• Transformer best practices?

---

Quick Revision

Topic	Summary
Transformer	Converts message payload
Payload	Business data
Headers	Message metadata
Header Enricher	Adds or updates headers
JSON Transformer	JSON ↔ Object
XML Transformer	XML ↔ Object
Object Mapper	Data conversion
Filter	Validates messages
Service Activator	Executes business logic
IntegrationFlow	Processing pipeline

---

Transformer Lifecycle

sequenceDiagram
Gateway->>Channel: Send JSON
Channel->>Transformer: Convert JSON
Transformer->>HeaderEnricher: Add Metadata
HeaderEnricher->>ServiceActivator: Business Object
ServiceActivator->>OutboundAdapter: Send XML
OutboundAdapter-->>Gateway: Response

---

Production Example – Banking Payment Transformation

A banking application receives payment requests from multiple channels.

Incoming Formats

• Mobile Banking → JSON
• ATM → XML
• Branch Application → Java Object
• SWIFT Gateway → ISO 20022 XML

Workflow

1. A JSON Transformer converts mobile requests into PaymentRequest.
2. A Header Enricher adds metadata such as channel=MOBILE and priority=HIGH.
3. The payment is processed by the business service.
4. An XML Transformer converts the response into ISO 20022 format before sending it to the Core Banking System.

.transform(
    Transformers.fromJson(
        PaymentRequest.class
    )
)
.enrichHeaders(
    h -> h.header(
        "channel",
        "MOBILE"
    )
)

flowchart LR

MobileApp --> JSONMessage

JSONMessage --> JSONTransformer

JSONTransformer --> PaymentRequest

PaymentRequest --> HeaderEnricher

HeaderEnricher --> PaymentService

PaymentService --> XMLTransformer

XMLTransformer --> CoreBankingSystem

This design keeps conversion logic independent of business processing, making the application easier to maintain and extend.

---

Key Takeaways

• Transformers convert messages from one format to another without introducing business logic.
• They primarily modify the payload while usually preserving message headers.
• Spring Integration provides built-in transformers for JSON, XML, Strings, byte arrays, and other common formats.
• Header Enrichers add metadata without changing the payload.
• Transformers should remain stateless, reusable, and focused solely on data conversion.
• Filters validate messages, while Transformers convert message content and Service Activators execute business operations.
• Separating transformation logic from business logic improves maintainability, testability, and reusability.
• Enterprise integration solutions rely heavily on transformers to connect systems that exchange data in different formats.