Salesforce → Salesforce Mulesoft Integration

Project: Salesforce to Salesforce Listing & Image Sync
Description: Implementation of Mulesoft as a middleware in between two Salesforce Instances.
Project Duration: June-25- Oct-25

Executive summary

Integrated Salesforce (advertisement and media data) with TARGET SALESFORCE SYSTEM to automate listing creation, updates, image uploads and deletions. The solution listens to Salesforce streaming events, enriches data via SOQL queries, transforms records into TARGET SALESFORCE SYSTEM payloads using DataWeave, and calls TARGET SALESFORCE SYSTEM REST APIs with OAuth2 token management. The integration reduced manual work, improved time-to-market for listings, and provided a reusable, secure API-led architecture for future channels.

Business problem

  • Manual publishing and image uploads from Salesforce to TARGET SALESFORCE SYSTEM (error-prone and slow).
  • Lack of reliable sync between Salesforce advertisement records and TARGET SALESFORCE SYSTEM listings.
  • Need for retryable, observable communication with TARGET SALESFORCE SYSTEM APIs and secure token management.

High-level solution overview

Two System APIs were implemented following MuleSoft’s API-led connectivity model to isolate concerns:

1. SFDC System API

  • Listens to Platform Events (Advertisement_Sync_Event__e)

  • Executes SOQL queries on Advertisement, Vehicle, and ModelVariant objects

  • Transforms data into the required Target Salesforce payload format using DataWeave 2.0

  • Triggers the Target Salesforce System API for further operations

2. Target Salesforce System API

  • Manages listing lifecycle (create, update, delete)

  • Handles image uploads, image set operations, and token management

  • Uses Object Store for OAuth token caching

  • Implements retry (Until Successful) and error-handling mechanisms for reliability

Architecture (Conceptual)

Salesforce Platform Events —> SFDC Sys API (Replay Listener)
|– SOQL: Advertisement, Vehicle, ModelVariant
|– DataWeave transforms –> Target Salesforce Sys API (HTTP)

Target Salesforce Sys API:
– Retrieve / Cache Access Token (ObjectStore)
– Create / Update / Delete Listing
– Upload Images (multipart/form-data)
– Set images on listing (PUT)

Monitoring & Resilience:
– Until Successful retries
– Global Error Handler + Logging
– Observability via Logger

Key Components & Flows

SFDC System API

  • Target Salesforce System-impl-flow: Replay-channel listener for Advertisement_Sync_Event__e

  • Executes SOQL queries to fetch Advertisement__c, Vehicle__c, and ModelVariant__c records

  • Uses DataWeave to prepare the listing payload

  • Triggers create/update/delete listing flows in the Target System

  • Manages image workflows by:

    • Querying FileUpload__c custom object

    • Fetching binary data from ContentVersion

    • Posting to Target Salesforce System image endpoint

    • Updating Salesforce with returned content IDs

Target Salesforce System API

  • sync-listing flow determines the operation (create/update/delete)

  • create-listing: POST request to Target Salesforce System

  • update-listing: PUT request

  • delete-listing: DELETE request

  • upload-image: Builds multipart/form-data body and uploads image

  • set-imageids: Attaches image IDs to listings

  • get-token & cache-token: Fetches and stores OAuth tokens with expiry handling

Technical Highlights

  • Data Transformations: Advanced DataWeave 2.0 mapping from Salesforce SObjects to Target payloads

  • Token Management: Object Store caching with TTL; automatic token renewal on 401 errors

  • Resilience & Retries: External API calls wrapped in Until Successful scopes

  • Observability: Consistent Logger placement and a Global Error Handler for uniform error output

  • File Handling: Binary conversions from Salesforce ContentVersion.VersionData using dw::core::Binaries

  • API Configurations: Centralized HTTP request configurations for dynamic environments

  • Salesforce Connectors: Used salesforce:replay-channel-listener, salesforce:query-all, and salesforce:update

Responsibilities (Delivered by GAGSTEK)

  • Designed end-to-end API-led integration between Salesforce and Target Salesforce System

  • Implemented event-based ingestion with Replay Listener and SOQL enrichment

  • Developed System APIs for listing lifecycle and image operations

  • Implemented Object Store caching for optimized OAuth token usage

  • Added retries, error handling, and DataWeave transformations for reliability

  • Configured binary handling for image uploads

  • Built monitoring hooks compatible with Anypoint Monitoring

Results & Business Benefits

  • ✅ Automated listing lifecycle (create, update, delete) eliminating manual errors

  • ✅ Reduced time-to-market for listings

  • ✅ Improved traceability with logs and Salesforce record updates

  • ✅ Reusable APIs enabling rapid onboarding of new marketplaces

Challenges & Solutions

ChallengeSolution
Token expiry & race conditionsUsed Object Store for token caching; re-fetch token and retry on 401
Binary image handlingConverted VersionData to multipart/form-data using DataWeave and Binaries module
Intermittent API errorsWrapped REST calls in Until Successful with controlled backoff and retries
Idempotency in listing creationExtracted listingId from Location header and stored in Salesforce

Security & Governance

  • OAuth2 token-based authentication

  • Secure storage of client ID/secret and Salesforce credentials

  • Object Store TTL to prevent stale tokens

  • Audit logs for every critical flow for investigation and monitoring

Reusable Patterns & Recommendations

  • Continue using Object Store for any external OAuth provider integration

  • Apply Until Successful for idempotent API calls

  • Abstract HTTP request configs into properties files

  • Follow System → Process → Experience API separation for scalability

Client Testimonial

“GAGSTEK built a resilient, event-driven integration that automatically syncs Salesforce advertisements and images to Target Salesforce System24.
It manages OAuth tokens securely, handles retries for transient failures, converts and uploads image binaries, and writes back identifiers to Salesforce — providing a fully auditable, low-touch publishing pipeline reusable for other marketplaces.”

Appendix — Useful Snippets & Notes

1. Token Storage (Object Store)

<os:store key=”Target_SF_access_token” objectStore=”Object_store” />

2. Example: Building Multipart Payload for Image Upload (DataWeave 2.0)

%dw 2.0
import * from dw::core::Binaries
output multipart/form-data
—
{
parts: {
file: {
content: Binaries.fromBase64(payload.binary),
headers: {
“Content-Disposition”: “form-data; name=\”file\”; filename=\”” ++ payload.fileName ++ “\””,
“Content-Type”: payload.mimeType
}
}
}
}

3. Retry Pattern

<until-successful maxRetries=”3″ millisBetweenRetries=”3000″>
<!– http request –>
</until-successful>

4. Salesforce Replay Listener

<salesforce:replay-channel-listener
streamingChannel=”/event/Advertisement_Sync_Event__e”
replayOption=”ONLY_NEW”/>