Transforms
Overview
A Transform in Condense is a real-time processing component that applies business logic, enrichment, or routing to streaming events within a pipeline.
It operates inside a workspace pipeline and processes messages as they flow between source connectors and sink connectors.
Transforms help:
Apply custom or prebuilt processing logic to streaming data without leaving Condense.
Implement domain-specific rules, validation, and enrichment directly in the event flow.
Maintain consistency by executing processing in the same environment as ingestion and delivery.
Enable both no-code and code-based logic through prebuilt utilities or custom applications.
What You Can Do with a Transform
Prebuilt Transforms
Ready-to-use, maintained by Condense, and optimized for low-latency execution.
Examples include
Geofence: Location-based event triggering.
Configuration is done entirely in the UI, no coding required.
Within the Prebuilt category, Condense also offers Utilities lightweight, no-code operators designed for essential, recurring logic tasks that can be quickly dropped into a pipeline without development effort.
Utilities
Utilities are lightweight, no-code operators provided under the Prebuilt category for essential, recurring logic.
Alerts
Function: Triggers alert events when a specified condition remains true for a defined duration.
Key Parameters
Duration: Time threshold for condition persistence.
Group By: Field to segment conditions (e.g., device ID).
Rule Groups: AND/OR logical rules for evaluation.
Logger
Function: Streams workspace logs to an external logging destination for monitoring and audit.
Key Parameters
Utility Name: Logical identifier.
Environment Variables: Logging target details.
Split
Function: Routes messages into different paths based on evaluated conditions.
Key Parameters
Rule Groups: Field, type, comparator, and value-based conditions.
Output Mapping: Definition of message structure per route.
Note: Additional utilities will be available in upcoming releases.
Custom Transforms
Developed in the Applications section of Condense using the inbuilt IDE, connected to a Git repository.
Support any programming language in Condense IDE.
Can perform advanced processing, cross-event correlation, or domain-specific analytics.
Once published, appear alongside prebuilt transforms for deployment into pipelines.
Transform Lifecycle
Every transform in Condense follows a defined lifecycle from creation and configuration to updates, monitoring, and eventual removal. This section explains the key stages in that lifecycle and the actions available at each step, starting with how to deploy a transform.
Deploy a Transform
Prebuilt Transform
Steps :
Go to the workspace pipeline editor.
Click Add Transform.
Select from the Prebuilt category or Utilities.
Configure parameters, rules, and mappings.
Save and deploy.
Custom Transform
Who can create: Workspace Admins, Maintainers, Developers
Steps:
Go to Applications in the workspace.
Create a new application in the IDE or import from Git.
Implement and test the logic.
Publish the application as a Custom Transform.
Deploy it to the pipeline.
Editing a deployed Transform
Editable: Rules, mappings, parameters, and, for custom transforms, the associated application version.
Non-editable: Name of the transform
Editing Deployed Transform
In the pipeline canvas, select the deployed transform block you wish to modify, then click the Edit button in the top-right corner.
Editable parameters vary depending on the transform type. Fields that cannot be modified will appear disabled with a lock icon, while editable fields will remain active. After making changes, click Save.
Deleting a custom Transform
Steps to delete the deployed
Warning!
Deletion is permanent; removed transforms cannot be restored.
All processing logic tied to the transform is removed from the pipeline and will no longer execute.
Transform Roles and Permissions
Transform Roles and Permissions
Roles are assigned per workspace, so a user’s transform permissions depend on their workspace role.
Operation
Admin
Maintainer
Developer
Viewer
Add prebuilt transform
✅
✅
✅
❌
Add custom transform
✅
✅
✅
❌
Configure transform parameters
✅
✅
✅
❌
Deploy transform to pipeline
✅
✅
❌
❌
Edit transform configuration
✅
✅
✅
❌
Delete transform
✅
✅
❌
❌
View transform list/config
✅
✅
✅
✅
Access transform logs
✅
✅
✅
✅
Role descriptions:
Admin: Full control over transforms and other pipeline resources.
Maintainer: Can create, configure, and deploy transforms, but not manage workspace settings.
Developer: Can create and configure transforms, but cannot deploy or delete them.
Viewer: Read-only access to transform details.
Monitoring and Audit
Transforms include monitoring and auditing tools for operational visibility:
Status (active, error)
Execution metrics (throughput, processing latency, error rate)
Version history for each transform
Activity logs for creation, modification, and deletion events
Sample Screens for Monitoring Deployed Transforms
In the pipeline view, click on a deployed transform block to open its detail page. Here you can view resource utilization metrics, execution statistics, and status information specific to that transform.
Best Practices
Use prebuilt transforms for rapid deployment of common logic.
Keep custom transform logic modular and version-controlled in Git.
Document each transform’s purpose and configuration for team clarity.
Regularly review transform performance metrics to detect bottlenecks early.
Common Pitfalls and How to Avoid Them
Deploying too many transforms in sequence Avoidance: Consolidate logic where possible to reduce processing latency.
Incorrect RBAC assumptions Avoidance: Confirm role permissions before assigning transform-related tasks.
Using the wrong transform type Avoidance: For quick standard tasks, use prebuilt utilities instead of building custom transforms.
Not testing before deployment Avoidance: Always validate logic using sample data before pushing to production.
Related Links
Last updated
Was this helpful?