# Automated SDLC Pipeline with Tray APIs

Setting Up an Automated SDLC Pipeline with Tray APIs

### Prerequisites

Before setting up the automated SDLC pipeline using Tray’s Projects and Solutions APIs, ensure the following prerequisites are met:

* **Documentation:** Familiarize yourself with Tray’s public developer documentation site at [developer.tray.io](https://developer.tray.io/), which provides a comprehensive reference for the [**Projects**](https://tray.ai/documentation/developer/platform-apis/projects) and [**Solutions**](https://tray.ai/documentation/developer/platform-apis/solutions) API endpoints, including intended use cases and available functions.
* **Source and destination workspaces:** You need an existing source and destination workspace for project deployment. Projects cannot currently be created via API, so they must be manually set up.
* **Authentication:** You must have valid authentication tokens (user or master tokens) to interact with Tray APIs. Tokens are environment-specific and region-sensitive.
* **Auth matching:** Ensure that all required authentications exist in both source and destination workspaces. Auth IDs must be mapped accordingly.
* **Versioning:** Declare a version of the project before attempting to deploy it. Versions can be created via the UI or API.

### Setup process

1. **Retrieve source project version information**

* **Purpose:** Ensures that you are deploying the correct version of the project.
* Call the `GET /core/v1/projects/{projectId}/versions `endpoint to fetch available versions.
* Ensure the latest version is retrieved for deployment to prevent outdated configurations.

2. **Export the source project version**

* **Purpose:** Exports the project version in a format suitable for migration.
* Use `GET /core/v1/projects/{projectId}/versions/{versionNumber}/export`
  to fetch the project version JSON.
* Convert the stringified JSON into a structured object for further processing to avoid parsing issues later.

3. **Check import requirements** (optional for initial deployments or authentication changes)

* **Purpose:** Determines whether additional configuration is needed before importing the project.
* Call  `POST /core/v1/projects/{projectId}/imports/requirements `with the exported project JSON.
* Identify any required authentication mappings or configuration mappings that need to be resolved.

4. **Preview the import**

* **Purpose:** Simulates the import process to detect potential issues before making changes.
* Call  `POST /core/v1/projects/{projectId}/imports/previews `to analyze potential changes before deployment.
* Ensure there are no conflicts or missing dependencies, preventing failed deployments.

5. **Import the project version**

* **Purpose:** Overwrites the project in the destination workspace with the new version.
* Call  `POST /core/v1/projects/{projectId}/imports `with the required mappings (that can be hardcoded in your pipeline) and project JSON.
* Validate the response for any errors to confirm a successful import.

6. **Create a version in the destination workspace**

* **Purpose:** Establishes a restore point for rollback capabilities in case of future issues and ensures version numbering consistency across environments.
* Call  `POST /core/v1/projects/{projectId}/versions/{versionNumber} `to create a version
* It is recommended to match the source version ID in the destination workspace to maintain consistency.

#### Additional steps for embedded solutions

1. Ensure solution exists
2. **Retrieve solution ID:** After import, use the response payload to extract the new solution ID.
3. **Publish the solution:**

* Call  `POST /core/v1/solutions/{solutionId}/releases/previews `to to analyze potential changes and solution instances impact before publishing.
* Call  `POST /core/v1/solutions/{solutionId}/releases `to publish the solution.

### Pipeline setup using Tray

You can build your automated pipeline using Tray workflows. [This video](https://www.loom.com/share/b369ce4146704e8693ba3b0e7a28ecb8) provides a walkthrough of an example implementation.
Here are some best practices for implementing pipelines in Tray:

* **Use config:** Store workspace IDs, auth mappings, and base URLs in config.
* **Implement error handling:** Monitor API responses and handle errors gracefully to allow retrying failed steps.
* **Leverage logs:** Utilize logs for debugging.
* **Automate with webhooks:** Consider triggering deployments automatically based on repository changes or CI/CD pipelines.
* **Ensure region compatibility:** Use the correct base URLs based on the region to prevent cross-region deployment issues.