Export and import a pipeline configuration

Overview

Use export and import to move a declarative pipeline project between projects, spaces, tenants, or environments without relying on a connected GitHub repository.

The workflow on this page shows the API-based process. You can also export and import project packages from the Qlik Cloud user interface.

A project package contains files such as qtcp_project.yaml, qtcp_bindings_definition.json, and files under qtcp_tasks/.

Export modes

When exporting through the API, use the mode field to choose the export mode:

• MINIMAL: YAML project structure with default values omitted.
• ALL: YAML project structure with all values included.
• LEGACY: legacy JSON project structure.

When declarative pipelines are enabled, MINIMAL is the default export mode.

Use MINIMAL for day-to-day deployment and promotion workflows. The resulting YAML contains only properties that differ from server defaults, which keeps package contents and diffs smaller.

Use ALL when you need to inspect all available properties or discover the full schema.

Use LEGACY only for backward compatibility with older import workflows.

Binding values are not included in exported packages by default. To include bindings.json in the exported package, set includeBindings to true.

Binding values

Binding variables separate the project definition from environment-specific values such as connections, schemas, warehouses, and other target-environment settings.

Binding variable declarations are stored in qtcp_bindings_definition.json. Use this file to identify which variables need values in the target environment.

Export and import workflow

Use this workflow to export a project package from a source environment and import it into a target project.

Step 1: Export the source project package

Call the export endpoint for the source project:

curl -X POST "https://<TENANT_URL>/api/v1/di-projects/<SOURCE_PROJECT_ID>/actions/export" \
  -H "Authorization: Bearer <BEARER_TOKEN>" \
  -H "Accept: application/octet-stream" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "MINIMAL"
  }' \
  --output my-project.zip

To include resolved binding values from the source environment, add includeBindings:

{
  "mode": "MINIMAL",
  "includeBindings": true
}

Step 2: Identify the required binding values

Inspect qtcp_bindings_definition.json in the exported package to identify the variables that need values in the target environment.

If the package includes bindings.json, you can use it as a reference for the source environment values.

Step 3: Create or select a target project

Import requires an existing target project. Create a target project in the target space, or select an existing compatible project.

The required project properties depend on the project type and target platform.

The following example creates a data pipeline project:

curl -X POST "https://<TENANT_URL>/api/v1/di-projects" \
  -H "Authorization: Bearer <BEARER_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "<PROJECT_NAME>",
    "space": "<SPACE_ID>",
    "type": "DATA_PIPELINE",
    "platformType": "<PLATFORM_TYPE>",
    "platformConnection": "<CONNECTION_ID>"
  }'

Save the target project ID from the response.

Step 4: Set binding values for the target project

If your zip file does not include the bindings.json file, you need to set the target environment values before importing the package by using the Bindings API:

curl -X PUT "https://<TENANT_URL>/api/v1/di-projects/<TARGET_PROJECT_ID>/bindings" \
  -H "Authorization: Bearer <BEARER_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "variables": {
      "<VARIABLE_NAME>": "<TARGET_VALUE>"
    }
  }'

Repeat the variable entries for each binding value required by the project.

Step 5: Optionally validate the package

You can run the validation endpoint before importing the project package. The validation endpoint currently performs limited checks, such as verifying the ZIP file and checking that task definitions include settings.

curl -X POST "https://<TENANT_URL>/api/v1/di-projects/utils/actions/validate-project-definitions" \
  -H "Authorization: Bearer <BEARER_TOKEN>" \
  -F "zip=@my-project.zip"

For more information, see Validate a declarative pipeline configuration.

Step 6: Import the package into the target project

Call the import endpoint with the project package:

curl -X POST "https://<TENANT_URL>/api/v1/di-projects/<TARGET_PROJECT_ID>/actions/import" \
  -H "Authorization: Bearer <BEARER_TOKEN>" \
  -F "zip=@my-project.zip"

A successful import returns 200 OK with an empty response object.

Verify the imported project

After importing the package, verify the target project in Qlik Cloud:

• The expected tasks are present.
• Binding values are set for the target environment.
• Task settings, schedules, and transformation rules match the imported project definition.
• The pipeline runs successfully in the target environment.

Project package structure

For project package structure details, see Project structure reference.

Troubleshooting

Issue: Import fails because binding values are missing

Confirm that the target project is compatible with the source project and has binding values set before import. Use qtcp_bindings_definition.json to identify the required variables, then set values by using the Bindings API or the Qlik Cloud user interface.

Issue: The imported project uses the wrong environment values

Confirm that each binding variable on the target project has the correct value for the target environment, if the package includes bindings.json.

Issue: The package is rejected

Confirm that the package contains supported declarative pipeline project files and that files are not empty. You can run the validation endpoint before import for a limited package check.

Issue: Import fails with a platform or project type mismatch

Confirm that the target project is compatible with the imported package. For example, do not import a package from a Snowflake project into a Databricks target project unless that cross-platform workflow is supported.