All Products
Search
Document Center

DataWorks:Code review

Last Updated:Nov 14, 2025

Code review ensures production code meets quality standards before deployment. Enable mandatory review to require approval before publishing tasks to production.

Function introduction

Enable mandatory code review to prevent task failures from non-compliant code and avoid wasting computing resources. When enabled, reviewers validate code quality before it reaches production.

  • If mandatory code review is disabled: You can manually initiate reviews when submitting nodes. Nodes can be published regardless of review status.

  • If mandatory code review is enabled: When you submit a node, you must select a qualified reviewer to perform code review and ensure code quality.

    • If approved, you can publish the node.

    • If rejected, you must revise the code based on feedback and resubmit for review. You can only publish after approval.

image

Limitations

(Optional) Administrator: Enable mandatory code review

A workspace administrator can enable mandatory code review for a workspace on the Data Studio settings page.

Note
  • When disabled, reviews can still be initiated but won't block publishing.

  • Code review is not triggered when deleting nodes.

Go to the Security Settings and Others page

  1. Go to the Workspaces page in the DataWorks console. In the top navigation bar, select a desired region. Find the desired workspace and choose Shortcuts > Data Studio in the Actions column.

  2. Go to the Data Studio settings page.

    1. Click the image icon in the lower-left corner and select Data Studio Settings.

    2. On the DataStudio settings page, click the Security Settings and Others tab.

Enable code review

On the Security Settings and Others tab, enable Force to review code.

Important

Code review features are only available in DataWorks Professional Edition and above.

Parameter

Description

Force to review code

Requires code review before publishing tasks to production.

Code reviewers

Specifies the users who can review the code in the workspace.

  • Any Developer Role: Select any user with the Developer role as a reviewer. The publisher chooses the specific reviewer when submitting.

  • Specify development role users: Requires approval from specified developers before publishing.

Note
  • You can select only users with the Developer role in the current workspace as reviewers.

  • Configure one or more reviewers. When multiple reviewers are assigned, approval from any one is sufficient.

Baseline scopes for code review

Specify which nodes require review when submitted.

You can specify which nodes require mandatory code review based on the priority of the baseline to which they belong. The logic is as follows:

  • If you set the baseline review scope to non-baseline tasks, newly created tasks in the workspace require approval.

  • When set to a specific baseline level, nodes at that level require approval.

  • Higher values indicate higher priority. Baseline tasks have priority over non-baseline tasks.

Note

Selecting non-baseline tasks means that all new nodes created in the workspace require approval.

DingTalk group robot webhook

Enter a custom DingTalk chatbot webhook URL to receive review status notifications. For more information, see Configure a custom chatbot.

Note

For the security settings of the DingTalk group chatbot, you must configure keywords, and the keywords must include DataWorks.

@everyone for code review?

When enabled, (@) mentions all group members on status changes.

Review initiator: Initiate a code review

Submitting a node triggers code review. Select a reviewer to generate the review request. When mandatory review is enabled, publishing is blocked until approval.

Note
  • Initiating a new review automatically invalidates any in-progress review.

  • Deleting nodes does not trigger code review.

  1. Publish the task.

    1. Click the image icon to save the node, and click the image icon to open the publishing page.

    2. Click Start Deployment to Production Environment. With mandatory review enabled, publishing is blocked until approval.

  2. Submit for review.

    Click Submit for Code Review. Select a reviewer and describe your changes.

    Configuration Item

    Description

    Code Reviewer

    Select one or more reviewers from the configured list.

    Change Description

    Describe your changes.

    Click Confirm to submit.

Reviewer: Process a review request

Reviewers can view and process requests under Code Review List > Reviewed by Me.

  1. Go to the code review page.

    In the left navigation pane of Data Studio, click the image icon to access your review queue under Reviewed by Me.

  2. Review and take action.

    View request details and available actions.

    • In the code review list page: Quickly Approve, Reject, Discard, Reopen, or Share requests.

    • In the code review detail page:

      • Compare code and schedule changes against production to decide on approval.

      • Available actions:

        • image: Approve.

        • image: Reject.

        • image: Discard.

        • image: Reopen.

        • Comment.

Note
  • Discard: Cancels the review and allows immediate publishing.

  • Reopen: Restarts a discarded review.

After completion, initiators can view results and manage requests.

Review initiator: View results

View your review results under Code Review List > Created by Me.

  1. Go to the code review page.

    In the left navigation pane of Data Studio, click the image icon to access your review queue under Created by Me.

  2. Check review status.

    View details and take action as needed.

    • In the code review list page: View results and Discard, Reopen, or Share requests.

      Note

      After discarding, resubmit for review or reopen the request to continue publishing.

    • In the code review detail page: Click a request to open details and decide whether to reopen or discard.

  3. If rejected, revise your code based on feedback and resubmit. You can only publish after approval.

More operations

After approval, publish your task to production. If rejected, revise based on feedback and resubmit.