Definition of “Done”

Use these checklists to determine whether the task/user story which you are reviewing is fully complete. Fully complete means releasable, and no further work required.

We might consider things “done” at different levels because features are composed from many sub-units, such as Task issues and PRs.

GitHub Pull Request

A single PR often won’t complete an entire feature. There may be multiple smaller PRs necessary. We consider a PR done if the master branch would remain “releaseable” after merging this PR.

Ready for review

Fields filled out correctly according to Pull Request template.
PR is not in “draft” state.
See also Code Review Guidelines.

Can be merged into `master`

Reviewed by someone other than the author.
Addresses the referenced task ticket in some way.
Automated CI passing.
Meets Coding Standards.
All review comments are addressed.

Sometimes it may be expedient to merge a PR without all review comments and/or coding standards being addressed. This should only be done if the PR author undertakes to immediately begin correcting those issues in their next PR.

Jira Task issue (Technical task)

The implementation of a User Story issue is divided into one or more Task issues. The technical task issues will address a smaller part of the feature.

Once a technical task is done, the assignee checks that it meets the necessary criteria and moves it into the “QA” column of the sprint board.

Ready for moving to QA state

Acceptance Criteria have been fulfilled.
Has adequate test coverage at the appropriate level(s).
Has documentation
- User Manual updated (if applicable)
- Design documentation added/updated (if applicable)
Task as a whole meets Coding Standards.

Done

Ticket has been checked by our QA engineer for adherence to standards and definitions of done.

User Story issues

See Process for an example feature description.

There are automated functional test cases covering the A/C of the User Story.
The U/S works coherently as a whole in the application.
Non-functional requirements have also been addressed, such as:
- Performance
- Logging
- Other operational concerns