Use Tanuki in your GitLab CI/CD pipelines to:

• Transform a Unity Test Runner report to a GitLab JUnit report
• Transform a Unity Project Auditor report to a GitLab Code Quality report
• Generate a static HTML Code Quality report for GitLab Merge Requests
  • Demo
• Print code coverage of a Unity Cobertura coverage report to GitLab Code Coverage

The following is a typical GitLab CI/CD configuration using Tanuki:

Code Quality:
  image: ghcr.io/zensharp/tanuki
  rules:
    - if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH
    - if: $CI_PIPELINE_SOURCE == "merge_request_event"
  script:
    - |
      # Obtain codequality.json here
      tanuki html codequality.json -o index.html --title "$CI_PROJECT_TITLE"
  artifacts:
    paths:
      - index.html
    expose_as: "Code Quality Report"
    reports:
      codequality: codequality.json

Run Unit Tests:
  image: ghcr.io/zensharp/tanuki
  rules:
    - if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH
    - if: $CI_PIPELINE_SOURCE == "merge_request_event"
  script:
    - |
      # Obtain results.xml and cobertura.xml here
      tanuki transform results.xml -o junit.xml
      tanuki coverage cobertura.xml --multiplier 100
  coverage: '/Code coverage is: \d+\.\d+/'
  artifacts:
    reports:
      junit: junit.xml

After every merge request pipeline, the merge request will display:

• Code coverage (i.e. Test coverage)
• Code Quality HTML report (i.e. Code Quality Report)
• Code Quality changes
• Test summary

Click "Code Quality Report" to view the HTML report.

# Unity Project Auditor JSON to Code Climate
tanuki transform report.json --codeclimate

# Unity Test Runner XML to JUnit
tanuki transform report.xml --junit

# or let Tanuki automatically determine the format
tanuki transform report.json
tanuki transform report.xml

# By default, Tanuki prints the result to STDOUT
# Use -o|--output to write to a file
tanuki transform report.json -o codeclimate.json
tanuki transform report.xml -o junit.xml

tanuki edit report.json [-o edited.json] [--location-prefix "src/Assets"] [--linter "Project Auditor"]

tanuki merge a.json b.json ... z.json [-o merged.json]

tanuki html report.json [-o index.html] [--base-url "https://docs.example.com"] [--title "My Code Quality Report"]

tanuki coverage cobertura.xml [--multiplier 100.0]

# Output
Code coverage is: 0.897

Add a Tanukifile at .tanuki/Tanukifile. See this example

Then trigger the file with:

tanuki trigger

tanuki delete report.json --where ".path == ^Packages/"

tanuki assert report.json --none ".severity == Blocker"

Tanuki offers a simple syntax language for querying issues. For example:

Here are the available patterns (see also IssueFilterer.cs):

• .body
• .category
• .check_name
• .description
• .fingerprint
• .linter
• .location.path
• .severity

Tanuki will search for a configuration file in the following order:

1. ./.tanuki.yml
2. ./.tanuki/tanuki.yml
3. ./.config/tanuki.yml
4. ~/.config/tanuki.yml

Add custom linters to the configuration file. When using tanuki html, the linters will get properly linked in the generated HTML.

linters:
  - name: Project Auditor
    url: https://docs.unity3d.com/Packages/com.unity.project-auditor@1.0/manual/index.html
  - name: Rubocop
    url: https://docs.rubocop.org

• Tanuki's report template is based this template from gitlab-org/ci-cd/codequality.