Skip to content
You're viewing an older version of this GitHub Action. Do you want to see the latest version instead?
file-text

GitHub Action

Test Reporting

v8

Test Reporting

file-text

Test Reporting

Shows test results in GitHub UI: .NET (xUnit, NUnit, MSTest), Dart, Flutter, Java (JUnit), JavaScript (JEST, Mocha)

Installation

Copy and paste the following snippet into your .yml file.

              

- name: Test Reporting

uses: phoenix-actions/test-reporting@v8

Learn more about this action in phoenix-actions/test-reporting

Choose a version

Test Reporting

A fork of Dorny's test reporter created primarily to add support for mochawesome-json. Also includes several further feature enhancements and bug fixes on the original.

This Github Action displays test results from popular testing frameworks directly in GitHub.

✔️ Parses test results in XML or JSON format and creates nice report as Github Check Run

✔️ Annotates code where it failed based on message and stack trace captured during test execution

✔️ Provides final conclusion and counts of passed, failed and skipped tests as output parameters

How it looks:

Supported languages / frameworks:

For more information see Supported formats section.

Do you miss support for your favorite language or framework? Please create Issue or contribute with PR.

Example

Following setup does not work in workflows triggered by pull request from forked repository. If that's fine for you, using this action is as simple as:

on:
  pull_request:
  push:
jobs:
  build-test:
    name: Build & Test
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2     # checkout the repo
      - run: npm ci                   # install packages
      - run: npm test                 # run tests (configured to use jest-junit reporter)

      - name: Test Report
        uses: phoenix-actions/test-reporting@v8
        id: test-report               # Set ID reference for step
        if: success() || failure()    # run this step even if previous step failed
        with:
          name: JEST Tests            # Name of the check run which will be created
          path: reports/jest-*.xml    # Path to test results
          reporter: jest-junit        # Format of test results

Action Outputs

A list of the possible action outputs are listed below.

  • runHtmlUrl

In case you want to extract and read an output from the action, see the below example code:

on:
  pull_request:
  push:
jobs:
  build-test:
    name: Build & Test
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2     # checkout the repo
      - run: npm ci                   # install packages
      - run: npm test                 # run tests (configured to use jest-junit reporter)

      - name: Test Report
        uses: phoenix-actions/test-reporting@v8
        id: test-report               # Set ID reference for step
        if: success() || failure()    # run this step even if previous step failed
        with:
          name: JEST Tests            # Name of the check run which will be created
          path: reports/jest-*.xml    # Path to test results
          reporter: jest-junit        # Format of test results

      - name: Read output variables
        run: |
          echo "url is ${{ steps.test-report.outputs.runHtmlUrl }}"

Recommended setup for public repositories

Workflows triggered by pull requests from forked repositories are executed with read-only token and therefore can't create check runs. To workaround this security restriction, it's required to use two separate workflows:

  1. CI runs in the context of the PR head branch with the read-only token. It executes the tests and uploads test results as a build artifact
  2. Test Report runs in the context of the repository main branch with read/write token. It will download test results and create reports

PR head branch: .github/workflows/ci.yml

name: 'CI'
on:
  pull_request:
jobs:
  build-test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2         # checkout the repo
      - run: npm ci                       # install packages
      - run: npm test                     # run tests (configured to use jest-junit reporter)
      - uses: actions/upload-artifact@v2  # upload test results
        if: success() || failure()        # run this step even if previous step failed
        with:
          name: test-results
          path: jest-junit.xml

default branch: .github/workflows/test-report.yml

name: 'Test Report'
on:
  workflow_run:
    workflows: ['CI']                     # runs after CI workflow
    types:
      - completed
jobs:
  report:
    runs-on: ubuntu-latest
    steps:
    - uses: phoenix-actions/test-reporting@v8
      id: test-report                     # Set ID reference for step
      with:
        artifact: test-results            # artifact name
        name: JEST Tests                  # Name of the check run which will be created
        path: '*.xml'                     # Path to test results (inside artifact .zip)
        reporter: jest-junit              # Format of test results

Usage

- uses: phoenix-actions/test-reporting@v8
  with:

    # Name or regex of artifact containing test results
    # Regular expression must be enclosed in '/'.
    # Values from captured groups will replace occurrences of $N in report name.
    # Example:
    #   artifact: /test-results-(.*)/
    #   name: 'Test report $1'
    #   -> Artifact 'test-result-ubuntu' would create report 'Test report ubuntu'
    artifact: ''

    # Name of the Check Run which will be created
    name: ''

    # Coma separated list of paths to test results
    # Supports wildcards via [fast-glob](https://github.com/mrmlnc/fast-glob)
    # All matched result files must be of the same format
    path: ''

    # The fast-glob library that is internally used interprets backslashes as escape characters.
    # If enabled, all backslashes in provided path will be replaced by forward slashes and act as directory separators.
    # It might be useful when path input variable is composed dynamically from existing directory paths on Windows.
    path-replace-backslashes: 'false'

    # Format of test results. Supported options:
    #   dart-json
    #   dotnet-trx
    #   flutter-json
    #   java-junit
    #   jest-junit
    #   mocha-json
    reporter: ''

    # Allows you to generate only the summary.
    # If enabled, the report will contain a table listing each test results file and the number of passed, failed, and skipped tests.
    # Detailed listing of test suites and test cases will be skipped.
    only-summary: 'false'

    # Limits which test suites are listed:
    #   all
    #   failed
    list-suites: 'all'

    # Limits which test cases are listed:
    #   all
    #   failed
    #   none
    list-tests: 'all'

    # Limits number of created annotations with error message and stack trace captured during test execution.
    # Must be less or equal to 50.
    max-annotations: '10'

    # Set action as failed if test report contains any failed test
    fail-on-error: 'true'

    # Relative path under $GITHUB_WORKSPACE where the repository was checked out.
    working-directory: ''

    # Personal access token used to interact with Github API
    # Default: ${{ github.token }}
    token: ''

Output parameters

Name Description
conclusion success or failure
passed Count of passed tests
failed Count of failed tests
skipped Count of skipped tests
time Test execution time [ms]

Supported formats

dart-json

Test run must be configured to use JSON reporter. You can configure it in dart_test.yaml:

file_reporters:
  json: reports/test-results.json

Or with CLI arguments:

dart test --file-reporter="json:test-results.json"

For more information see:

dotnet-trx

Test execution must be configured to produce Visual Studio Test Results files (TRX). To get test results in TRX format you can execute your tests with CLI arguments:

dotnet test --logger "trx;LogFileName=test-results.trx"

Or you can configure TRX test output in *.csproj or Directory.Build.props:

<PropertyGroup>
  <VSTestLogger>trx%3bLogFileName=$(MSBuildProjectName).trx</VSTestLogger>
  <VSTestResultsDirectory>$(MSBuildThisFileDirectory)/TestResults/$(TargetFramework)</VSTestResultsDirectory>
</PropertyGroup>

Supported testing frameworks:

For more information see dotnet test

flutter-json

Test run must be configured to use JSON reporter. You can configure it in dart_test.yaml:

file_reporters:
  json: reports/test-results.json

Or with (undocumented) CLI argument:

flutter test --machine > test-results.json

According to documentation dart_test.yaml should be at the root of the package, next to the package's pubspec. On current stable and beta channels it doesn't work, and you have to put dart_test.yaml inside your test folder. On dev channel, it's already fixed.

For more information see:

java-junit (Experimental)

Support for JUnit XML is experimental - should work but it was not extensively tested. To have code annotations working properly, it's required your directory structure matches the package name. This is due to the fact Java stack traces don't contain a full path to the source file. Some heuristic was necessary to figure out the mapping between the line in the stack trace and an actual source file.

jest-junit

JEST testing framework support requires the usage of jest-junit reporter. It will create test results in junit XML format which can be then processed by this action. You can use the following example configuration in package.json:

"scripts": {
  "test": "jest --ci --reporters=default --reporters=jest-junit"
},
"devDependencies": {
  "jest": "^26.5.3",
  "jest-junit": "^12.0.0"
},
"jest-junit": {
  "outputDirectory": "reports",
  "outputName": "jest-junit.xml",
  "ancestorSeparator": "",
  "uniqueOutputName": "false",
  "suiteNameTemplate": "{filepath}",
  "classNameTemplate": "{classname}",
  "titleTemplate": "{title}"
}

Configuration of uniqueOutputName, suiteNameTemplate, classNameTemplate, titleTemplate is important for proper visualization of test results.

mocha-json

Mocha testing framework support requires:

  • Mocha version v9.1.0 or higher
  • Usage of json reporter.

You can use the following example configuration in package.json:

"scripts": {
  "test": "mocha --reporter json --reporter-option output=filename.json"
}

For Mocha < v9.1, the command should look like this:

"scripts": {
  "test": "mocha --reporter json > test-results.json"
}

Additionally, test processing might fail if any of your tests write anything on standard output. Before version v9.1.0, Mocha doesn't have the option to store json output directly to the file, and we have to rely on redirecting its standard output (mocha#4607). Please update Mocha to version v9.1.0 or above if you encounter this issue.

mochawesome-json

Mochawesome testing framework support requires:

  • Mochawesome version v7.0.0 or higher

You can use the following example configuration in package.json:

"scripts": {
  "test": "mocha test.js --reporter mochawesome --reporter-options reportDir=customReportDir,reportFilename=customReportFilename"
}

GitHub limitations

Unfortunately, there are some known issues and limitations caused by GitHub API:

  • Test report (i.e. Check Run summary) is markdown text. No custom styling or HTML is possible.
  • Maximum report size is 65535 bytes. Input parameters list-suites and list-tests will be automatically adjusted if max size is exceeded.
  • Test report can't reference any additional files (e.g. screenshots). You can use actions/upload-artifact@v2 to upload them and inspect them manually.
  • Check Runs are created for specific commit SHA. It's not possible to specify under which workflow test report should belong if more workflows are running for the same SHA. Thanks to this GitHub "feature" it's possible your test report will appear in an unexpected place in GitHub UI. For more information, see #67.

See also

  • paths-filter - Conditionally run actions based on files modified by PR, feature branch, or pushed commits

License

The scripts and documentation in this project are released under the MIT License