GitHub Action
Test Reporter
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.
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: dorny/test-reporter@v1
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
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:
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 artifactTest 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: dorny/test-reporter@v1
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
- uses: dorny/test-reporter@v1
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: ''
# Format of test results. Supported options:
# dart-json
# dotnet-trx
# flutter-json
# java-junit
# jest-junit
# mocha-json
reporter: ''
# 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: ${{ github.token }}
token: ''
Name | Description |
---|---|
conclusion | success or failure |
passed | Count of passed tests |
failed | Count of failed tests |
skipped | Count of skipped tests |
time | Test execution time [ms] |
dart-json
Test run must be configured to use JSON reporter.
You can configure it in dart_test.yaml
:
file_reporters:
json: reports/test-results.json
Or with CLI arguments:
dart test --file-reporter="json:test-results.json"
For more information see:
dotnet-trx
Test execution must be configured to produce Visual Studio Test Results files (TRX). To get test results in TRX format you can execute your tests with CLI arguments:
dotnet test --logger "trx;LogFileName=test-results.trx"
Or you can configure TRX test output in *.csproj
or Directory.Build.props
:
<PropertyGroup>
<VSTestLogger>trx%3bLogFileName=$(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:
You can use the following example configuration in package.json
:
"scripts": {
"test": "mocha --reporter json > test-results.json"
}
Test processing might fail if any of your tests write anything on standard output.
Mocha, unfortunately, doesn't have the option to store json
output directly to the file, and we have to rely on redirecting its standard output.
There is a work in progress to fix it: mocha#4607
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
andlist-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.
- paths-filter - Conditionally run actions based on files modified by PR, feature branch, or pushed commits
The scripts and documentation in this project are released under the MIT License