gradle/actions
1
0
Fork 0
mirror of https://github.com/gradle/actions synced 2024-11-27 11:52:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Restructure documentation and add a dependency-submission FAQ (#110)

Browse source
This commit is contained in:
Daz DeBoer 2024-04-04 16:38:14 -06:00 committed by GitHub
parent e24011a3b5
commit 0b06ce12c8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 1203 additions and 1089 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
README.md
View file

@ -8,7 +8,7 @@ The `setup-gradle` action can be used to configure Gradle for optimal execution
This replaces the previous `gradle/gradle-build-action`, which now delegates to this implementation.
The recommended way to execute any Gradle build is with the help of the [Gradle Wrapper](https://docs.gradle.org/current/userguide/gradle_wrapper.html), and the examples assume that the Gradle Wrapper has been configured for the project. See [this example](setup-gradle/README.md#build-with-a-specific-gradle-version) if your project doesn't use the Gradle Wrapper.
The recommended way to execute any Gradle build is with the help of the [Gradle Wrapper](https://docs.gradle.org/current/userguide/gradle_wrapper.html), and the examples assume that the Gradle Wrapper has been configured for the project. See [this example](docs/setup-gradle.md#build-with-a-specific-gradle-version) if your project doesn't use the Gradle Wrapper.
### Example usage
@ -29,7 +29,7 @@ jobs:
run: ./gradlew build
```
See the [full action documentation](setup-gradle/README.md) for more advanced usage scenarios.
See the [full action documentation](docs/setup-gradle.md) for more advanced usage scenarios.
## The `dependency-submission` action
@ -58,4 +58,4 @@ jobs:
uses: gradle/actions/dependency-submission@v3
```
See the [full action documentation](dependency-submission/README.md) for more advanced usage scenarios.
See the [full action documentation](docs/dependency-submission.md) for more advanced usage scenarios.

388
dependency-submission/README.md
View file

@ -1,19 +1,6 @@
# The `dependency-submission` action
## The `dependency-submission` action
The `gradle/actions/dependency-submission` action provides the simplest (and recommended) way to generate a
dependency graph for your project. This action will attempt to detect all dependencies used by your build
without building and testing the project itself.
The dependency graph snapshot is generated via integration with the [GitHub Dependency Graph Gradle Plugin](https://plugins.gradle.org/plugin/org.gradle.github-dependency-graph-gradle-plugin), and submitted to your repository via the
[GitHub Dependency Submission API](https://docs.github.com/en/rest/dependency-graph/dependency-submission).
The generated snapshot files can be submitted in the same job, or saved for submission in a subsequent job.
The generated dependency graph includes all of the dependencies in your build, and is used by GitHub to generate
[Dependabot Alerts](https://docs.github.com/en/code-security/dependabot/dependabot-alerts/about-dependabot-alerts)
for vulnerable dependencies, as well as to populate the
[Dependency Graph insights view](https://docs.github.com/en/code-security/supply-chain-security/understanding-your-software-supply-chain/exploring-the-dependencies-of-a-repository#viewing-the-dependency-graph).
## General usage
Generates and submits a dependency graph for a Gradle project, allowing GitHub to alert about reported vulnerabilities in your project dependencies.
The following workflow will generate a dependency graph for a Gradle project and submit it immediately to the repository via the
Dependency Submission API. For most projects, this default configuration should be all that you need.
@ -38,373 +25,4 @@ jobs:
uses: gradle/actions/dependency-submission@v3
```
### Configuration parameters
In some cases, the default action configuration will not be sufficient, and additional action parameters will need to be specified.
See the example below for a summary, and the [Action Metadata file](action.yml) for a more detailed description of each input parameter.
```yaml
name: Dependency Submission with advanced config
on: [ push ]
permissions:
contents: read
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and save dependency graph
uses: gradle/actions/dependency-submission@v3
with:
# Use a particular Gradle version instead of the configured wrapper.
gradle-version: 8.6
# The gradle project is not in the root of the repository.
build-root-directory: my-gradle-project
# Enable configuration-cache reuse for this build.
cache-encryption-key: ${{ secrets.GRADLE_ENCRYPTION_KEY }}
# Do not attempt to submit the dependency-graph. Save it as a workflow artifact.
dependency-graph: generate-and-upload
```
# Resolving a dependency vulnerability
## Finding the source of a dependency vulnerability
Once you have submitted a dependency graph, you may receive Dependabot Alerts warning about vulnerabilities in
dependencies of your project. In the case of transitive dependencies, it may not be obvious how that dependency is
used or what you can do to address the vulnerability alert.
The first step to investigating a Dependabot Alert is to determine the source of the dependency. One of the best ways to
do so is with a free Develocity Build Scan®, which makes it easy to explore the dependencies resolved in your build.
<img width="1069" alt="image" src="https://github.com/gradle/actions/assets/179734/3a637dfd-396c-4e94-8332-dcc6eb5a35ac">
In this example, we are searching for dependencies matching the name 'com.squareup.okio:okio' in the _Build Dependencies_ of
the project. You can easily see that this dependency originates from 'com.github.ben-manes:gradle-versions-plugin'.
Knowing the source of the dependency can help determine how to deal with the Dependabot Alert.
Note that you may need to look at both the _Dependencies_ and the _Build Dependencies_ of your project to find the
offending dependency.
### Publishing a Develocity Build Scan® from your dependency submission workflow
You can automatically publish a Build Scan on every run of `gradle/actions/dependency-submission`. Three input parameters are
required, one to enable publishing and two more to accept the [Develocity terms of use](https://gradle.com/help/legal-terms-of-use).
```yaml
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
with:
build-scan-publish: true
build-scan-terms-of-use-url: "https://gradle.com/terms-of-service"
build-scan-terms-of-use-agree: "yes"
```
### When you cannot publish a Build Scan®
If publishing a free Build Scan to https://scans.gradle.com isn't an option, and you don't have access to a private [Develocity
server](https://gradle.com/) for your project, you can obtain information about the each resolved dependency by running the `dependency-submission` workflow with debug logging enabled.
The simplest way to do so is to re-run the dependency-submission job with debug logging enabled:
<img width="665" alt="image" src="https://github.com/gradle/actions/assets/179734/d95b889a-09fb-4731-91f2-baebbf647e31">
When you do so, the Gradle build that generates the dependency-graph will include a log message for each dependency version included in the graph.
Given the details in one log message, you can run (locally) the built-in [dependencyInsight](https://docs.gradle.org/current/userguide/viewing_debugging_dependencies.html#dependency_insights) task
to determine exactly how the dependency was resolved.
For example, given the following message in the logs:
```
Detected dependency 'com.google.guava:guava:32.1.3-jre': project = ':my-subproject', configuration = 'compileClasspath'
```
You would run the following command locally:
```
./gradlew :my-subproject:dependencyInsight --configuration compileClasspath --dependency com.google.guava:guava:32.1.3-jre
```
#### Dealing with 'classpath' configuration
If the configuration value in the log message is "classpath" then instead of running `dependency-insight` you'll need to run the Gradle
`buildEnvironment` task.
For example, given the following message in the logs:
```
Detected dependency 'xerces:xercesImpl:2.12.2': project = ':my-subproject', configuration = 'classpath'
```
You would run the following command locally to expose the `xercesImpl` dependency:
```
./gradlew :my-subproject:buildEnvironment | grep -C 5 xercesImpl
```
## Updating the dependency version
Once you've discovered the source of the dependency, the most obvious fix is to update the dependency to a patched version that does not
suffer the vulnerability. For direct dependencies, this is often straightforward. But for transitive dependencies it can be tricky.
### Dependency source is specified directly in the build
If the dependency is used to compile your code or run your tests, it's normal for the underlying "source" of the dependency to have a
version configured directly in the build. For example, if you have a vulnerable version of `com.squareup.okio:okio` in your `compileClasspath`, then
it's likely you have a dependency like `com.squareup.moshi:moshi` configured as an `api` or `implementation` dependency.
In this case there are 2 possibilities:
1. There is a newer, compatible version of `com.squareup.moshi:moshi` available, and you can just bump the version number.
2. There isn't a newer, compatible version of `com.squareup.moshi:moshi`
In the second case, you can add a Dependency Constraint, to force the use of the newest version of `com.squareup.okio`:
```kotlin
dependencies {
implementation("com.squareup.moshi:moshi:1.12.0")
constraints {
// Force a newer version of okio in transitive resolution
implementation("com.squareup.okio:okio:3.6.0")
}
}
```
### Dependency source is a plugin classpath
If the vulnerable dependency is introduced by a Gradle plugin, again the best option is to look for a newer version of the plugin.
But if none is available, you can still use a dependency constraint to force a newer transitive version to be used.
The dependency constraint must be added to the `classpath` configuration of the buildscript that loads the plugin.
```kotlin
buildscript {
repositories {
gradlePluginPortal()
}
dependencies {
constraints {
// Force a newer version of okio in transitive resolution
classpath("com.squareup.okio:okio:3.6.0")
}
}
}
plugins {
id("com.github.ben-manes.versions") version("0.51.0")
}
```
## Limiting the dependencies that appear in the dependency graph
By default, the `dependency-submission` action attempts to detect all dependencies declared and used by your Gradle build.
At times it may helpful to limit the dependencies reported to GitHub, to avoid security alerts for dependencies that
don't form a critical part of your product. For example, a vulnerability in the tool you use to generate documentation
may not be as important as a vulnerability in one of your runtime dependencies.
The `dependency-submission` action provides a convenient mechanism to filter the projects and configurations that
contribute to the dependency graph.
> [!NOTE]
> Ideally, all dependencies involved in building and testing a project will be extracted and reported in a dependency graph.
> These dependencies would be assigned to different scopes (eg development, runtime, testing) and the GitHub UI would make it easy to opt-in to security alerts for different dependency scopes.
> However, this functionality does not yet exist.
### Excluding certain Gradle projects from the dependency graph
If you do not want the dependency graph to include dependencies from every project in your build,
you can easily exclude certain projects from the dependency extraction process.
To restrict which Gradle subprojects contribute to the report, specify which projects to exclude via a regular expression.
You can provide this value via the `DEPENDENCY_GRAPH_EXCLUDE_PROJECTS` environment variable or system property.
Note that excluding a project in this way only removes dependencies that are _resolved_ as part of that project, and may
not necessarily remove all dependencies _declared_ in that project. If another project depends on the excluded project
then it may transitively resolve dependencies declared in the excluded project: these dependencies will still be included
in the generated dependency graph.
### Excluding certain Gradle configurations from the dependency graph
Similarly to Gradle projects, it is possible to exclude a set of configuration instances from dependency graph generation,
so that dependencies resolved by those configurations are not included.
To restrict which Gradle configurations contribute to the report, specify which configurations to exclude via a regular expression.
You can provide this value via the `DEPENDENCY_GRAPH_EXCLUDE_CONFIGURATIONS` environment variable or system property.
Note that configuration exclusion applies to the configuration in which the dependency is _resolved_ which is not necessarily
the configuration where the dependency is _declared_. For example if you decare a dependency as `implementation` in
a Java project, that dependency will be resolved in `compileClasspath`, `runtimeClasspath` and possibly other configurations.
### Example of project and configuration filtering
For example, if you want to exclude dependencies in the `buildSrc` project, and exclude dependencies from the `testCompileClasspath` and `testRuntimeClasspath` configurations, you would use the following configuration:
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
env:
# Exclude all dependencies that originate solely in the 'buildSrc' project
DEPENDENCY_GRAPH_EXCLUDE_PROJECTS: ':buildSrc'
# Exclude dependencies that are only resolved in test classpaths
DEPENDENCY_GRAPH_EXCLUDE_CONFIGURATIONS: '.*[Tt]est(Compile|Runtime)Classpath'
```
### Other filtering options
The [GitHub Dependency Graph Gradle Plugin](https://plugins.gradle.org/plugin/org.gradle.github-dependency-graph-gradle-plugin)
has other filtering options that may be useful.
See [the docs](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#filtering-which-gradle-configurations-contribute-to-the-dependency-graph) for details.
# Advance usage scenarios
## Using a custom plugin repository
By default, the action downloads the `github-dependency-graph-gradle-plugin` from the Gradle Plugin Portal (https://plugins.gradle.org). If your GitHub Actions environment does not have access to this URL, you can specify a custom plugin repository to use.
Do so by setting the `GRADLE_PLUGIN_REPOSITORY_URL` environment variable.
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
env:
GRADLE_PLUGIN_REPOSITORY_URL: "https://gradle-plugins-proxy.mycorp.com"
```
## Integrating the `dependency-review-action`
The GitHub [dependency-review-action](https://github.com/actions/dependency-review-action) helps you
understand dependency changes (and the security impact of these changes) for a pull request,
by comparing the dependency graph for the pull-request with that of the HEAD commit.
Example of a pull request workflow that executes a build for a pull request and runs the `dependency-review-action`:
```yaml
name: Dependency review for pull requests
on: [ pull_request ]
permissions:
contents: write
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
dependency-review:
needs: dependency-submission
runs-on: ubuntu-latest
steps:
- name: Perform dependency review
uses: actions/dependency-review-action@v3
```
Note that the `dependency-submission` action submits the dependency graph at the completion of the workflow Job.
For this reason, the `dependency-review-action` must be executed in a dependent job, and not as a subsequent step in the job that generates the dependency graph.
## Usage with pull requests from public forked repositories
This `contents: write` permission is [not available for any workflow that is triggered by a pull request submitted from a public forked repository](https://docs.github.com/en/actions/security-guides/automatic-token-authentication#permissions-for-the-github_token).
This limitation is designed to prevent a malicious pull request from effecting repository changes.
Because of this restriction, we require 2 separate workflows in order to generate and submit a dependency graph:
1. The first workflow runs directly against the pull request sources and will `generate-and-upload` the dependency graph.
2. The second workflow is triggered on `workflow_run` of the first workflow, and will `download-and-submit` the previously saved dependency graph.
***Main workflow file***
```yaml
name: Generate and save dependency graph
on: [ pull_request ]
permissions:
contents: read # 'write' permission is not available
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and save dependency graph
uses: gradle/actions/dependency-submission@v3
with:
dependency-graph: generate-and-upload
```
***Dependent workflow file***
```yaml
name: Download and submit dependency graph
on:
workflow_run:
workflows: ['Generate and save dependency graph']
types: [completed]
permissions:
contents: write
jobs:
submit-dependency-graph:
runs-on: ubuntu-latest
steps:
- name: Download and submit dependency graph
uses: gradle/actions/dependency-submission@v3
with:
dependency-graph: download-and-submit # Download saved dependency-graph and submit
```
### Integrating `dependency-review-action` for pull requests from public forked repositories
To integrate the `dependency-review-action` into the pull request workflows above, a third workflow file is required.
This workflow will be triggered directly on `pull_request`, but will wait until the dependency graph results are
submitted before the dependency review can complete. The period to wait is controlled by the `retry-on-snapshot-warnings` input parameters.
Here's an example of a separate "Dependency Review" workflow that will wait for 10 minutes for the above PR check workflow to complete.
```yaml
name: dependency-review
on: [ pull_request ]
permissions:
contents: read
jobs:
dependency-review:
runs-on: ubuntu-latest
steps:
- name: 'Dependency Review'
uses: actions/dependency-review-action@v3
with:
retry-on-snapshot-warnings: true
retry-on-snapshot-warnings-timeout: 600
```
The `retry-on-snapshot-warnings-timeout` (in seconds) needs to be long enough to allow the entire `Generate and save dependency graph` and `Download and submit dependency graph` workflows (above) to complete.
# Gradle version compatibility
Dependency-graph generation is compatible with most versions of Gradle >= `5.2`, and is tested regularly against
Gradle versions `5.2.1`, `5.6.4`, `6.0.1`, `6.9.4`, `7.1.1` and `7.6.3`, as well as all patched versions of Gradle 8.x.
A known exception to this is that Gradle `7.0`, `7.0.1` and `7.0.2` are not supported.
See [here](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#gradle-compatibility) for complete compatibility information.
See the [full action documentation](../docs/dependency-submission.md) for more advanced usage scenarios.

58
docs/dependency-submission-faq.md Normal file
View file

@ -0,0 +1,58 @@
# Dependency submission FAQ
Implementing a `dependency-submission` workflow for your repository is documented in the
[core documentation](dependency-submission.md).
But getting it working is the easy part: the dependency alerts you recieve can be confusing and surprising.
Here are some common questions answered.
### How can I easily try this out without experimenting on my main repository?
The https://github.com/gradle/github-dependency-submission-demo repository is setup as a tutorial for you to fork and play with.
### How can I tell if the `dependency-submission` action is working?
Inspect the Dependency Graph for your project (Insights -> Dependency Graph). You should see some dependencies annotated with "Detected by GitHub Dependency Graph Gradle Plugin"
### Why is `(Maven)` stated for all dependencies submitted by this action? I'm not using Maven.
This simply indicates that the dependency was resolved from a standard Gradle/Maven artifact repository. It does not imply which build tool is used.
### Why is every dependency attributed to `settings.gradle.kts`?
All dependendies detected by the `dependency-submission` action are attributed to the Gradle project as a whole. We found that the best way is to link to the project `Settings` file.
We do not currently attempt to attribute dependencies to the actual file where they were declared.
### Why aren't dependencies be linked to the source file where they are declared?
There are a couple of reasons for this:
1. Gradle doesn't currently provide a mechanism to determine the location where a dependency is declared. In fact, the resulting dependency version can be influenced by many different sources within a Gradle project.
2. The GitHub Dependency Graph was modelled heavily on NPM and doesn't really map well to having multiple source locations for a single dependency declaration.
We have long-term plans to improve the first point, and we are working with GitHub to resolve the second. However, at this stage the behaviour your are experiencing is what is expected.
### My repository dependency graph contains a dependency that isn't anywhere in my build. Why is the `dependency-submission` action reporting dependencies I'm not using?
If you see a particular dependency version reported in the dependency graph, it means your build is resolving that dependency at some point.
You may be surprised what transitive dependencies are brought in by declared dependencies and applied plugins in your build.
[See here for a HOW-TO](dependency-submission.md#resolving-a-dependency-vulnerability) on getting the bottom of why the dependency is being resolved.
### I see multiple versions of the same dependency in the dependency graph, but I'm only declaring a single version in my build. Why is the action reporting dependency versions I'm not using?
This is almost certainly because the dependency in question is actually being resolved with different versions in different dependency configurations.
For example, you may have one version brought in as a plugin dependency (resolved in the `classpath` configuration) and another used directly as a code dependency (resolved in the `compileClasspath` configuration).
[See here for a HOW-TO](dependency-submission.md#resolving-a-dependency-vulnerability) on getting the bottom of why the dependency is being resolved.
By far the easiest way is to publish a Build Scan® for the workflow run: [this is easily achieved with some additional action configuration](dependency-submission.md#publishing-a-develocity-build-scan-from-your-dependency-submission-workflow).
### I'm not seeing any security vulnerabilities for any of my dependencies. How can I be sure this is working?
First check that [Dependabot Alerts](https://docs.github.com/en/code-security/dependabot/dependabot-alerts/about-dependabot-alerts) are enabled for your repository.
Without this, your dependency graph may be populated but you won't see which dependencies are potentially vulnerable.
### How can I use Dependabot Security Updates to generate a PR to update my vulnerable dependencies?
In most cases, the Dependabot Security Updates feature is not able to automatically generate a PR to update a dependency version.
This can be due to the vulnerable dependency being transitive, or because the Dependabot implementation doesn't understand how to update the dependency version.
In a few select cases the Dependabot security update will work and successfully generate a pull-request. For example when a direct dependency version is listed in a TOML dependency catalog.
### I'm getting many false positive Dependabot Alerts for dependencies that aren't used by my project. Why are these dependencies being reported?
The `dependency-submission` action resolves all of the dependencies in your build. This includes plugins, dependencies you've declared, test dependencies, and all transitive dependencies of these.
It doesn't matter how the dependencies are declared: the ones being resolved by Gradle are the ones being reported.
Many people are surprised to see what dependencies are actually being resolved when they run their builds, but I'm yet to see a case where the dependencies being reported are actually incorrect.
Please [follow the instructions here](dependency-submission.md#finding-the-source-of-a-dependency-vulnerability) to identify the source of the dependency version that is being reported.
Once you have worked out why it is being resolved, you can either [update the dependency version](dependency-submission.md#updating-the-dependency-version)
or [exclude it from the submitted dependency graph](dependency-submission.md#limiting-the-dependencies-that-appear-in-the-dependency-graph).

410
docs/dependency-submission.md Normal file
View file

@ -0,0 +1,410 @@
# The `dependency-submission` action
The `gradle/actions/dependency-submission` action provides the simplest (and recommended) way to generate a
dependency graph for your project. This action will attempt to detect all dependencies used by your build
without building and testing the project itself.
The dependency graph snapshot is generated via integration with the [GitHub Dependency Graph Gradle Plugin](https://plugins.gradle.org/plugin/org.gradle.github-dependency-graph-gradle-plugin), and submitted to your repository via the
[GitHub Dependency Submission API](https://docs.github.com/en/rest/dependency-graph/dependency-submission).
The generated snapshot files can be submitted in the same job, or saved for submission in a subsequent job.
The generated dependency graph includes all of the dependencies in your build, and is used by GitHub to generate
[Dependabot Alerts](https://docs.github.com/en/code-security/dependabot/dependabot-alerts/about-dependabot-alerts)
for vulnerable dependencies, as well as to populate the
[Dependency Graph insights view](https://docs.github.com/en/code-security/supply-chain-security/understanding-your-software-supply-chain/exploring-the-dependencies-of-a-repository#viewing-the-dependency-graph).
## General usage
The following workflow will generate a dependency graph for a Gradle project and submit it immediately to the repository via the
Dependency Submission API. For most projects, this default configuration should be all that you need.
Simply add this as a new workflow file to your repository (eg `.github/workflows/dependency-submission.yml`).
```yaml
name: Dependency Submission
on: [ push ]
permissions:
contents: write
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
```
### Configuration parameters
In some cases, the default action configuration will not be sufficient, and additional action parameters will need to be specified.
See the example below for a summary, and the [Action Metadata file](action.yml) for a more detailed description of each input parameter.
```yaml
name: Dependency Submission with advanced config
on: [ push ]
permissions:
contents: read
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and save dependency graph
uses: gradle/actions/dependency-submission@v3
with:
# Use a particular Gradle version instead of the configured wrapper.
gradle-version: 8.6
# The gradle project is not in the root of the repository.
build-root-directory: my-gradle-project
# Enable configuration-cache reuse for this build.
cache-encryption-key: ${{ secrets.GRADLE_ENCRYPTION_KEY }}
# Do not attempt to submit the dependency-graph. Save it as a workflow artifact.
dependency-graph: generate-and-upload
```
# Resolving a dependency vulnerability
## Finding the source of a dependency vulnerability
Once you have submitted a dependency graph, you may receive Dependabot Alerts warning about vulnerabilities in
dependencies of your project. In the case of transitive dependencies, it may not be obvious how that dependency is
used or what you can do to address the vulnerability alert.
The first step to investigating a Dependabot Alert is to determine the source of the dependency. One of the best ways to
do so is with a free Develocity Build Scan®, which makes it easy to explore the dependencies resolved in your build.
<img width="1069" alt="image" src="https://github.com/gradle/actions/assets/179734/3a637dfd-396c-4e94-8332-dcc6eb5a35ac">
In this example, we are searching for dependencies matching the name 'com.squareup.okio:okio' in the _Build Dependencies_ of
the project. You can easily see that this dependency originates from 'com.github.ben-manes:gradle-versions-plugin'.
Knowing the source of the dependency can help determine how to deal with the Dependabot Alert.
Note that you may need to look at both the _Dependencies_ and the _Build Dependencies_ of your project to find the
offending dependency.
### Publishing a Develocity Build Scan® from your dependency submission workflow
You can automatically publish a Build Scan on every run of `gradle/actions/dependency-submission`. Three input parameters are
required, one to enable publishing and two more to accept the [Develocity terms of use](https://gradle.com/help/legal-terms-of-use).
```yaml
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
with:
build-scan-publish: true
build-scan-terms-of-use-url: "https://gradle.com/terms-of-service"
build-scan-terms-of-use-agree: "yes"
```
### When you cannot publish a Build Scan®
If publishing a free Build Scan to https://scans.gradle.com isn't an option, and you don't have access to a private [Develocity
server](https://gradle.com/) for your project, you can obtain information about the each resolved dependency by running the `dependency-submission` workflow with debug logging enabled.
The simplest way to do so is to re-run the dependency-submission job with debug logging enabled:
<img width="665" alt="image" src="https://github.com/gradle/actions/assets/179734/d95b889a-09fb-4731-91f2-baebbf647e31">
When you do so, the Gradle build that generates the dependency-graph will include a log message for each dependency version included in the graph.
Given the details in one log message, you can run (locally) the built-in [dependencyInsight](https://docs.gradle.org/current/userguide/viewing_debugging_dependencies.html#dependency_insights) task
to determine exactly how the dependency was resolved.
For example, given the following message in the logs:
```
Detected dependency 'com.google.guava:guava:32.1.3-jre': project = ':my-subproject', configuration = 'compileClasspath'
```
You would run the following command locally:
```
./gradlew :my-subproject:dependencyInsight --configuration compileClasspath --dependency com.google.guava:guava:32.1.3-jre
```
#### Dealing with 'classpath' configuration
If the configuration value in the log message is "classpath" then instead of running `dependency-insight` you'll need to run the Gradle
`buildEnvironment` task.
For example, given the following message in the logs:
```
Detected dependency 'xerces:xercesImpl:2.12.2': project = ':my-subproject', configuration = 'classpath'
```
You would run the following command locally to expose the `xercesImpl` dependency:
```
./gradlew :my-subproject:buildEnvironment | grep -C 5 xercesImpl
```
## Updating the dependency version
Once you've discovered the source of the dependency, the most obvious fix is to update the dependency to a patched version that does not
suffer the vulnerability. For direct dependencies, this is often straightforward. But for transitive dependencies it can be tricky.
### Dependency source is specified directly in the build
If the dependency is used to compile your code or run your tests, it's normal for the underlying "source" of the dependency to have a
version configured directly in the build. For example, if you have a vulnerable version of `com.squareup.okio:okio` in your `compileClasspath`, then
it's likely you have a dependency like `com.squareup.moshi:moshi` configured as an `api` or `implementation` dependency.
In this case there are 2 possibilities:
1. There is a newer, compatible version of `com.squareup.moshi:moshi` available, and you can just bump the version number.
2. There isn't a newer, compatible version of `com.squareup.moshi:moshi`
In the second case, you can add a Dependency Constraint, to force the use of the newest version of `com.squareup.okio`:
```kotlin
dependencies {
implementation("com.squareup.moshi:moshi:1.12.0")
constraints {
// Force a newer version of okio in transitive resolution
implementation("com.squareup.okio:okio:3.6.0")
}
}
```
### Dependency source is a plugin classpath
If the vulnerable dependency is introduced by a Gradle plugin, again the best option is to look for a newer version of the plugin.
But if none is available, you can still use a dependency constraint to force a newer transitive version to be used.
The dependency constraint must be added to the `classpath` configuration of the buildscript that loads the plugin.
```kotlin
buildscript {
repositories {
gradlePluginPortal()
}
dependencies {
constraints {
// Force a newer version of okio in transitive resolution
classpath("com.squareup.okio:okio:3.6.0")
}
}
}
plugins {
id("com.github.ben-manes.versions") version("0.51.0")
}
```
## Limiting the dependencies that appear in the dependency graph
By default, the `dependency-submission` action attempts to detect all dependencies declared and used by your Gradle build.
At times it may helpful to limit the dependencies reported to GitHub, to avoid security alerts for dependencies that
don't form a critical part of your product. For example, a vulnerability in the tool you use to generate documentation
may not be as important as a vulnerability in one of your runtime dependencies.
The `dependency-submission` action provides a convenient mechanism to filter the projects and configurations that
contribute to the dependency graph.
> [!NOTE]
> Ideally, all dependencies involved in building and testing a project will be extracted and reported in a dependency graph.
> These dependencies would be assigned to different scopes (eg development, runtime, testing) and the GitHub UI would make it easy to opt-in to security alerts for different dependency scopes.
> However, this functionality does not yet exist.
### Excluding certain Gradle projects from the dependency graph
If you do not want the dependency graph to include dependencies from every project in your build,
you can easily exclude certain projects from the dependency extraction process.
To restrict which Gradle subprojects contribute to the report, specify which projects to exclude via a regular expression.
You can provide this value via the `DEPENDENCY_GRAPH_EXCLUDE_PROJECTS` environment variable or system property.
Note that excluding a project in this way only removes dependencies that are _resolved_ as part of that project, and may
not necessarily remove all dependencies _declared_ in that project. If another project depends on the excluded project
then it may transitively resolve dependencies declared in the excluded project: these dependencies will still be included
in the generated dependency graph.
### Excluding certain Gradle configurations from the dependency graph
Similarly to Gradle projects, it is possible to exclude a set of configuration instances from dependency graph generation,
so that dependencies resolved by those configurations are not included.
To restrict which Gradle configurations contribute to the report, specify which configurations to exclude via a regular expression.
You can provide this value via the `DEPENDENCY_GRAPH_EXCLUDE_CONFIGURATIONS` environment variable or system property.
Note that configuration exclusion applies to the configuration in which the dependency is _resolved_ which is not necessarily
the configuration where the dependency is _declared_. For example if you decare a dependency as `implementation` in
a Java project, that dependency will be resolved in `compileClasspath`, `runtimeClasspath` and possibly other configurations.
### Example of project and configuration filtering
For example, if you want to exclude dependencies in the `buildSrc` project, and exclude dependencies from the `testCompileClasspath` and `testRuntimeClasspath` configurations, you would use the following configuration:
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
env:
# Exclude all dependencies that originate solely in the 'buildSrc' project
DEPENDENCY_GRAPH_EXCLUDE_PROJECTS: ':buildSrc'
# Exclude dependencies that are only resolved in test classpaths
DEPENDENCY_GRAPH_EXCLUDE_CONFIGURATIONS: '.*[Tt]est(Compile|Runtime)Classpath'
```
### Other filtering options
The [GitHub Dependency Graph Gradle Plugin](https://plugins.gradle.org/plugin/org.gradle.github-dependency-graph-gradle-plugin)
has other filtering options that may be useful.
See [the docs](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#filtering-which-gradle-configurations-contribute-to-the-dependency-graph) for details.
# Advance usage scenarios
## Using a custom plugin repository
By default, the action downloads the `github-dependency-graph-gradle-plugin` from the Gradle Plugin Portal (https://plugins.gradle.org). If your GitHub Actions environment does not have access to this URL, you can specify a custom plugin repository to use.
Do so by setting the `GRADLE_PLUGIN_REPOSITORY_URL` environment variable.
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
env:
GRADLE_PLUGIN_REPOSITORY_URL: "https://gradle-plugins-proxy.mycorp.com"
```
## Integrating the `dependency-review-action`
The GitHub [dependency-review-action](https://github.com/actions/dependency-review-action) helps you
understand dependency changes (and the security impact of these changes) for a pull request,
by comparing the dependency graph for the pull-request with that of the HEAD commit.
Example of a pull request workflow that executes a build for a pull request and runs the `dependency-review-action`:
```yaml
name: Dependency review for pull requests
on: [ pull_request ]
permissions:
contents: write
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and submit dependency graph
uses: gradle/actions/dependency-submission@v3
dependency-review:
needs: dependency-submission
runs-on: ubuntu-latest
steps:
- name: Perform dependency review
uses: actions/dependency-review-action@v3
```
Note that the `dependency-submission` action submits the dependency graph at the completion of the workflow Job.
For this reason, the `dependency-review-action` must be executed in a dependent job, and not as a subsequent step in the job that generates the dependency graph.
## Usage with pull requests from public forked repositories
This `contents: write` permission is [not available for any workflow that is triggered by a pull request submitted from a public forked repository](https://docs.github.com/en/actions/security-guides/automatic-token-authentication#permissions-for-the-github_token).
This limitation is designed to prevent a malicious pull request from effecting repository changes.
Because of this restriction, we require 2 separate workflows in order to generate and submit a dependency graph:
1. The first workflow runs directly against the pull request sources and will `generate-and-upload` the dependency graph.
2. The second workflow is triggered on `workflow_run` of the first workflow, and will `download-and-submit` the previously saved dependency graph.
***Main workflow file***
```yaml
name: Generate and save dependency graph
on: [ pull_request ]
permissions:
contents: read # 'write' permission is not available
jobs:
dependency-submission:
runs-on: ubuntu-latest
steps:
- name: Checkout sources
uses: actions/checkout@v4
- name: Generate and save dependency graph
uses: gradle/actions/dependency-submission@v3
with:
dependency-graph: generate-and-upload
```
***Dependent workflow file***
```yaml
name: Download and submit dependency graph
on:
workflow_run:
workflows: ['Generate and save dependency graph']
types: [completed]
permissions:
contents: write
jobs:
submit-dependency-graph:
runs-on: ubuntu-latest
steps:
- name: Download and submit dependency graph
uses: gradle/actions/dependency-submission@v3
with:
dependency-graph: download-and-submit # Download saved dependency-graph and submit
```
### Integrating `dependency-review-action` for pull requests from public forked repositories
To integrate the `dependency-review-action` into the pull request workflows above, a third workflow file is required.
This workflow will be triggered directly on `pull_request`, but will wait until the dependency graph results are
submitted before the dependency review can complete. The period to wait is controlled by the `retry-on-snapshot-warnings` input parameters.
Here's an example of a separate "Dependency Review" workflow that will wait for 10 minutes for the above PR check workflow to complete.
```yaml
name: dependency-review
on: [ pull_request ]
permissions:
contents: read
jobs:
dependency-review:
runs-on: ubuntu-latest
steps:
- name: 'Dependency Review'
uses: actions/dependency-review-action@v3
with:
retry-on-snapshot-warnings: true
retry-on-snapshot-warnings-timeout: 600
```
The `retry-on-snapshot-warnings-timeout` (in seconds) needs to be long enough to allow the entire `Generate and save dependency graph` and `Download and submit dependency graph` workflows (above) to complete.
# Gradle version compatibility
Dependency-graph generation is compatible with most versions of Gradle >= `5.2`, and is tested regularly against
Gradle versions `5.2.1`, `5.6.4`, `6.0.1`, `6.9.4`, `7.1.1` and `7.6.3`, as well as all patched versions of Gradle 8.x.
A known exception to this is that Gradle `7.0`, `7.0.1` and `7.0.2` are not supported.
See [here](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#gradle-compatibility) for complete compatibility information.

716
docs/setup-gradle.md Normal file
View file

@ -0,0 +1,716 @@
# Configure Gradle for GitHub Actions workflows
This GitHub Action can be used to configure Gradle for optimal execution on any platform supported by GitHub Actions.
## Why use the `setup-gradle` action?
It is possible to directly invoke Gradle in your workflow, and the `actions/setup-java@v4` action provides a simple way to cache Gradle dependencies.
However, the `setup-gradle` action offers a several advantages over this approach:
- Easily [configure your workflow to use a specific version of Gradle](#choose-a-specific-gradle-version) using the `gradle-version` parameter. Gradle distributions are automatically downloaded and cached.
- More sophisticated and more efficient caching of Gradle User Home between invocations, compared to `setup-java` and most custom configurations using `actions/cache`. [More details below](#caching-build-state-between-jobs).
- Detailed reporting of cache usage and cache configuration options allow you to [optimize the use of the GitHub actions cache](#optimizing-cache-effectiveness).
- [Generate and Submit a GitHub Dependency Graph](#github-dependency-graph-support) for your project, enabling Dependabot security alerts.
- [Automatic capture of Build Scan® links](#build-reporting) from the build, making them easier to locate in workflow runs.
The `setup-gradle` action is designed to provide these benefits with minimal configuration.
These features work both when Gradle is executed via `setup-gradle` and for any Gradle execution in subsequent steps.
## General usage
The `setup-gradle` action works by configuring environment variables and by adding a set of Gradle init-scripts to the Gradle User Home. These will apply to all Gradle executions on the runner, no matter how Gradle is invoked.
This means that if you have an existing workflow that executes Gradle with a `run` step, you can add an initial "Setup Gradle" Step to benefit from caching, build-scan capture, and other features of this action.
The recommended way to execute any Gradle build is with the help of the [Gradle Wrapper](https://docs.gradle.org/current/userguide/gradle_wrapper.html), and the following examples assume that the Gradle Wrapper has been configured for the project. See [this example](#build-with-a-specific-gradle-version) if your project doesn't use the Gradle Wrapper.
```yaml
name: Run Gradle on every push
on: push
jobs:
gradle:
strategy:
matrix:
os: [ubuntu-latest, macos-latest, windows-latest]
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout@v4
- uses: actions/setup-java@v4
with:
distribution: temurin
java-version: 11
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
- name: Execute Gradle build
run: ./gradlew build
```
## Build with a specific Gradle version
The `setup-gradle` action can download and install a specified Gradle version, adding this installed version to the PATH.
Downloaded Gradle versions are stored in the GitHub Actions cache, to avoid having to download them again later.
```yaml
- name: Setup Gradle 8.5
uses: gradle/actions/setup-gradle@v3
with:
gradle-version: 8.5
- name: Build with Gradle 8.5
run: gradle build
```
The `gradle-version` parameter can be set to any valid Gradle version.
Moreover, you can use the following aliases:
| Alias | Selects |
| --- |---|
| `wrapper` | The Gradle wrapper's version (default, useful for matrix builds) |
| `current` | The current [stable release](https://gradle.org/install/) |
| `release-candidate` | The current [release candidate](https://gradle.org/release-candidate/) if any, otherwise fallback to `current` |
| `nightly` | The latest [nightly](https://gradle.org/nightly/), fails if none. |
| `release-nightly` | The latest [release nightly](https://gradle.org/release-nightly/), fails if none. |
This can be handy to automatically verify your build works with the latest release candidate of Gradle:
The actual Gradle version used is available as an action output: `gradle-version`.
```yaml
name: Test latest Gradle RC
on:
schedule:
- cron: 0 0 * * * # daily
jobs:
gradle-rc:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: actions/setup-java@v4
with:
distribution: temurin
java-version: 11
- uses: gradle/actions/setup-gradle@v3
id: setup-gradle
with:
gradle-version: release-candidate
- run: gradle build --dry-run # just test build configuration
- run: echo "The release-candidate version was ${{ steps.setup-gradle.outputs.gradle-version }}"
```
## Caching build state between Jobs
The `setup-gradle` action will use the GitHub Actions cache to save and restore reusable state that may speed up subsequent build invocations. This includes most content that is downloaded from the internet as part of a build, as well as expensive to create content like compiled build scripts, transformed Jar files, etc.
The cached state includes:
- Any distributions downloaded to satisfy a `gradle-version` parameter.
- A subset of the Gradle User Home directory, including downloaded dependencies, wrapper distributions, and the local build cache.
To reduce the space required for caching, this action attempts to reduce duplication in cache entries on a best effort basis.
The state will be restored from the cache during the first `setup-gradle` step for any workflow job, and cache entries will be written back to the cache at the end of the job after all Gradle executions have been completed.
### Disabling caching
Caching is enabled by default. You can disable caching for the action as follows:
```yaml
cache-disabled: true
```
### Using the cache read-only
By default, The `setup-gradle` action will only write to the cache from Jobs on the default (`main`/`master`) branch.
Jobs on other branches will read entries from the cache but will not write updated entries.
See [Optimizing cache effectiveness](#select-which-branches-should-write-to-the-cache) for a more detailed explanation.
In some circumstances, it makes sense to change this default and configure a workflow Job to read existing cache entries but not to write changes back.
You can configure read-only caching for `setup-gradle` as follows:
```yaml
cache-read-only: true
```
You can also configure read-only caching only for certain branches:
```yaml
# Only write to the cache for builds on the 'main' and 'release' branches. (Default is 'main' only.)
# Builds on other branches will only read existing entries from the cache.
cache-read-only: ${{ github.ref != 'refs/heads/main' && github.ref != 'refs/heads/release' }}
```
### Using the cache write-only
In certain circumstances it may be desirable to start with a clean Gradle User Home state, but to save the state at the end of a workflow Job:
```yaml
cache-write-only: true
```
### Overwriting an existing Gradle User Home
When the action detects that the Gradle User Home caches directory already exists (`~/.gradle/caches`), then by default it will not overwrite the existing content of this directory.
This can occur when a prior action initializes this directory, or when using a self-hosted runner that retains this directory between uses.
In this case, the Job Summary will display a message like:
> Caching for Gradle actions was disabled due to pre-existing Gradle User Home
If you want to override the default and have the caches of the `setup-gradle` action overwrite existing content in the Gradle User Home, you can set the `cache-overwrite-existing` parameter to `true`:
```yaml
cache-overwrite-existing: true
```
### Saving configuration-cache data
When Gradle is executed with the [configuration-cache](https://docs.gradle.org/current/userguide/configuration_cache.html) enabled, the configuration-cache data is stored
in the project directory, at `<project-dir>/.gradle/configuration-cache`. Due to the way the configuration-cache works, [this file may contain stored credentials and other
secrets](https://docs.gradle.org/release-nightly/userguide/configuration_cache.html#config_cache:secrets), and this data needs to be encrypted to be safely stored in the GitHub Actions cache.
To benefit from configuration caching in your GitHub Actions workflow, you must:
- Execute your build with Gradle 8.6 or newer. This can be achieved directly or via the Gradle Wrapper.
- Enable the configuration cache for your build.
- Generate a [valid Gradle encryption key](https://docs.gradle.org/8.6/userguide/configuration_cache.html#config_cache:secrets:configuring_encryption_key) and save it as a [GitHub Actions secret](https://docs.github.com/en/actions/security-guides/using-secrets-in-github-actions).
- Provide the secret key via the `cache-encryption-key` action parameter.
```yaml
jobs:
gradle-with-configuration-cache:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: gradle/actions/setup-gradle@v3
with:
gradle-version: 8.6
cache-encryption-key: ${{ secrets.GradleEncryptionKey }}
- run: gradle build --configuration-cache
```
### Incompatibility with other caching mechanisms
When using `setup-gradle` we recommend that you avoid using other mechanisms to save and restore the Gradle User Home.
Specifically:
- Avoid using `actions/cache` configured to cache the Gradle User Home, [as described in this example](https://github.com/actions/cache/blob/main/examples.md#java---gradle).
- Avoid using `actions/setup-java` with the `cache: gradle` option, [as described here](https://github.com/actions/setup-java#caching-gradle-dependencies).
Using either of these mechanisms may interfere with the caching provided by this action. If you choose to use a different mechanism to save and restore the Gradle User Home, you should disable the caching provided by this action, as described above.
## How Gradle User Home caching works
### Properties of the GitHub Actions cache
The GitHub Actions cache has some properties that present problems for efficient caching of the Gradle User Home.
- Immutable entries: once a cache entry is written for a key, it cannot be overwritten or changed.
- Branch scope: cache entries written for a Git branch are not visible from actions running against different branches. Entries written for the default branch are visible to all. https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows#restrictions-for-accessing-a-cache
- Restore keys: if no exact match is found, a set of partial keys can be provided that will match by cache key prefix. https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows#matching-a-cache-key
Each of these properties has influenced the design and implementation of the caching in `setup-gradle`, as described below.
### Which content is cached
Using experiments and observations, we have attempted to identify which Gradle User Home content is worth saving and restoring between build invocations. We considered both the respective size of the content and the impact this content has on build times. As well as the obvious candidates like downloaded dependencies, we saw that compiled build scripts, transformed Jar files, and other content can also have a significant impact.
In the end, we opted to save and restore as much content as is practical, including:
- `caches/<version>/generated-gradle-jars`: These files are generated on the first use of a particular Gradle version, and are expensive to recreate
- `caches/<version>/kotlin-dsl` and `caches/<version>/scripts`: These are the compiled build scripts. The Kotlin ones in particular can benefit from caching.
- `caches/modules-2`: The downloaded dependencies
- `caches/transforms-3`: The results of artifact transforms
- `caches/jars-9`: Jar files that have been processed/instrumented by Gradle
- `caches/build-cache-1`: The local build cache
In certain cases, a particular section of Gradle User Home will be too large to make caching effective. In these cases, particular subdirectories can be excluded from caching. See [Exclude content from Gradle User Home cache](#exclude-content-from-gradle-user-home-cache).
### Cache keys
The actual content of the Gradle User Home after a build is the result of many factors, including:
- Core Gradle build files (`settings.gradle[.kts]`, `build.gradle[.kts]`, `gradle.properties`)
- Associated Gradle configuration files (`gradle-wrapper.properties`, `dependencies.toml`, etc)
- The entire content of `buildSrc` or any included builds that provide plugins.
- The entire content of the repository, in the case of the local build cache.
- The actual build command that was invoked, including system properties and environment variables.
For this reason, it's very difficult to create a cache key that will deterministically map to a saved Gradle User Home state. So instead of trying to reliably hash all of these inputs to generate a cache key, the Gradle User Home cache key is based on the currently executing Job and the current commit hash for the repository.
The Gradle User Home cache key is composed of:
- The current operating system (`RUNNER_OS`)
- The Job id
- A hash of the Job matrix parameters and the workflow name
- The git SHA for the latest commit
Specifically, the cache key is: `${cache-protocol}-gradle|${runner-os}|${job-id}[${hash-of-job-matrix-and-workflow-name}]-${git-sha}`
As such, the cache key is likely to change on each subsequent run of GitHub actions.
This allows the most recent state to always be available in the GitHub actions cache.
### Finding a matching cache entry
In most cases, no exact match will exist for the cache key. Instead, the Gradle User Home will be restored for the closest matching cache entry, using a set of "restore keys". The entries will be matched with the following precedence:
- An exact match on OS, job id, workflow name, matrix, and Git SHA
- The most recent entry saved for the same OS, job id, workflow name, and matrix values
- The most recent entry saved for the same OS and job id
- The most recent entry saved for the same OS
Due to branch scoping of cache entries, the above match will be first performed for entries from the same branch, and then for the default ('main') branch.
After the Job is complete, the current Gradle User Home state will be collected and written as a new cache entry with the complete cache key. Old entries will be expunged from the GitHub Actions cache on a least recently used basis.
Note that while effective, this mechanism is not inherently efficient. It requires the entire Gradle User Home directory to be stored separately for each branch, for every OS+Job+Matrix combination. In addition, it writes a new cache entry on every GitHub Actions run.
This inefficiency is effectively mitigated by [Deduplication of Gradle User Home cache entries](#deduplication-of-gradle-user-home-cache-entries) and can be further optimized for a workflow using the techniques described in [Optimizing cache effectiveness](#optimizing-cache-effectiveness).
### Deduplication of Gradle User Home cache entries
To reduce duplication between cache entries, certain artifacts in Gradle User Home are extracted and cached independently based on their identity. This allows each Gradle User Home cache entry to be relatively small, sharing common elements between them without duplication.
Artifacts that are cached independently include:
- Downloaded dependencies
- Downloaded wrapper distributions
- Generated Gradle API jars
- Downloaded Java Toolchains
For example, this means that all jobs executing a particular version of the Gradle wrapper will share a single common entry for this wrapper distribution and one for each of the generated Gradle API jars.
### Stopping the Gradle daemon
By default, the action will stop all running Gradle daemons in the post-action step, before saving the Gradle User Home state.
This allows for any Gradle User Home cleanup to occur, and avoid file-locking issues on Windows.
If caching is disabled or the cache is in read-only mode, the daemon will not be stopped and will continue running after the job is completed.
## Optimizing cache effectiveness
Cache storage space for GitHub actions is limited, and writing new cache entries can trigger the deletion of existing entries.
Eviction of shared cache entries can reduce cache effectiveness, slowing down your `setup-gradle` steps.
There are a several actions you can take if your cache use is less effective due to entry eviction.
At the end of a Job, The `setup-gradle` action will write a summary of the Gradle builds executed, together with a detailed report of the cache entries that were read and written during the Job. This report can provide valuable insights that may help to determine the right way to optimize the cache usage for your workflow.
### Select which jobs should write to the cache
Consider a workflow that first runs a Job "compile-and-unit-test" to compile the code and run some basic unit tests, which is followed by a matrix of parallel "integration-test" jobs that each run a set of integration tests for the repository. Each "integration test" Job requires all of the dependencies required by "compile-and-unit-test", and possibly one or 2 additional dependencies.
By default, a new cache entry will be written on completion of each integration test job. If no additional dependencies were downloaded then this cache entry will share the "dependencies" entry with the "compile-and-unit-test" job, but if a single dependency was downloaded then an entirely new "dependencies" entry would be written. (The `setup-gradle` action does not _yet_ support a layered cache that could do this more efficiently). If each of these "integration-test" entries with their different "dependencies" entries is too large, then it could result in other important entries being evicted from the GitHub Actions cache.
Some techniques can be used to avoid/mitigate this issue:
- Configure the "integration-test" jobs with `cache-read-only: true`, meaning that the Job will use the entry written by the "compile-and-unit-test" job. This will avoid the overhead of cache entries for each of these jobs, at the expense of re-downloading any additional dependencies required by "integration-test".
- Add a step to the "compile-and-unit-test" job which downloads all dependencies required by the integration-test jobs but does not execute the tests. This will allow the "dependencies" entry for "compile-and-unit-test" to be shared among all cache entries for "integration-test". The resulting "integration-test" entries should be much smaller, reducing the potential for eviction.
- Combine the above 2 techniques, so that no cache entry is written by "integration-test" jobs, but all required dependencies are already present from the restored "compile-and-unit-test" entry.
### Select which branches should write to the cache
GitHub cache entries are not shared between builds on different branches.
Workflow runs can restore caches created in either the current branch or the default branch (usually main).
This means that each branch will have its own Gradle User Home cache scope, and will not benefit from cache entries written for other (non-default) branches.
By default, The `setup-gradle` action will only _write_ to the cache for builds run on the default (`master`/`main`) branch.
Jobs running on other branches will only read from the cache. In most cases, this is the desired behavior.
This is because Jobs running on other branches will benefit from the cached Gradle User Home from `main`,
without writing private cache entries which could lead to evicting these shared entries.
If you have other long-lived development branches that would benefit from writing to the cache,
you can configure this by disabling the `cache-read-only` action parameter for these branches.
See [Using the cache read-only](#using-the-cache-read-only) for more details.
Note there are some cases where writing cache entries is typically unhelpful (these are disabled by default):
- For `pull_request` triggered runs, the cache scope is limited to the merge ref (`refs/pull/.../merge`) and can only be restored by re-runs of the same pull request.
- For `merge_group` triggered runs, the cache scope is limited to a temporary branch with a special prefix created to validate pull request changes, and won't be available on subsequent Merge Queue executions.
### Exclude content from Gradle User Home cache
As well as any wrapper distributions, the action will attempt to save and restore the `caches` and `notifications` directories from Gradle User Home.
Each build is different, and some builds produce more Gradle User Home content than others.
[Cache debugging ](#cache-debugging-and-analysis) can provide insight into which cache entries are the largest,
and the contents to be cached can be fine-tuned by including and excluding certain paths within the Gradle User Home.
```yaml
# Cache downloaded JDKs in addition to the default directories.
gradle-home-cache-includes: |
caches
notifications
jdks
# Exclude the local build-cache and keyrings from the directories cached.
gradle-home-cache-excludes: |
caches/build-cache-1
caches/keyrings
```
You can specify any number of fixed paths or patterns to include or exclude.
File pattern support is documented at https://docs.github.com/en/actions/learn-github-actions/workflow-syntax-for-github-actions#patterns-to-match-file-paths.
### Remove unused files from Gradle User Home before saving to the cache
The Gradle User Home directory tends to grow over time. When you switch to a new Gradle wrapper version or upgrade a dependency version
the old files are not automatically and immediately removed. While this can make sense in a local environment, in a GitHub Actions environment
it can lead to ever-larger Gradle User Home cache entries being saved and restored.
To avoid this situation, The `setup-gradle` action supports the `gradle-home-cache-cleanup` parameter.
When enabled, this feature will attempt to delete any files in the Gradle User Home that were not used by Gradle during the GitHub Actions Workflow, before saving the Gradle User Home to the GitHub Actions cache.
Gradle Home cache cleanup is considered experimental and is disabled by default. You can enable this feature for the action as follows:
```yaml
gradle-home-cache-cleanup: true
```
## Debugging and Troubleshooting
To debug a failed job, it can be useful to run with [debug logging enabled](https://docs.github.com/en/actions/monitoring-and-troubleshooting-workflows/enabling-debug-logging).
You can enable debug logging either by:
1. Adding an `ACTIONS_STEP_DEBUG` variable to your repository configuration ([see here](https://docs.github.com/en/actions/monitoring-and-troubleshooting-workflows/enabling-debug-logging#enabling-step-debug-logging)).
2. By re-running a Job and checking the "Enable debug logging" box ([see here](https://github.blog/changelog/2022-05-24-github-actions-re-run-jobs-with-debug-logging/)).
### Increased logging from Gradle builds
When debug logging is enabled, this action will cause all builds to run with the `--info` and `--stacktrace` options.
This is done by inserting the relevant [Gradle properties](https://docs.gradle.org/current/userguide/build_environment.html#sec:gradle_configuration_properties)
at the top of the `${GRADLE_USER_HOME}/gradle.properties` file.
If the additional Gradle logging produced is problematic, you may opt out of this behavior by setting these properties manually in your project `gradle.properties` file:
```properties
# default lifecycle
org.gradle.logging.level=lifecycle
org.gradle.logging.stacktrace=internal
```
### Cache debugging and analysis
A report of all cache entries restored and saved is printed to the Job Summary when saving the cache entries.
This report can provide valuable insight into how much cache space is being used.
When debug logging is enabled, more detailed logging of cache operations is included in the GitHub actions log.
This includes a breakdown of the contents of the Gradle User Home directory, which may assist in cache optimization.
## Build reporting
The `setup-gradle` action collects information about any Gradle executions that occur in a workflow, including the root project,
requested tasks, build outcome, and any Build Scan link generated. Details of cache entries read and written are also collected.
These details are compiled into a Job Summary, which is visible in the GitHub Actions UI.
Generation of a Job Summary is enabled by default for all Jobs using The `setup-gradle` action. This feature can be configured
so that a Job Summary is never generated, or so that a Job Summary is only generated on build failure:
```yaml
add-job-summary: 'on-failure' # Valid values are 'always' (default), 'never', and 'on-failure'
```
### Adding Job Summary as a Pull Request comment
It is sometimes more convenient to view the results of a GitHub Actions Job directly from the Pull Request that triggered
the Job. For this purpose, you can configure the action so that Job Summary data is added as a Pull Request comment.
```yaml
name: CI
on:
pull_request:
permissions:
pull-requests: write
jobs:
run-gradle-build:
runs-on: ubuntu-latest
steps:
- name: Checkout project sources
uses: actions/checkout@v4
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
with:
add-job-summary-as-pr-comment: on-failure # Valid values are 'never' (default), 'always', and 'on-failure'
- run: ./gradlew build --scan
```
Note that to add a Pull Request comment, the workflow must be configured with the `pull-requests: write` permission.
### Build Scan® link as Step output
As well as reporting all [Build Scan](https://gradle.com/build-scans/) links in the Job Summary,
The `setup-gradle` action makes this link available as an output of any Step that executes Gradle.
The output name is `build-scan-url`. You can then use the build scan link in subsequent actions of your workflow.
### Saving arbitrary build outputs
By default, a GitHub Actions workflow using `setup-gradle` will record the log output and any Build Scan
links for your build, but any output files generated by the build will not be saved.
To save selected files from your build execution, you can use the core [Upload-Artifact](https://github.com/actions/upload-artifact) action.
For example:
```yaml
jobs:
gradle:
runs-on: ubuntu-latest
steps:
- name: Checkout project sources
uses: actions/checkout@v4
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
- name: Run build with Gradle wrapper
run: ./gradlew build --scan
- name: Upload build reports
uses: actions/upload-artifact@v3
if: always()
with:
name: build-reports
path: build/reports/
```
### Use of custom init-scripts in Gradle User Home
Note that the action collects information about Gradle invocations via an [Initialization Script](https://docs.gradle.org/current/userguide/init_scripts.html#sec:using_an_init_script)
located at `USER_HOME/.gradle/init.d/gradle-actions.build-result-capture.init.gradle`.
If you are adding any custom init scripts to the `USER_HOME/.gradle/init.d` directory, it may be necessary to ensure these files are applied before `gradle-actions.build-result-capture.init.gradle`.
Since Gradle applies init scripts in alphabetical order, one way to ensure this is via file naming.
## Support for GitHub Enterprise Server (GHES)
You can use the `setup-gradle` action on GitHub Enterprise Server, and benefit from the improved integration with Gradle. Depending on the version of GHES you are running, certain features may be limited:
- Build Scan links are captured and displayed in the GitHub Actions UI
- Easily run your build with different versions of Gradle
- Save/restore of Gradle User Home (requires GHES v3.5+ : GitHub Actions cache was introduced in GHES 3.5)
- Support for GitHub Actions Job Summary (requires GHES 3.6+ : GitHub Actions Job Summary support was introduced in GHES 3.6). In earlier versions of GHES, the build-results summary and caching report will be written to the workflow log, as part of the post-action step.
## GitHub Dependency Graph support
> [!IMPORTANT]
> The simplest (and recommended) way to generate a dependency graph is via a separate workflow
> using `gradle/actions/dependency-submission`. This action will attempt to detect all dependencies used by your build
> without building and testing the project itself.
>
> See the [dependency-submission documentation](dependency-submission.md) for up-to-date documentation.
The `setup-gradle` action has support for submitting a [GitHub Dependency Graph](https://docs.github.com/en/code-security/supply-chain-security/understanding-your-software-supply-chain/about-the-dependency-graph) snapshot via the [GitHub Dependency Submission API](https://docs.github.com/en/rest/dependency-graph/dependency-submission?apiVersion=2022-11-28).
The dependency graph snapshot is generated via integration with the [GitHub Dependency Graph Gradle Plugin](https://plugins.gradle.org/plugin/org.gradle.github-dependency-graph-gradle-plugin) and saved as a workflow artifact. The generated snapshot files can be submitted either in the same job or in a subsequent job (in the same or a dependent workflow).
The generated dependency graph snapshot reports all of the dependencies that were resolved during a build execution, and is used by GitHub to generate [Dependabot Alerts](https://docs.github.com/en/code-security/dependabot/dependabot-alerts/about-dependabot-alerts) for vulnerable dependencies, as well as to populate the [Dependency Graph insights view](https://docs.github.com/en/code-security/supply-chain-security/understanding-your-software-supply-chain/exploring-the-dependencies-of-a-repository#viewing-the-dependency-graph).
### Basic usage
You enable GitHub Dependency Graph support by setting the `dependency-graph` action parameter. Valid values are:
| Option | Behaviour |
| --- | --- |
| `disabled` | Do not generate a dependency graph for any build invocations.<p>This is the default. |
| `generate` | Generate a dependency graph snapshot for each build invocation. |
| `generate-and-submit` | Generate a dependency graph snapshot for each build invocation, and submit these via the Dependency Submission API on completion of the job. |
| `generate-and-upload` | Generate a dependency graph snapshot for each build invocation, saving it as a workflow artifact. |
| `download-and-submit` | Download any previously saved dependency graph snapshots, and submit them via the Dependency Submission API. This can be useful to submit [dependency graphs for pull requests submitted from repository forks](#dependency-graphs-for-pull-request-workflows). |
Example of a CI workflow that generates and submits a dependency graph:
```yaml
name: CI build
on:
push:
permissions:
contents: write
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to generate and submit dependency graphs
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
- name: Run the usual CI build (dependency-graph will be generated and submitted post-job)
run: ./gradlew build
```
The `contents: write` permission is required to submit (but not generate) the dependency graph file.
Depending on [repository settings](https://docs.github.com/en/actions/security-guides/automatic-token-authentication#permissions-for-the-github_token), this permission may be available by default or may need to be explicitly enabled in the workflow file (as above).
> [!IMPORTANT]
> The above configuration will work for workflows that run as a result of commits to a repository branch,
> but not when a workflow is triggered by a PR from a repository fork.
> This is because the `contents: write` permission is not available when executing a workflow
> for a PR submitted from a forked repository.
> For a configuration that supports this setup, see [Dependency Graphs for pull request workflows](#dependency-graphs-for-pull-request-workflows).
### Making dependency graph failures cause Job failures
By default, if a failure is encountered when generating or submitting the dependency graph, the action will log the failure as a warning and continue.
This allows your workflow to be resilient to dependency graph failures, in case dependency graph production is a side-effect rather than the primary purpose of a workflow.
If instead, you have a workflow whose primary purpose is to generate and submit a dependency graph, it makes sense for this workflow to fail if the dependency
graph cannot be generated or submitted. You can enable this behavior with the `dependency-graph-continue-on-failure` parameter, which defaults to `true`.
```yaml
# Ensure that the workflow Job will fail if the dependency graph cannot be submitted
- uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
dependency-graph-continue-on-failure: false
```
### Using a custom plugin repository
By default, the action downloads the `github-dependency-graph-gradle-plugin` from the Gradle Plugin Portal (https://plugins.gradle.org). If your GitHub Actions environment does not have access to this URL, you can specify a custom plugin repository to use.
Do so by setting the `GRADLE_PLUGIN_REPOSITORY_URL` environment variable with your Gradle invocation.
The `GRADLE_PLUGIN_REPOSITORY_USERNAME` and `GRADLE_PLUGIN_REPOSITORY_PASSWORD` can be used when the plugin repository requires authentication.
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to generate and submit dependency graphs
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
- name: Run a build, resolving the 'dependency-graph' plugin from the plugin portal proxy
run: ./gradlew build
env:
GRADLE_PLUGIN_REPOSITORY_URL: "https://gradle-plugins-proxy.mycorp.com"
# Set the following variables if your custom plugin repository requires authentication
# GRADLE_PLUGIN_REPOSITORY_USERNAME: "username"
# GRADLE_PLUGIN_REPOSITORY_PASSWORD: ${secrets.MY_REPOSITORY_PASSWORD}
```
### Choosing which Gradle invocations will generate a dependency graph
Once you enable the dependency graph support for a workflow job (via the `dependency-graph` parameter), dependencies will be collected and reported for all subsequent Gradle invocations.
If you have a Gradle build step that you want to exclude from dependency graph generation, you can set the `GITHUB_DEPENDENCY_GRAPH_ENABLED` environment variable to `false`.
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to generate and submit dependency graphs
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
- name: Build the app, generating a graph of dependencies required
run: ./gradlew :my-app:assemble
- name: Run all checks, disabling dependency graph generation
run: ./gradlew check
env:
GITHUB_DEPENDENCY_GRAPH_ENABLED: false
```
### Filtering which Gradle Configurations contribute to the dependency graph
If you do not want the dependency graph to include every dependency configuration in every project in your build,
you can limit the dependency extraction to a subset of these.
See the documentation for [dependency-submission](dependency-submission.md) and the
[GitHub Dependency Graph Gradle Plugin](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#filtering-which-gradle-configurations-contribute-to-the-dependency-graph) for details.
### Gradle version compatibility
Dependency-graph generation is compatible with most versions of Gradle >= `5.2`, and is tested regularly against
Gradle versions `5.2.1`, `5.6.4`, `6.0.1`, `6.9.4`, `7.1.1` and `7.6.3`, as well as all patched versions of Gradle 8.x.
A known exception to this is that Gradle `7.0`, `7.0.1`, and `7.0.2` are not supported.
See [here](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#gradle-compatibility) for complete compatibility information.
### Reducing storage costs for saved dependency graph artifacts
When `generate` or `generate-and-submit` is used with the action, the dependency graph that is generated is stored as a workflow artifact.
By default, these artifacts are retained for 30 days (or as configured for the repository).
To reduce storage costs for these artifacts, you can set the `artifact-retention-days` value to a lower number.
```yaml
steps:
- name: Generate dependency graph, but only retain artifact for one day
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate
artifact-retention-days: 1
```
# Develocity plugin injection
The `setup-gradle` action provides support for injecting and configuring the Develocity Gradle plugin into any Gradle build, without any modification to the project sources.
This is achieved via an init-script installed into Gradle User Home, which is enabled and parameterized via environment variables.
The same auto-injection behavior is available for the Common Custom User Data Gradle plugin, which enriches any build scans published with additional useful information.
## Enabling Develocity injection
To enable Develocity injection for your build, you must provide the required configuration via environment variables.
Here's a minimal example:
```yaml
name: Run build with Develocity injection
env:
DEVELOCITY_INJECTION_ENABLED: true
DEVELOCITY_URL: https://develocity.your-server.com
DEVELOCITY_PLUGIN_VERSION: 3.17
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
- name: Run a Gradle build with Develocity injection enabled
run: ./gradlew build
```
This configuration will automatically apply `v3.17` of the [Develocity Gradle plugin](https://docs.gradle.com/develocity/gradle-plugin/), and publish build scans to https://develocity.your-server.com.
This example assumes that the `develocity.your-server.com` server allows anonymous publishing of build scans.
In the likely scenario that your Develocity server requires authentication, you will also need to configure an additional environment variable
with a valid [Develocity access key](https://docs.gradle.com/develocity/gradle-plugin/#via_environment_variable).
## Configuring Develocity injection
The `init-script` supports several additional configuration parameters that you may find useful. All configuration options (required and optional) are detailed below:
| Variable | Required | Description |
|--------------------------------------| --- |-------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| DEVELOCITY_INJECTION_ENABLED | :white_check_mark: | enables Develocity injection |
| DEVELOCITY_URL | :white_check_mark: | the URL of the Develocity server |
| DEVELOCITY_ALLOW_UNTRUSTED_SERVER | | allow communication with an untrusted server; set to _true_ if your Develocity instance is using a self-signed certificate |
| DEVELOCITY_CAPTURE_FILE_FINGERPRINTS | | enables capturing the paths and content hashes of each individual input file |
| DEVELOCITY_ENFORCE_URL | | enforce the configured Develocity URL over a URL configured in the project's build; set to _true_ to enforce publication of build scans to the configured Develocity URL |
| DEVELOCITY_PLUGIN_VERSION | :white_check_mark: | the version of the [Develocity Gradle plugin](https://docs.gradle.com/develocity/gradle-plugin/) to apply |
| DEVELOCITY_CCUD_PLUGIN_VERSION | | the version of the [Common Custom User Data Gradle plugin](https://github.com/gradle/common-custom-user-data-gradle-plugin) to apply, if any |
| GRADLE_PLUGIN_REPOSITORY_URL | | the URL of the repository to use when resolving the Develocity and CCUD plugins; the Gradle Plugin Portal is used by default |
| GRADLE_PLUGIN_REPOSITORY_USERNAME | | the username for the repository URL to use when resolving the Develocity and CCUD plugins |
| GRADLE_PLUGIN_REPOSITORY_PASSWORD | | the password for the repository URL to use when resolving the Develocity and CCUD plugins; Consider using secrets to pass the value to this variable |
## Publishing to scans.gradle.com
Develocity injection is designed to enable the publishing of build scans to a Develocity instance,
but is also useful for publishing to the public Build Scans instance (https://scans.gradle.com).
To publish to https://scans.gradle.com, you must specify in your workflow that you accept the [Gradle Terms of Use](https://gradle.com/help/legal-terms-of-use).
```yaml
name: Run build and publish Build Scan
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to publish build scans
uses: gradle/actions/setup-gradle@v3
with:
build-scan-publish: true
build-scan-terms-of-use-url: "https://gradle.com/terms-of-service"
build-scan-terms-of-use-agree: "yes"
- name: Run a Gradle build - a build scan will be published automatically
run: ./gradlew build
```

714
setup-gradle/README.md
View file

@ -1,716 +1,28 @@
# Configure Gradle for GitHub Actions workflows
## The `setup-gradle` action
This GitHub Action can be used to configure Gradle for optimal execution on any platform supported by GitHub Actions.
The `setup-gradle` action can be used to configure Gradle for optimal execution on any platform supported by GitHub Actions.
## Why use the `setup-gradle` action?
This replaces the previous `gradle/gradle-build-action`, which now delegates to this implementation.
It is possible to directly invoke Gradle in your workflow, and the `actions/setup-java@v4` action provides a simple way to cache Gradle dependencies.
However, the `setup-gradle` action offers a several advantages over this approach:
- Easily [configure your workflow to use a specific version of Gradle](#choose-a-specific-gradle-version) using the `gradle-version` parameter. Gradle distributions are automatically downloaded and cached.
- More sophisticated and more efficient caching of Gradle User Home between invocations, compared to `setup-java` and most custom configurations using `actions/cache`. [More details below](#caching-build-state-between-jobs).
- Detailed reporting of cache usage and cache configuration options allow you to [optimize the use of the GitHub actions cache](#optimizing-cache-effectiveness).
- [Generate and Submit a GitHub Dependency Graph](#github-dependency-graph-support) for your project, enabling Dependabot security alerts.
- [Automatic capture of Build Scan® links](#build-reporting) from the build, making them easier to locate in workflow runs.
The `setup-gradle` action is designed to provide these benefits with minimal configuration.
These features work both when Gradle is executed via `setup-gradle` and for any Gradle execution in subsequent steps.
## General usage
The `setup-gradle` action works by configuring environment variables and by adding a set of Gradle init-scripts to the Gradle User Home. These will apply to all Gradle executions on the runner, no matter how Gradle is invoked.
This means that if you have an existing workflow that executes Gradle with a `run` step, you can add an initial "Setup Gradle" Step to benefit from caching, build-scan capture, and other features of this action.
The recommended way to execute any Gradle build is with the help of the [Gradle Wrapper](https://docs.gradle.org/current/userguide/gradle_wrapper.html), and the following examples assume that the Gradle Wrapper has been configured for the project. See [this example](#build-with-a-specific-gradle-version) if your project doesn't use the Gradle Wrapper.
The recommended way to execute any Gradle build is with the help of the [Gradle Wrapper](https://docs.gradle.org/current/userguide/gradle_wrapper.html), and the examples assume that the Gradle Wrapper has been configured for the project. See [this example](../docs/setup-gradle.md#build-with-a-specific-gradle-version) if your project doesn't use the Gradle Wrapper.
### Example usage
```yaml
name: Run Gradle on every push
on: push
name: Build
on: [ push ]
jobs:
gradle:
strategy:
matrix:
os: [ubuntu-latest, macos-latest, windows-latest]
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout@v4
- uses: actions/setup-java@v4
with:
distribution: temurin
java-version: 11
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
- name: Execute Gradle build
run: ./gradlew build
```
## Build with a specific Gradle version
The `setup-gradle` action can download and install a specified Gradle version, adding this installed version to the PATH.
Downloaded Gradle versions are stored in the GitHub Actions cache, to avoid having to download them again later.
```yaml
- name: Setup Gradle 8.5
uses: gradle/actions/setup-gradle@v3
with:
gradle-version: 8.5
- name: Build with Gradle 8.5
run: gradle build
```
The `gradle-version` parameter can be set to any valid Gradle version.
Moreover, you can use the following aliases:
| Alias | Selects |
| --- |---|
| `wrapper` | The Gradle wrapper's version (default, useful for matrix builds) |
| `current` | The current [stable release](https://gradle.org/install/) |
| `release-candidate` | The current [release candidate](https://gradle.org/release-candidate/) if any, otherwise fallback to `current` |
| `nightly` | The latest [nightly](https://gradle.org/nightly/), fails if none. |
| `release-nightly` | The latest [release nightly](https://gradle.org/release-nightly/), fails if none. |
This can be handy to automatically verify your build works with the latest release candidate of Gradle:
The actual Gradle version used is available as an action output: `gradle-version`.
```yaml
name: Test latest Gradle RC
on:
schedule:
- cron: 0 0 * * * # daily
jobs:
gradle-rc:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: actions/setup-java@v4
with:
distribution: temurin
java-version: 11
- uses: gradle/actions/setup-gradle@v3
id: setup-gradle
with:
gradle-version: release-candidate
- run: gradle build --dry-run # just test build configuration
- run: echo "The release-candidate version was ${{ steps.setup-gradle.outputs.gradle-version }}"
```
## Caching build state between Jobs
The `setup-gradle` action will use the GitHub Actions cache to save and restore reusable state that may speed up subsequent build invocations. This includes most content that is downloaded from the internet as part of a build, as well as expensive to create content like compiled build scripts, transformed Jar files, etc.
The cached state includes:
- Any distributions downloaded to satisfy a `gradle-version` parameter.
- A subset of the Gradle User Home directory, including downloaded dependencies, wrapper distributions, and the local build cache.
To reduce the space required for caching, this action attempts to reduce duplication in cache entries on a best effort basis.
The state will be restored from the cache during the first `setup-gradle` step for any workflow job, and cache entries will be written back to the cache at the end of the job after all Gradle executions have been completed.
### Disabling caching
Caching is enabled by default. You can disable caching for the action as follows:
```yaml
cache-disabled: true
```
### Using the cache read-only
By default, The `setup-gradle` action will only write to the cache from Jobs on the default (`main`/`master`) branch.
Jobs on other branches will read entries from the cache but will not write updated entries.
See [Optimizing cache effectiveness](#select-which-branches-should-write-to-the-cache) for a more detailed explanation.
In some circumstances, it makes sense to change this default and configure a workflow Job to read existing cache entries but not to write changes back.
You can configure read-only caching for `setup-gradle` as follows:
```yaml
cache-read-only: true
```
You can also configure read-only caching only for certain branches:
```yaml
# Only write to the cache for builds on the 'main' and 'release' branches. (Default is 'main' only.)
# Builds on other branches will only read existing entries from the cache.
cache-read-only: ${{ github.ref != 'refs/heads/main' && github.ref != 'refs/heads/release' }}
```
### Using the cache write-only
In certain circumstances it may be desirable to start with a clean Gradle User Home state, but to save the state at the end of a workflow Job:
```yaml
cache-write-only: true
```
### Overwriting an existing Gradle User Home
When the action detects that the Gradle User Home caches directory already exists (`~/.gradle/caches`), then by default it will not overwrite the existing content of this directory.
This can occur when a prior action initializes this directory, or when using a self-hosted runner that retains this directory between uses.
In this case, the Job Summary will display a message like:
> Caching for Gradle actions was disabled due to pre-existing Gradle User Home
If you want to override the default and have the caches of the `setup-gradle` action overwrite existing content in the Gradle User Home, you can set the `cache-overwrite-existing` parameter to `true`:
```yaml
cache-overwrite-existing: true
```
### Saving configuration-cache data
When Gradle is executed with the [configuration-cache](https://docs.gradle.org/current/userguide/configuration_cache.html) enabled, the configuration-cache data is stored
in the project directory, at `<project-dir>/.gradle/configuration-cache`. Due to the way the configuration-cache works, [this file may contain stored credentials and other
secrets](https://docs.gradle.org/release-nightly/userguide/configuration_cache.html#config_cache:secrets), and this data needs to be encrypted to be safely stored in the GitHub Actions cache.
To benefit from configuration caching in your GitHub Actions workflow, you must:
- Execute your build with Gradle 8.6 or newer. This can be achieved directly or via the Gradle Wrapper.
- Enable the configuration cache for your build.
- Generate a [valid Gradle encryption key](https://docs.gradle.org/8.6/userguide/configuration_cache.html#config_cache:secrets:configuring_encryption_key) and save it as a [GitHub Actions secret](https://docs.github.com/en/actions/security-guides/using-secrets-in-github-actions).
- Provide the secret key via the `cache-encryption-key` action parameter.
```yaml
jobs:
gradle-with-configuration-cache:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: gradle/actions/setup-gradle@v3
with:
gradle-version: 8.6
cache-encryption-key: ${{ secrets.GradleEncryptionKey }}
- run: gradle build --configuration-cache
```
### Incompatibility with other caching mechanisms
When using `setup-gradle` we recommend that you avoid using other mechanisms to save and restore the Gradle User Home.
Specifically:
- Avoid using `actions/cache` configured to cache the Gradle User Home, [as described in this example](https://github.com/actions/cache/blob/main/examples.md#java---gradle).
- Avoid using `actions/setup-java` with the `cache: gradle` option, [as described here](https://github.com/actions/setup-java#caching-gradle-dependencies).
Using either of these mechanisms may interfere with the caching provided by this action. If you choose to use a different mechanism to save and restore the Gradle User Home, you should disable the caching provided by this action, as described above.
## How Gradle User Home caching works
### Properties of the GitHub Actions cache
The GitHub Actions cache has some properties that present problems for efficient caching of the Gradle User Home.
- Immutable entries: once a cache entry is written for a key, it cannot be overwritten or changed.
- Branch scope: cache entries written for a Git branch are not visible from actions running against different branches. Entries written for the default branch are visible to all. https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows#restrictions-for-accessing-a-cache
- Restore keys: if no exact match is found, a set of partial keys can be provided that will match by cache key prefix. https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows#matching-a-cache-key
Each of these properties has influenced the design and implementation of the caching in `setup-gradle`, as described below.
### Which content is cached
Using experiments and observations, we have attempted to identify which Gradle User Home content is worth saving and restoring between build invocations. We considered both the respective size of the content and the impact this content has on build times. As well as the obvious candidates like downloaded dependencies, we saw that compiled build scripts, transformed Jar files, and other content can also have a significant impact.
In the end, we opted to save and restore as much content as is practical, including:
- `caches/<version>/generated-gradle-jars`: These files are generated on the first use of a particular Gradle version, and are expensive to recreate
- `caches/<version>/kotlin-dsl` and `caches/<version>/scripts`: These are the compiled build scripts. The Kotlin ones in particular can benefit from caching.
- `caches/modules-2`: The downloaded dependencies
- `caches/transforms-3`: The results of artifact transforms
- `caches/jars-9`: Jar files that have been processed/instrumented by Gradle
- `caches/build-cache-1`: The local build cache
In certain cases, a particular section of Gradle User Home will be too large to make caching effective. In these cases, particular subdirectories can be excluded from caching. See [Exclude content from Gradle User Home cache](#exclude-content-from-gradle-user-home-cache).
### Cache keys
The actual content of the Gradle User Home after a build is the result of many factors, including:
- Core Gradle build files (`settings.gradle[.kts]`, `build.gradle[.kts]`, `gradle.properties`)
- Associated Gradle configuration files (`gradle-wrapper.properties`, `dependencies.toml`, etc)
- The entire content of `buildSrc` or any included builds that provide plugins.
- The entire content of the repository, in the case of the local build cache.
- The actual build command that was invoked, including system properties and environment variables.
For this reason, it's very difficult to create a cache key that will deterministically map to a saved Gradle User Home state. So instead of trying to reliably hash all of these inputs to generate a cache key, the Gradle User Home cache key is based on the currently executing Job and the current commit hash for the repository.
The Gradle User Home cache key is composed of:
- The current operating system (`RUNNER_OS`)
- The Job id
- A hash of the Job matrix parameters and the workflow name
- The git SHA for the latest commit
Specifically, the cache key is: `${cache-protocol}-gradle|${runner-os}|${job-id}[${hash-of-job-matrix-and-workflow-name}]-${git-sha}`
As such, the cache key is likely to change on each subsequent run of GitHub actions.
This allows the most recent state to always be available in the GitHub actions cache.
### Finding a matching cache entry
In most cases, no exact match will exist for the cache key. Instead, the Gradle User Home will be restored for the closest matching cache entry, using a set of "restore keys". The entries will be matched with the following precedence:
- An exact match on OS, job id, workflow name, matrix, and Git SHA
- The most recent entry saved for the same OS, job id, workflow name, and matrix values
- The most recent entry saved for the same OS and job id
- The most recent entry saved for the same OS
Due to branch scoping of cache entries, the above match will be first performed for entries from the same branch, and then for the default ('main') branch.
After the Job is complete, the current Gradle User Home state will be collected and written as a new cache entry with the complete cache key. Old entries will be expunged from the GitHub Actions cache on a least recently used basis.
Note that while effective, this mechanism is not inherently efficient. It requires the entire Gradle User Home directory to be stored separately for each branch, for every OS+Job+Matrix combination. In addition, it writes a new cache entry on every GitHub Actions run.
This inefficiency is effectively mitigated by [Deduplication of Gradle User Home cache entries](#deduplication-of-gradle-user-home-cache-entries) and can be further optimized for a workflow using the techniques described in [Optimizing cache effectiveness](#optimizing-cache-effectiveness).
### Deduplication of Gradle User Home cache entries
To reduce duplication between cache entries, certain artifacts in Gradle User Home are extracted and cached independently based on their identity. This allows each Gradle User Home cache entry to be relatively small, sharing common elements between them without duplication.
Artifacts that are cached independently include:
- Downloaded dependencies
- Downloaded wrapper distributions
- Generated Gradle API jars
- Downloaded Java Toolchains
For example, this means that all jobs executing a particular version of the Gradle wrapper will share a single common entry for this wrapper distribution and one for each of the generated Gradle API jars.
### Stopping the Gradle daemon
By default, the action will stop all running Gradle daemons in the post-action step, before saving the Gradle User Home state.
This allows for any Gradle User Home cleanup to occur, and avoid file-locking issues on Windows.
If caching is disabled or the cache is in read-only mode, the daemon will not be stopped and will continue running after the job is completed.
## Optimizing cache effectiveness
Cache storage space for GitHub actions is limited, and writing new cache entries can trigger the deletion of existing entries.
Eviction of shared cache entries can reduce cache effectiveness, slowing down your `setup-gradle` steps.
There are a several actions you can take if your cache use is less effective due to entry eviction.
At the end of a Job, The `setup-gradle` action will write a summary of the Gradle builds executed, together with a detailed report of the cache entries that were read and written during the Job. This report can provide valuable insights that may help to determine the right way to optimize the cache usage for your workflow.
### Select which jobs should write to the cache
Consider a workflow that first runs a Job "compile-and-unit-test" to compile the code and run some basic unit tests, which is followed by a matrix of parallel "integration-test" jobs that each run a set of integration tests for the repository. Each "integration test" Job requires all of the dependencies required by "compile-and-unit-test", and possibly one or 2 additional dependencies.
By default, a new cache entry will be written on completion of each integration test job. If no additional dependencies were downloaded then this cache entry will share the "dependencies" entry with the "compile-and-unit-test" job, but if a single dependency was downloaded then an entirely new "dependencies" entry would be written. (The `setup-gradle` action does not _yet_ support a layered cache that could do this more efficiently). If each of these "integration-test" entries with their different "dependencies" entries is too large, then it could result in other important entries being evicted from the GitHub Actions cache.
Some techniques can be used to avoid/mitigate this issue:
- Configure the "integration-test" jobs with `cache-read-only: true`, meaning that the Job will use the entry written by the "compile-and-unit-test" job. This will avoid the overhead of cache entries for each of these jobs, at the expense of re-downloading any additional dependencies required by "integration-test".
- Add a step to the "compile-and-unit-test" job which downloads all dependencies required by the integration-test jobs but does not execute the tests. This will allow the "dependencies" entry for "compile-and-unit-test" to be shared among all cache entries for "integration-test". The resulting "integration-test" entries should be much smaller, reducing the potential for eviction.
- Combine the above 2 techniques, so that no cache entry is written by "integration-test" jobs, but all required dependencies are already present from the restored "compile-and-unit-test" entry.
### Select which branches should write to the cache
GitHub cache entries are not shared between builds on different branches.
Workflow runs can restore caches created in either the current branch or the default branch (usually main).
This means that each branch will have its own Gradle User Home cache scope, and will not benefit from cache entries written for other (non-default) branches.
By default, The `setup-gradle` action will only _write_ to the cache for builds run on the default (`master`/`main`) branch.
Jobs running on other branches will only read from the cache. In most cases, this is the desired behavior.
This is because Jobs running on other branches will benefit from the cached Gradle User Home from `main`,
without writing private cache entries which could lead to evicting these shared entries.
If you have other long-lived development branches that would benefit from writing to the cache,
you can configure this by disabling the `cache-read-only` action parameter for these branches.
See [Using the cache read-only](#using-the-cache-read-only) for more details.
Note there are some cases where writing cache entries is typically unhelpful (these are disabled by default):
- For `pull_request` triggered runs, the cache scope is limited to the merge ref (`refs/pull/.../merge`) and can only be restored by re-runs of the same pull request.
- For `merge_group` triggered runs, the cache scope is limited to a temporary branch with a special prefix created to validate pull request changes, and won't be available on subsequent Merge Queue executions.
### Exclude content from Gradle User Home cache
As well as any wrapper distributions, the action will attempt to save and restore the `caches` and `notifications` directories from Gradle User Home.
Each build is different, and some builds produce more Gradle User Home content than others.
[Cache debugging ](#cache-debugging-and-analysis) can provide insight into which cache entries are the largest,
and the contents to be cached can be fine-tuned by including and excluding certain paths within the Gradle User Home.
```yaml
# Cache downloaded JDKs in addition to the default directories.
gradle-home-cache-includes: |
caches
notifications
jdks
# Exclude the local build-cache and keyrings from the directories cached.
gradle-home-cache-excludes: |
caches/build-cache-1
caches/keyrings
```
You can specify any number of fixed paths or patterns to include or exclude.
File pattern support is documented at https://docs.github.com/en/actions/learn-github-actions/workflow-syntax-for-github-actions#patterns-to-match-file-paths.
### Remove unused files from Gradle User Home before saving to the cache
The Gradle User Home directory tends to grow over time. When you switch to a new Gradle wrapper version or upgrade a dependency version
the old files are not automatically and immediately removed. While this can make sense in a local environment, in a GitHub Actions environment
it can lead to ever-larger Gradle User Home cache entries being saved and restored.
To avoid this situation, The `setup-gradle` action supports the `gradle-home-cache-cleanup` parameter.
When enabled, this feature will attempt to delete any files in the Gradle User Home that were not used by Gradle during the GitHub Actions Workflow, before saving the Gradle User Home to the GitHub Actions cache.
Gradle Home cache cleanup is considered experimental and is disabled by default. You can enable this feature for the action as follows:
```yaml
gradle-home-cache-cleanup: true
```
## Debugging and Troubleshooting
To debug a failed job, it can be useful to run with [debug logging enabled](https://docs.github.com/en/actions/monitoring-and-troubleshooting-workflows/enabling-debug-logging).
You can enable debug logging either by:
1. Adding an `ACTIONS_STEP_DEBUG` variable to your repository configuration ([see here](https://docs.github.com/en/actions/monitoring-and-troubleshooting-workflows/enabling-debug-logging#enabling-step-debug-logging)).
2. By re-running a Job and checking the "Enable debug logging" box ([see here](https://github.blog/changelog/2022-05-24-github-actions-re-run-jobs-with-debug-logging/)).
### Increased logging from Gradle builds
When debug logging is enabled, this action will cause all builds to run with the `--info` and `--stacktrace` options.
This is done by inserting the relevant [Gradle properties](https://docs.gradle.org/current/userguide/build_environment.html#sec:gradle_configuration_properties)
at the top of the `${GRADLE_USER_HOME}/gradle.properties` file.
If the additional Gradle logging produced is problematic, you may opt out of this behavior by setting these properties manually in your project `gradle.properties` file:
```properties
# default lifecycle
org.gradle.logging.level=lifecycle
org.gradle.logging.stacktrace=internal
```
### Cache debugging and analysis
A report of all cache entries restored and saved is printed to the Job Summary when saving the cache entries.
This report can provide valuable insight into how much cache space is being used.
When debug logging is enabled, more detailed logging of cache operations is included in the GitHub actions log.
This includes a breakdown of the contents of the Gradle User Home directory, which may assist in cache optimization.
## Build reporting
The `setup-gradle` action collects information about any Gradle executions that occur in a workflow, including the root project,
requested tasks, build outcome, and any Build Scan link generated. Details of cache entries read and written are also collected.
These details are compiled into a Job Summary, which is visible in the GitHub Actions UI.
Generation of a Job Summary is enabled by default for all Jobs using The `setup-gradle` action. This feature can be configured
so that a Job Summary is never generated, or so that a Job Summary is only generated on build failure:
```yaml
add-job-summary: 'on-failure' # Valid values are 'always' (default), 'never', and 'on-failure'
```
### Adding Job Summary as a Pull Request comment
It is sometimes more convenient to view the results of a GitHub Actions Job directly from the Pull Request that triggered
the Job. For this purpose, you can configure the action so that Job Summary data is added as a Pull Request comment.
```yaml
name: CI
on:
pull_request:
permissions:
pull-requests: write
jobs:
run-gradle-build:
runs-on: ubuntu-latest
steps:
- name: Checkout project sources
- name: Checkout sources
uses: actions/checkout@v4
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
with:
add-job-summary-as-pr-comment: on-failure # Valid values are 'never' (default), 'always', and 'on-failure'
- run: ./gradlew build --scan
```
Note that to add a Pull Request comment, the workflow must be configured with the `pull-requests: write` permission.
### Build Scan® link as Step output
As well as reporting all [Build Scan](https://gradle.com/build-scans/) links in the Job Summary,
The `setup-gradle` action makes this link available as an output of any Step that executes Gradle.
The output name is `build-scan-url`. You can then use the build scan link in subsequent actions of your workflow.
### Saving arbitrary build outputs
By default, a GitHub Actions workflow using `setup-gradle` will record the log output and any Build Scan
links for your build, but any output files generated by the build will not be saved.
To save selected files from your build execution, you can use the core [Upload-Artifact](https://github.com/actions/upload-artifact) action.
For example:
```yaml
jobs:
gradle:
runs-on: ubuntu-latest
steps:
- name: Checkout project sources
uses: actions/checkout@v4
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
- name: Run build with Gradle wrapper
run: ./gradlew build --scan
- name: Upload build reports
uses: actions/upload-artifact@v3
if: always()
with:
name: build-reports
path: build/reports/
```
### Use of custom init-scripts in Gradle User Home
Note that the action collects information about Gradle invocations via an [Initialization Script](https://docs.gradle.org/current/userguide/init_scripts.html#sec:using_an_init_script)
located at `USER_HOME/.gradle/init.d/gradle-actions.build-result-capture.init.gradle`.
If you are adding any custom init scripts to the `USER_HOME/.gradle/init.d` directory, it may be necessary to ensure these files are applied before `gradle-actions.build-result-capture.init.gradle`.
Since Gradle applies init scripts in alphabetical order, one way to ensure this is via file naming.
## Support for GitHub Enterprise Server (GHES)
You can use the `setup-gradle` action on GitHub Enterprise Server, and benefit from the improved integration with Gradle. Depending on the version of GHES you are running, certain features may be limited:
- Build Scan links are captured and displayed in the GitHub Actions UI
- Easily run your build with different versions of Gradle
- Save/restore of Gradle User Home (requires GHES v3.5+ : GitHub Actions cache was introduced in GHES 3.5)
- Support for GitHub Actions Job Summary (requires GHES 3.6+ : GitHub Actions Job Summary support was introduced in GHES 3.6). In earlier versions of GHES, the build-results summary and caching report will be written to the workflow log, as part of the post-action step.
## GitHub Dependency Graph support
> [!IMPORTANT]
> The simplest (and recommended) way to generate a dependency graph is via a separate workflow
> using `gradle/actions/dependency-submission`. This action will attempt to detect all dependencies used by your build
> without building and testing the project itself.
>
> See the [dependency-submission documentation](../dependency-submission/README.md) for up-to-date documentation.
The `setup-gradle` action has support for submitting a [GitHub Dependency Graph](https://docs.github.com/en/code-security/supply-chain-security/understanding-your-software-supply-chain/about-the-dependency-graph) snapshot via the [GitHub Dependency Submission API](https://docs.github.com/en/rest/dependency-graph/dependency-submission?apiVersion=2022-11-28).
The dependency graph snapshot is generated via integration with the [GitHub Dependency Graph Gradle Plugin](https://plugins.gradle.org/plugin/org.gradle.github-dependency-graph-gradle-plugin) and saved as a workflow artifact. The generated snapshot files can be submitted either in the same job or in a subsequent job (in the same or a dependent workflow).
The generated dependency graph snapshot reports all of the dependencies that were resolved during a build execution, and is used by GitHub to generate [Dependabot Alerts](https://docs.github.com/en/code-security/dependabot/dependabot-alerts/about-dependabot-alerts) for vulnerable dependencies, as well as to populate the [Dependency Graph insights view](https://docs.github.com/en/code-security/supply-chain-security/understanding-your-software-supply-chain/exploring-the-dependencies-of-a-repository#viewing-the-dependency-graph).
### Basic usage
You enable GitHub Dependency Graph support by setting the `dependency-graph` action parameter. Valid values are:
| Option | Behaviour |
| --- | --- |
| `disabled` | Do not generate a dependency graph for any build invocations.<p>This is the default. |
| `generate` | Generate a dependency graph snapshot for each build invocation. |
| `generate-and-submit` | Generate a dependency graph snapshot for each build invocation, and submit these via the Dependency Submission API on completion of the job. |
| `generate-and-upload` | Generate a dependency graph snapshot for each build invocation, saving it as a workflow artifact. |
| `download-and-submit` | Download any previously saved dependency graph snapshots, and submit them via the Dependency Submission API. This can be useful to submit [dependency graphs for pull requests submitted from repository forks](#dependency-graphs-for-pull-request-workflows). |
Example of a CI workflow that generates and submits a dependency graph:
```yaml
name: CI build
on:
push:
permissions:
contents: write
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to generate and submit dependency graphs
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
- name: Run the usual CI build (dependency-graph will be generated and submitted post-job)
- name: Build with Gradle
run: ./gradlew build
```
The `contents: write` permission is required to submit (but not generate) the dependency graph file.
Depending on [repository settings](https://docs.github.com/en/actions/security-guides/automatic-token-authentication#permissions-for-the-github_token), this permission may be available by default or may need to be explicitly enabled in the workflow file (as above).
> [!IMPORTANT]
> The above configuration will work for workflows that run as a result of commits to a repository branch,
> but not when a workflow is triggered by a PR from a repository fork.
> This is because the `contents: write` permission is not available when executing a workflow
> for a PR submitted from a forked repository.
> For a configuration that supports this setup, see [Dependency Graphs for pull request workflows](#dependency-graphs-for-pull-request-workflows).
### Making dependency graph failures cause Job failures
By default, if a failure is encountered when generating or submitting the dependency graph, the action will log the failure as a warning and continue.
This allows your workflow to be resilient to dependency graph failures, in case dependency graph production is a side-effect rather than the primary purpose of a workflow.
If instead, you have a workflow whose primary purpose is to generate and submit a dependency graph, it makes sense for this workflow to fail if the dependency
graph cannot be generated or submitted. You can enable this behavior with the `dependency-graph-continue-on-failure` parameter, which defaults to `true`.
```yaml
# Ensure that the workflow Job will fail if the dependency graph cannot be submitted
- uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
dependency-graph-continue-on-failure: false
```
### Using a custom plugin repository
By default, the action downloads the `github-dependency-graph-gradle-plugin` from the Gradle Plugin Portal (https://plugins.gradle.org). If your GitHub Actions environment does not have access to this URL, you can specify a custom plugin repository to use.
Do so by setting the `GRADLE_PLUGIN_REPOSITORY_URL` environment variable with your Gradle invocation.
The `GRADLE_PLUGIN_REPOSITORY_USERNAME` and `GRADLE_PLUGIN_REPOSITORY_PASSWORD` can be used when the plugin repository requires authentication.
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to generate and submit dependency graphs
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
- name: Run a build, resolving the 'dependency-graph' plugin from the plugin portal proxy
run: ./gradlew build
env:
GRADLE_PLUGIN_REPOSITORY_URL: "https://gradle-plugins-proxy.mycorp.com"
# Set the following variables if your custom plugin repository requires authentication
# GRADLE_PLUGIN_REPOSITORY_USERNAME: "username"
# GRADLE_PLUGIN_REPOSITORY_PASSWORD: ${secrets.MY_REPOSITORY_PASSWORD}
```
### Choosing which Gradle invocations will generate a dependency graph
Once you enable the dependency graph support for a workflow job (via the `dependency-graph` parameter), dependencies will be collected and reported for all subsequent Gradle invocations.
If you have a Gradle build step that you want to exclude from dependency graph generation, you can set the `GITHUB_DEPENDENCY_GRAPH_ENABLED` environment variable to `false`.
```yaml
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to generate and submit dependency graphs
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate-and-submit
- name: Build the app, generating a graph of dependencies required
run: ./gradlew :my-app:assemble
- name: Run all checks, disabling dependency graph generation
run: ./gradlew check
env:
GITHUB_DEPENDENCY_GRAPH_ENABLED: false
```
### Filtering which Gradle Configurations contribute to the dependency graph
If you do not want the dependency graph to include every dependency configuration in every project in your build,
you can limit the dependency extraction to a subset of these.
See the documentation for [dependency-submission](../dependency-submission/README.md) and the
[GitHub Dependency Graph Gradle Plugin](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#filtering-which-gradle-configurations-contribute-to-the-dependency-graph) for details.
### Gradle version compatibility
Dependency-graph generation is compatible with most versions of Gradle >= `5.2`, and is tested regularly against
Gradle versions `5.2.1`, `5.6.4`, `6.0.1`, `6.9.4`, `7.1.1` and `7.6.3`, as well as all patched versions of Gradle 8.x.
A known exception to this is that Gradle `7.0`, `7.0.1`, and `7.0.2` are not supported.
See [here](https://github.com/gradle/github-dependency-graph-gradle-plugin?tab=readme-ov-file#gradle-compatibility) for complete compatibility information.
### Reducing storage costs for saved dependency graph artifacts
When `generate` or `generate-and-submit` is used with the action, the dependency graph that is generated is stored as a workflow artifact.
By default, these artifacts are retained for 30 days (or as configured for the repository).
To reduce storage costs for these artifacts, you can set the `artifact-retention-days` value to a lower number.
```yaml
steps:
- name: Generate dependency graph, but only retain artifact for one day
uses: gradle/actions/setup-gradle@v3
with:
dependency-graph: generate
artifact-retention-days: 1
```
# Develocity plugin injection
The `setup-gradle` action provides support for injecting and configuring the Develocity Gradle plugin into any Gradle build, without any modification to the project sources.
This is achieved via an init-script installed into Gradle User Home, which is enabled and parameterized via environment variables.
The same auto-injection behavior is available for the Common Custom User Data Gradle plugin, which enriches any build scans published with additional useful information.
## Enabling Develocity injection
To enable Develocity injection for your build, you must provide the required configuration via environment variables.
Here's a minimal example:
```yaml
name: Run build with Develocity injection
env:
DEVELOCITY_INJECTION_ENABLED: true
DEVELOCITY_URL: https://develocity.your-server.com
DEVELOCITY_PLUGIN_VERSION: 3.17
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle
uses: gradle/actions/setup-gradle@v3
- name: Run a Gradle build with Develocity injection enabled
run: ./gradlew build
```
This configuration will automatically apply `v3.17` of the [Develocity Gradle plugin](https://docs.gradle.com/develocity/gradle-plugin/), and publish build scans to https://develocity.your-server.com.
This example assumes that the `develocity.your-server.com` server allows anonymous publishing of build scans.
In the likely scenario that your Develocity server requires authentication, you will also need to configure an additional environment variable
with a valid [Develocity access key](https://docs.gradle.com/develocity/gradle-plugin/#via_environment_variable).
## Configuring Develocity injection
The `init-script` supports several additional configuration parameters that you may find useful. All configuration options (required and optional) are detailed below:
| Variable | Required | Description |
|--------------------------------------| --- |-------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| DEVELOCITY_INJECTION_ENABLED | :white_check_mark: | enables Develocity injection |
| DEVELOCITY_URL | :white_check_mark: | the URL of the Develocity server |
| DEVELOCITY_ALLOW_UNTRUSTED_SERVER | | allow communication with an untrusted server; set to _true_ if your Develocity instance is using a self-signed certificate |
| DEVELOCITY_CAPTURE_FILE_FINGERPRINTS | | enables capturing the paths and content hashes of each individual input file |
| DEVELOCITY_ENFORCE_URL | | enforce the configured Develocity URL over a URL configured in the project's build; set to _true_ to enforce publication of build scans to the configured Develocity URL |
| DEVELOCITY_PLUGIN_VERSION | :white_check_mark: | the version of the [Develocity Gradle plugin](https://docs.gradle.com/develocity/gradle-plugin/) to apply |
| DEVELOCITY_CCUD_PLUGIN_VERSION | | the version of the [Common Custom User Data Gradle plugin](https://github.com/gradle/common-custom-user-data-gradle-plugin) to apply, if any |
| GRADLE_PLUGIN_REPOSITORY_URL | | the URL of the repository to use when resolving the Develocity and CCUD plugins; the Gradle Plugin Portal is used by default |
| GRADLE_PLUGIN_REPOSITORY_USERNAME | | the username for the repository URL to use when resolving the Develocity and CCUD plugins |
| GRADLE_PLUGIN_REPOSITORY_PASSWORD | | the password for the repository URL to use when resolving the Develocity and CCUD plugins; Consider using secrets to pass the value to this variable |
## Publishing to scans.gradle.com
Develocity injection is designed to enable the publishing of build scans to a Develocity instance,
but is also useful for publishing to the public Build Scans instance (https://scans.gradle.com).
To publish to https://scans.gradle.com, you must specify in your workflow that you accept the [Gradle Terms of Use](https://gradle.com/help/legal-terms-of-use).
```yaml
name: Run build and publish Build Scan
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Setup Gradle to publish build scans
uses: gradle/actions/setup-gradle@v3
with:
build-scan-publish: true
build-scan-terms-of-use-url: "https://gradle.com/terms-of-service"
build-scan-terms-of-use-agree: "yes"
- name: Run a Gradle build - a build scan will be published automatically
run: ./gradlew build
```
See the [full action documentation](../docs/setup-gradle.md) for more advanced usage scenarios.