leakcanary/docs/dev-env.md

Dev Environment for LeakCanary contributors

Setup

• Download Android Studio.
• We use two spaces code indentation, use SquareAndroid code style settings from https://github.com/square/java-code-styles.
• Build with ./gradlew build.
• Running the failing UI tests to confirm leak detection correctly fails UI tests: ./gradlew leakcanary-android-sample:connectedCheck.
• Normal UI tests: ./gradlew leakcanary-android-core:connectedCheck.

Static Code Analysis

• LeakCanary uses Detekt for static Code analysis.
• Analyze the entire project with ./gradlew check or particular modules with ./gradlew :module-name:check. Detekt will fail the build if any ruleset violations are found. You should fix all issues before pushing the branch to remote.
  • There's also a git pre-push hook that will run analysis automatically before pushing a branch to the remote. If there are any violations - it will prevent the push. Fix the issues!
  • You can bypass the git hook though; Travis CI will still run checks and will fail if any violations are found.
• Detekt report will be printed in the console and saved to /moduleDir/build/reports/.

Unit Testing with Synthetic Heap Dumps

When testing heap analysis functionality, create synthetic heap dumps programmatically instead of committing binary .hprof files:

val heapDump = dump {
  "com.example.MyClass" watchedInstance {
    field["fieldName"] = BooleanHolder(true)
  }
}

• Requires testImplementation(projects.shark.sharkHprofTest).
• See LeakStatusTest or AndroidObjectInspectorsTest for examples.

Deploying locally

To deploy LeakCanary to your local maven repository, run the following command, changing the path to the path of your local repository:

./gradlew uploadArchives -PSNAPSHOT_REPOSITORY_URL=file:///Users/py/.m2/repository

Then add the SNAPSHOT dependency and mavenLocal() repository to your project:

dependencies {
  debugImplementation 'com.squareup.leakcanary:leakcanary-android:{{ leak_canary.next_release }}-SNAPSHOT'
}

repositories {
  mavenLocal()
}

Deploying the docs locally

Installing or updating the docs dependencies:

pip install --requirement docs/requirements.txt

Deploying locally

mkdocs serve