JaCoCo VS proguard-core

I will use here JaCoCo, where also the JaCoCo-Maven-lugin exists for the usage in your Maven builds. This article will show how to configure the code coverage to finally get the results for unit- and integration-tests.

In protection rules, I added build workflow in Require status checks to pass before merging. This is to ensure that before merging code in master branch, build should run successfully. I also added Jacoco Code Coverage to make sure that enough unit tests are available in project and Detekt to make sure that code in project is readable. I added them in build configuration. Even if one of them gives error, build will fail. Whenever, someone push code in pull request, build action will run and check if build is running successfully or not.

Code coverage analysis tools quantify the amount of tested code, serving as a valuable tool to inform on code structure and testing related decisions. We will make use of JaCoCo, JaCoCo produces reports on multiple kinds of code coverage metrics including instructions, line and branch coverage.

One specific example where I regularly use a profile in this way is for configuring code coverage. In all of my Java projects, I use JaCoCo for generating code coverage reports. I use JaCoCo during the Maven test phase. However, while developing I find it useful at times to exclude coverage reporting to reduce the build time. But in my CI/CD workflows in GitHub Actions, I activate the code coverage profile during pull-requests and pushes to the default branch. For pull-requests, my GitHub Actions workflow comments the code coverage on the PR and uploads the coverage report as a workflow artifact, where I can inspect it as necessary. And during a push to the default branch, my workflow updates coverage badges to keep them up to date with the current state of the default branch. I can also activate the code coverage profile locally while developing, such as prior to submitting a pull-request, to ensure that I didn't miss testing something.

In Unit tests, individual software code components are checked if it is working as expected or not. Unit tests isolate a function or module of code and verify its correctness. We can use tools like JaCoCo for Java and Mocha, and Jasmine for NodeJS to generate unit test reports. We can also send these reports to SonarQube which shows us code coverage and the percentage of your code covered by your test cases.

And there is the original jacoco/jacoco: (0.8.7: released this on May 5, 2021), but it's for Java. I'm not sure if we can use it with multiple flavors on Android.

Make sure to also update to Jacoco 0.8.7 to avoid test issues: https://github.com/jacoco/jacoco/releases/tag/v0.8.7

Anyways, have you looked at Proguard core?

For generating Java bytecode/class files you could take a look at ProGuardCORE . It's a bytecode manipulation and analysis library that grew out of ProGuard.

The MethodBuilder "low-level" example looks like the ProGuardCORE API for building classes & code sequences, which can be written as:

Something else that could make your code generation for your JVM language easier: ProGuardCORE (https://github.com/Guardsquare/proguard-core). It can be used to read, generate and analyse Java bytecode.

Some examples for code generation: ProGuard where the project originated (https://github.com/Guardsquare/proguard), Brainf*ck compiler (https://github.com/mrjameshamilton/bf), Lox compiler (https://github.com/mrjameshamilton/klox)

Disclaimer: I work at Guardsquare on ProGuardCORE so may be biased ;)

I implemented Lox in Kotlin based on the book's Java interpreter then added a JVM backend using ProGuardCORE: https://github.com/mrjameshamilton/klox

You can find out more about ProGuardCORE on GitHub and the manual contains some further examples.

Not a book or guide but a useful library for working with/generating JVM class files is ProGuardCORE. It's the underlying library used by ProGuard and I also recently used it to add a JVM backend to my implementation of Lox from the Crafting Interpreters book.

After completing the first part of the book (using Kotlin instead of Java), I added a JVM backend using ProGuardCORE (https://github.com/Guardsquare/proguard-core).

The biggest difference is our initial goal of creating a tool for application developers rather than a security team. This means we aim to exclude information that the application developer definitely knows and doesn't want to see again, e.g. the list of permissions an application is requesting. Furthermore we try to focus on nice and intuitive UX that is familiar to this audience, enabling an application developer to be very efficient in reading & interpreting the results (e.g. easy comparison of two builds). Also note that this is built on the same core technology as ProGuard. This foundation of compiler components (e.g. our partial evaluator) and the knowledge and experience at Guardsquare that comes with it will unlock many more in-depth code checks.