Save Energy

Reduce your carbon footprint by saving energy on GitHub.

How we can save energy on GitHub

Consider sustainability by saving energy for each inactive GitHub repository.

• Reduce the interval for automatically rebuilding the project or entirely remove regular automatic builds, e.g. for non-production code.
• Replace the trigger on every build by a trigger on pull request to main. Add a trigger to allow building manually on demand.
• Actively avoid running a CI pipeline if no relevant change was made, see the Skipping workflow runs article
• Reduce the frequency of dependabot checks or deactivate dependabot.
• Limit the retention duration for uploaded artifacts to the minimum.
• If you don't work on the code any longer, then download the repository to an archive on your local machine, archive it on GitHub or delete from GitHub.

Vision

The Save Energy application makes it easy to check which of your GitHub repositories have been inactive for a long time. It provides you with a list of repositories that you might want to consider archiving or deleting to save energy.

This shows the vision for a minimum viable product:

Development and Support Status

I am developing during my spare time and use this project for learning purposes. Please assume that I will need some days to answer your questions. At some point I might lose interest in the project. Please keep this in mind when using this project in a production environment.

Building the Project

To compile, test and run this project the latest LTS release of the .NET SDK is required on your machine.

Run the following commands from the folder containing the .sln file in order to build the project:

dotnet build

Running the Application

This application uses the GitHub API to retrieve information about repositories. You need to configure this mechanism of authentication as described in the Authentication Setup file.

The application must be run from a folder containing the appsettings.json file. You can either run it from the src/SaveEnergy folder or from the build output folder.

cd src/SaveEnergy; \
DOTNET_ENVIRONMENT=Development dotnet run --project SaveEnergy.csproj; \
cd ../..

Running from inside the src/SaveEnergy directory makes sure that the application reads the configuration file linked in the project file.

The DOTNET_ENVIRONMENT=Development parameter is required for the application to receive the user secrets.

Running the Tests

Test Overview

The project contains unit tests in the ./src/SaveEnergy.Tests folder and integration tests in the ./src/SaveEnergy.Specs folder.

Prerequisites: Coverlet, ReportGenerator

If you would like to build HTML reports from the test coverage, install the ReportGenerator tool:

dotnet tool install --global dotnet-reportgenerator-globaltool

Build, Test, Create Coverage Report

Run the following commands from the folder containing the .sln file in order to build and test.

# Build the project
dotnet build

# Run the tests once
dotnet test

To create a coverage report, run the following command:

# Cleanup test results from previous runs
rm -r ./src/SaveEnergy.Tests/TestResults ./src/SaveEnergy.Specs/TestResults

# Run the tests and create a coverage report in Cobertura format
dotnet test --no-restore --verbosity normal /p:CollectCoverage=true /p:CoverletOutputFormat=cobertura /p:CoverletOutput='./TestResults/coverage.cobertura.xml'

# Convert the Cobertura report to LCOV and HTML
reportgenerator "-reports:src/SaveEnergy.Tests/TestResults/*.xml;src/SaveEnergy.Specs/TestResults/*.xml" "-targetdir:report" "-reporttypes:Html;lcov" "-title:Save Energy"

# Open the HTML report in the browser
open report/index.html

Relevant GitHub API Endpoints

./docs/README.md shows how you can run a collection of GitHub API requests relevant to the project.

Acknowledgements

This project is inspired by Uwe Friedrichsen's talk "Patterns of Sustainability – Going Green in IT" held at the OOP Conference in 2023.

Many thanks to JetBrains who provide an Open Source License for this project ❤️.

Before Creating a Pull Request

Fix Static Code Analysis Warnings

Fix static code analysis warnings reported by SonarLint.

Apply Code Formatting Rules

# Install https://csharpier.io globally, once
dotnet tool install -g csharpier

# Format code
dotnet csharpier .

Remove temporary and generated files

Temporary and generated files create false positives during static code analysis.

# Remove temporary and generated files
# 1. dry run
git clean -ndx

# 2. Remove the files shown by the dry run
git clean -fdx

Check Code Metrics

Check code metrics using metrix++

• Configure the location of the cloned metrix++ scripts

  export METRIXPP=/path/to/metrixplusplus

• Collect metrics

  python "$METRIXPP/metrix++.py" collect --std.code.complexity.cyclomatic --std.code.lines.code --std.code.todo.comments --std.code.maintindex.simple -- .

• Get an overview

  python "$METRIXPP/metrix++.py" view --db-file=./metrixpp.db

• Apply thresholds

  python "$METRIXPP/metrix++.py" limit --db-file=./metrixpp.db --max-limit=std.code.complexity:cyclomatic:5 --max-limit=std.code.lines:code:25:function --max-limit=std.code.todo:comments:0 --max-limit=std.code.mi:simple:1

At the time of writing, I want to stay below the following thresholds:

--max-limit=std.code.complexity:cyclomatic:5
--max-limit=std.code.lines:code:25:function
--max-limit=std.code.todo:comments:0
--max-limit=std.code.mi:simple:1

Remove Code Duplication Where Appropriate

To detect duplicates I use the CPD Copy Paste Detector tool from the PMD Source Code Analyzer Project.

If you have installed PMD by download & unzip, replace pmd by ./run.sh. The homebrew pmd formula makes the pmd command globally available.

# Identify duplicated code in files to push to GitHub
pmd cpd --minimum-tokens 50 --language cs --dir .

References

• GitHub REST API
• GitHub: Skipping workflow runs