dotnet-probability-space-calculation

Visualisation using WPF dynamically generated heatmap.

Overview

Exploratory research project enabling setup of various automated CI/CD components and services.

Problem domain used is the calculation of a Bivariate Poisson 2d probability space. The focus of this exercise is more to understand options, dependencies and complexity of automating C#/.Net build pipelines than the domain itself.

Stack

• .NetCore 2.2
• C# 7
• Math.net Numerics
• Coverlet (C# coverage testing)
• NUnit

Components under consideration / evaluation

• Cake
• Travis CI
• Jenkins
• Code Climate
• nUnit / xUnit / MSTest

Running Notes

Example running Cake file created / included in root i.e.

./build.ps1

To execute from the command line

dotnet restore
dotnet test ./PredictiveModellingTest/PredictiveModellingTest.csproj 
    /p:CollectCoverage=true /p:CoverletOutputFormat=cobertura 
	/p:CoverletOutput=./PredictiveModellingTest/cobertura.xml

Status / Summary

• Travis CI automated build / badge status update [working].
• Jenkins automated build using Cake build script [working].
• Code Climate integration (code analysis) [working].
• Code Climate coverage reporting [broken - suspected upstream / platform issue].

Follow-on (Future possibly private dev't)

• Explore Azure Devops (https://azure.microsoft.com/en-gb/product-categories/devops/)
• Explore AppVeyor (https://www.appveyor.com/docs/)
• Explore use of Sonar Cloud (https://sonarcloud.io/about)
• Explore integration of selenium / appium testing, integrated with Cake / Travis.
• Extend / enhance WPF viewer to enable dynamic update from NATS / RabbitMQ / SQS etc.
• Alternative visualations showing realtime outcome distribution.