My website is brand new and one of the few requirements of testings I had to implement was code coverage. Never though it was so hard to implement using a #docker container. Given my lack of experience, every attempt I tried on making a simple code coverage test using the 4 combinations of #TravisCI, #CircleCi with #Coveralls, #Codecov I failed. The main problem was I was generating the .coverage file within the docker container and couldn't access it with #TravisCi or #CircleCi, every attempt to solve this problem seems to be very hacky and this was not the kind of complexity I want to introduce to my newborn website. This problem was solved using a specific action for #GitHubActions, it was a 3 line solution I had to put in my github workflow file and I was able to access the .coverage file from my docker container and get the coverage report with #Codecov.

As we all know testing is an important part of any application. To assist with our testing we are going to use both Cypress and Jest. We feel these tools complement each other and will help us get good coverage of our code. We will use Cypress for our end to end testing as we've found it quite user friendly. Jest will be used for our unit tests because we've seen how many larger companies use it with great success.

Postman will be used to do integration testing with the backend API we create. It offers a clean interface to create many requests, and you can even organize these requests into collections. It helps to test the backend API first to make sure it's working before using it in the front-end. Jest can also be used for testing and is already embedded into React. Not only does it offer unit testing support in javascript, it can also do snapshot testing for the front-end to make sure components are rendering correctly. Enzyme is complementary to Jest and offers more functions such as shallow rendering. UnitTest will be used for Python testing as it is simple, has a lot of functionality and already built in with python. Sentry will be used for keeping track of errors as it is also easily integratable with Heroku because they offer it as an add-on. LogDNA will be used for tracking logs which are not errors and is also a Heroku add-on. Its good to have a separate service to record logs, monitor, track and even fix errors in real-time so our application can run more smoothly.

We were long time users of TravisCI, but switched to CircleCI because of the better user interface and pricing. Version 2.0 has had a couple of trips and hiccups; but overall we've been very happy with the continuous integration it provides. Continuous Integration is a must-have for building software, and CircleCI continues to surprise as they roll out ideas and features. It's leading the industry in terms of innovation and new ideas, and it's exciting to see what new things they keep rolling out.

Jenkins is a pretty flexible, complete tool. Especially I love the possibility to configure jobs as a code with Jenkins pipelines.

CircleCI is well suited for small projects where the main task is to run continuous integration as quickly as possible. Travis CI is recommended primarily for open-source projects that need to be tested in different environments.

And for something a bit larger I prefer to use Jenkins because it is possible to make serious system configuration thereby different plugins. In Jenkins, I can change almost anything. But if you want to start the CI chain as soon as possible, Jenkins may not be the right choice.