Product Tour

Owned by Mark Moloney
Jul 12, 2020
4 min read

DEVsheds is embedded in Bitbucket. Installation instructions are included in the readme of the test Bitbucket repository provided. This repository has the app preinstalled.

Location of Test Bitbucket Repository with preinstalled app - https://bitbucket.org/markm01/test-repo

  1. From the repository sidebar menu click on the "DEVsheds" link, which will take you to the Launchpad page.

     

  2. First build your "shed" (Development Workspace) by clicking on the "Build Shed" primary action button in the top navigation bar.

     

  3. This opens a window to show progress. It is building a custom Development Workspace (container) based on the source code in the repository.

     

  4. When this is done, the first two tool selections will be enabled.

     

  5. Start by clicking on the Data Science card. This opens a JupyterLab environment for working with data science notebooks. You can open the existing notebook "My Notebook.ipynb" and edit the contents. Code cells allow you to write and execute Python code. A range of data science and machine learning packages, such as Numpy, Pandas, and Scikit-learn, have been preinstalled.

     

  6. Next, we'll switch over to the Software Engineering View. In the top-left of the navigation bar is an App Switcher, which you can use to select "Code" or go back to the Launchpad. Clicking on "Code" opens up Theia - an integrated software development environment based on the popular VS Code.

     

  7. Within Theia, you can edit files, open the Git view to maintain the version history and commit changes, and open a command line terminal.

     

  8. For the next step in the tour, we need to come back to the Launchpad. From the Launchpad, click on the "Deploy System" button in the top navigation bar. This opens a progress window. In the background, the system is being built, tested, and deployed to Kubernetes to be able to see the running application. This may take a while depending on the complexity of the application.

     

  9. Once done, the "Services" card is enabled. Click on this to enter the DevOps/Services View.

     

  10. The first view you will see is the set of modules and tests contained within the source repository and their status.

     

  11. Clicking on the "Stack Graph" menu item in the sidebar will open a dependency view of the modules. Clicking on a module will display more detail about its state such as whether all tests have passed.

     

  12. You will notice links that have been added to the top navigation bar. These are links to the live application that has been deployed into a Kubernetes cluster for the purpose of seeing the live application as code is developed. Any code change results in the automatic retesting and update of the running application to close the feedback loop between development and user experience.

     

  13. Close any popup window and return to the Launchpad. If you have made any code changes, these have been automatically committed to a feature branch that was created when you first built your shed. You have the option of switching to the "Branches" view of Bitbucket to create a pull request from this feature branch. Alternatively, from the DEVsheds Launchpad, you can click on the "Merge & close" button to enter a commit message, which will squash your feature branch commits and merge back to the master branch. This will also shutdown your Development Workspace and free up computing capacity for other users.

     

  14. Alternatively, when you are done, click on the "Free up shed" button to shutdown your Development Workspace. (This will be done automatically after 30 minutes of inactivity, but it's good practice to free up computing capacity when you are done with your session.)

 

This concludes a brief walkthrough of key functionality and thank you for your attention.