Versions Compared

Old Version 4

changes.mady.by.user Tim Donohue

Saved on

compared with

New Version 5

changes.mady.by.user Tim Donohue

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Testing DSpace 7 Pull Requests - How to use existing Docker scripts to spin up PRs more easily locally, in order to test them or review them.
  • Spin up code in virtual environment (quickly) for easier reviews/testing
    • GitHub CodeSpaces: https://docs.github.com/en/codespaces/overview
      • Tim Donohue has played with this, and it's possible to spin up DSpace 7 in CodeSpaces.  However, you need to use a 4 Core / 8GB RAM machine type (otherwise, the UI will often fail to build with an "error 137" - meaning it ran out of memory during the build process).  Connecting the UI to a running Backend is also not super-easy right now, as codespaces assigns a random URL.  We'd need to likely provide a Codespaces configuration for DSpace to make this easier.
    • GitPod: https://www.gitpod.io/for/opensource 
  • Automated Code Reviewing resources.  Tools/resources exist which can do some automatic checking/verification of code quality in Pull Requests.  Some examples include:
    • Code Scanning in GitHub.  We already do some of this, but it is more currently we only scan for security-issue oriented.oriented code issues.
      • Free & integrated into GitHub.  Interface is a bit clunky at times though.
      • Highly Configurable (e.g. see query types)
      • Could configure this to also check PR code quality against coding best practices (currently we only scan for major bugs / security issues).  See these settings
    • SonarCloud.io - This is a hosted version of SonarQube.
      • Free for open source projects. Integrates with GitHub & supports both Java and TypeScript. Can run on every new PR. 
      • Example projects: https://sonarcloud.io/explore/projects
      • Test analysis run by Tim on DSpace backend/frontend: https://sonarcloud.io/organizations/tdonohue/projects  (Keep in mind, there are definitely false positives listed here. These are just raw reports)
      • Pros: HIGHLY CONFIGURABLE. Highly Configurable Used by other major OS projects like Apache.  Good documentation/resources on how to fix any issues that are found. SonarQube is open source itself.
    • DeepSource
      • Free for open source projects. Integrates with GitHub & supports both Java and TypeScript. Can run on every new PR. 
      • Test analysis run by Tim on DSpace backend/frontend (Keep in mind, there are definitely false positives listed in both. These are just raw reports)
      • Not as configurable, but able to turn off individual rules if they are too "noisy" or not useful.
      • Pros: Some "autofix" options. Good documentation/resources on how to fix any issues that are found.

...