Skip to content

A new perspective

We work issue branches defined as follows:

  • Must come off of main
  • Must have the issue number it relates to as the first part of the name followed by a dash and then the title of the issue (whitespaces replaced by underscores)
gitGraph
commit id: '34'
commit id: '56'
branch "32-issue"
checkout "32-issue"
commit id: 'a'

When an issue is worked on

  • An existing branch that matches the definition of an issue branch must be reused if it exists either locally or remote
  • or else it is created
  • Must be registered with the issue as a development branch (gh issue develop ...) (implies being made made available to origin immediately upon creation)
  • Must have it’s upstream set to a branch with the same name in origin

If the repo is connected to a kanban board:

  • The issue must be added to the project if it isn’t already
  • The issue must be updated to status ‘work in progress’
  • The person who began working on the issue must be assigned to the issue

During development the developer can share the branch with co-workers (workon, pull, push) and create as many commits as desired

gitGraph
commit id: '34'
commit id: '56'
branch "32-issue"
checkout "32-issue"
commit id: 'a'
commit id: 'b'
commit id: 'c'