Orb-level parameters

Ecosystem Orbs
1

Are there ways to set a parameter that affects the behavior of the entire orb?

For example, let’s say I have 5 workflow steps that all override the cache-key with the same value. I would have to set that parameter on each of those, right?

Additionally, it might be nice to invoke an orb once with a top level param that influences the behavior of other commands. For example, a node orb, being able to set whether npm or yarn is the package manager might be a better way of influencing defaults (see, e.g., discussion in https://github.com/CircleCI-Public/circleci-orbs/pull/85)

2 Likes
2

@wyardley This is good food for thought!

I know that the team is currently sketching out ideas for some kind of top-level concept of orb parameters—i.e., defining parameters for an entire orb, which would then filter down and/or influence the behavior of individual parameters in commands/jobs.

Good to know that we’re on the right track :slight_smile:

1 Like
3

Glad to see this issue already raised!

I would like to submit my use-case for this feature.

I have some config like this:

  python: dialogue/sandbox@dev:python
  app: dialogue/sandbox@dev:lambda-python

workflows:
  main:
    jobs:
      - python/pylama:
          name: pylama
          executor:
            name: python/python3_6
            version: "3.6"
      - python/isort:
          name: isort
          executor:
            name: python/python
            version: "3.6"
          context: org-global-v2
      - python/black:
          name: black
          executor:
            name: python/python
            version: "3.6"
          context: org-global-v2
      - app/deploy:
          name: deploy
          context: org-global-v2
          executor:
            name: python/python
            version: "3.6"

But I wish I could something like this:

  python: dialogue/sandbox@dev:python
  app:
    from: dialogue/sandbox@dev:lambda-python
    python_version: 3.6

workflows:
  main:
    jobs:
      - python/pylama:
          name: pylama
      - python/isort:
          name: isort
          context: org-global-v2
      - python/black:
          name: black
          context: org-global-v2
      - app/deploy:
          name: deploy
          context: org-global-v2

Also… while I’m here, although off-topic, I really wish CCI would apply a default context and/or allow something like:

workflows:
  default_context: org-global-v2
  main:
    jobs:
      - python/pylama:
          name: pylama
      - python/isort:
          name: isort
      - python/black:
          name: black
      - app/deploy:
          name: deploy
1 Like
4

Has there been any movement on this? I would also love to have orb-level parameters, particularly for version parameters that should be consistent across commands.

5

Hi there! Is some progress on this topic? It may be useful for those developers who work on huge projects with big amounts of repos