hull workspace plan

Synopsis

hull workspace plan reads hull-workspace.yaml and prints the order in which hull workspace install and upgrade will process the members. Nothing is applied and the cluster is not contacted — it is a preview of the dependency ordering. With --levels it groups the members by depth, so you can see which ones are eligible to run in parallel.

When to use it

  • Before an install or upgrade, to confirm the dependency order is what you expect — especially after editing a dependsOn list.
  • To see where parallelism kicks in: any level with more than one member can run those members concurrently under --parallel.
  • To catch a dependency cycle early — a cycle stops the command with an error naming the members involved.

What happens

  1. Reads hull-workspace.yaml from --dir (default .).
  2. Builds the dependency graph from every member’s dependsOn list.
  3. Sorts the members so each one comes after everything it depends on.
  4. Prints the order: a numbered flat list by default, or one group per level with --levels.

Usage 

hull workspace plan [flags]

Flags

Flag Type Default Description
--dir string . Directory containing hull-workspace.yaml. Point it elsewhere to plan a workspace in another directory.
--levels bool false Group the output by dependency depth instead of a flat list, so members that can run in parallel appear together.

Inherits the global flags.

Worked example

INPUT — hull-workspace.yaml with two members, where api depends on postgres:

apiVersion: hull/v1
defaults:
  namespace: apps
members:
  - name: postgres
    path: ./postgres
  - name: api
    path: ./api
    dependsOn: [postgres]

Run it:

hull workspace plan

OUTPUT:

1. postgres (path=./postgres, ns=apps, profile=)
2. api (path=./api, ns=apps, profile=)

postgres is listed first because api declares dependsOn: [postgres], so it must come up before api. Both inherit ns=apps from defaults.namespace, and profile= is empty because neither member sets one.

Group the same two members by level:

hull workspace plan --levels

OUTPUT:

level 0 (1 members, parallelisable):
  - postgres (path=./postgres, ns=apps)
level 1 (1 members, parallelisable):
  - api (path=./api, ns=apps)

postgres has no dependencies, so it sits alone at level 0; api waits for level 0 to finish, so it lands at level 1. Add a second dependency-free package and it would join postgres at level 0 and run alongside it under --parallel.

See also