Break up large spec files

[dacook]

What? Why?

Knapsack evenly distributes spec files to run in parallel, but can't break up spec files to run in parallel.
I noticed that a couple of large spec files were much longer than others, and therefore causing a bottleneck.

Also, I think we can change one of the runners to run more system specs, because the models are nice and quick.

What should we test?

• green specs
• quicker runtime
• review the 'Usage' list to see individual runtimes (scroll to bottom left of Actions summary).

Release notes

Changelog Category (reviewers may add a label for the release notes):

• User facing changes
• API changes (V0, V1, DFC or Webhook)
• Technical changes only
• Feature toggled

The title of the pull request will be included in the release notes.

Dependencies

Documentation updates

[dacook]

One system_admin spec was still 7m, but maybe after this is merged, Knapsack will have better data to more evenly distribute specs.

---

Excerpt from logs shows that spring is used for both rake and rspec commands. I couldn't see timestamps to prove the time saving.

Run bin/rake db:create db:schema:load

Heads up! 🔥

👉 StimulusReflex requires caching to be enabled. Caching allows the session to be modified during ActionCable requests.

To enable caching in development, run:

  rails dev:cache

Running via Spring preloader in process 2883
Database 'open_food_network_test' already exists

Run bin/rake knapsack_pro:queue:rspec
Running via Spring preloader in process 2891

[mkllnk]

Thank you! This is great.

[dacook]

Thanks, I just had a look at the latest master build, and it was done in 6m 19s.
The longest runner was 6m 9s, executing only one file spec/system/admin/order_spec.rb.
Second longest was 5m 24s, for spec/system/admin/subscriptions/crud_spec.rb.

So those are the next bottlenecks to attack..

[dacook]

FYI @filipefurtad0 I made a few tweaks to CI here ^