[Rails 6.0] Fix order adjustments spec

[andrewpbrett]

What? Why?

Paging @Matt-Yorkley to take a look at this... I'm a bit puzzled as to what would have changed in Rails 6 to flip this spec to red. Let's see if the fix works on the current master.

What should we test?

Green build

Release notes

Updated an Order spec to prepare for Rails 6

Changelog Category: Technical changes

Dependencies

Documentation updates

[codecov]

Codecov Report

Merging #7589 (231f01d) into master (2910cbf) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master    #7589   +/-   ##
=======================================
  Coverage   93.20%   93.20%           
=======================================
  Files         635      635           
  Lines       18136    18136           
=======================================
  Hits        16903    16903           
  Misses       1233     1233           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2910cbf...231f01d. Read the comment docs.

[luisramos0]

Could it be related to autoloading?
https://guides.rubyonrails.org/upgrading_ruby_on_rails.html#autoloading

[Matt-Yorkley]

What was the error? It's difficult to review without the context.

[Matt-Yorkley]

We could maybe just do order.all_adjustments.each(&:finalize!) here for simplicity? Not a big deal though.

[andrewpbrett]

@Matt-Yorkley you can see it in this build: https://github.com/openfoodfoundation/openfoodnetwork/runs/2536112257?check_suite_focus=true

2) Spree::Order a completed order with shipping and transaction fees removing line_items when finalized fee adjustments exist on the order does not attempt to update such adjustments
     Failure/Error: expect(order.adjustment_total).to eq expected_fees
     
       expected: 16
            got: 0.13e2
     
       (compared using ==)
     # ./spec/models/spree/order_spec.rb:1158:in `block (5 levels) in <top (required)>'

From what I could tell the math it's doing is 2*5 + 1*3, when it should be doing 2*(5+3) (see line 1118 of the spec file).

[Matt-Yorkley]

Alright, cool. Seems like it's an issue with the test setup as opposed to the code itself 👌

[Matt-Yorkley]

👍