The fixed partial was used twice on the same page. So the used element
id appeared twice on the page which is invalid HTML. It also confused
the spec.
Now we look at the actually displayed text to select the correct
element.
https://github.com/openfoodfoundation/openfoodnetwork/issues/3477
The admin orders edit form was displaying adjustments even if they were
"not eligible". For example, an additional fee with amount `0` is not
eligible. They were already hidden in the adjustments view and we are
hiding them in reports.
The changed spec was failing from time to time. This change seems to
make it stable. It may be worth moving tag entering into a helper that
verifies that the tag was entered (waits for it).
A previous pull request added support for flexible decimal characters
when editing money amounts.
https://github.com/openfoodfoundation/openfoodnetwork/pull/1831
This pull request applies the same principle to the weight calculator
which was missed in the previous pull request.
The `registration_path` helper resolves to `/signup` in Spree
controllers due to spree_auth_device > config > routes.rb.
We worked around that in:
https://github.com/openfoodfoundation/openfoodnetwork/pull/3174
Here we add a spec for this so that we can test more easily if we
remove that workaround or detect it's accidental removal.
The `registration_path` helper resolves to `/signup` in Spree
controllers due to spree_auth_device > config > routes.rb.
We worked around that in:
https://github.com/openfoodfoundation/openfoodnetwork/pull/3174
Here we add a spec for this so that we can test more easily if we can
remove that workaround or detect it's accidental removal.
We keep tests running by default without cache, keeping it as it was before 10a79d5a65
Additionally, cached_products_renderer no longer depends on Rails.env, so here we remove that dependency from it's spec.
The shipping methods are updated to their target settings after the
subscription order has been created, so the order is created with the
"require_ship_address" factory default which is "true".
The shipping method setting at time of order creation has implications
on whether a shipment is set up for the order or not.
This fixes some assertions from using the same subscription shipping and
billing addresses, and distributor address.
One of the specs also pass because the subscription shipping address
matches the distributor address. This commit makes that scenario
clearer.
