- presence: true is redundant since Rails 5.0 BUT applies
with new default config of
belongs_to_required_by_default to true.
Lots of files with belongs_to_required_by_default = false
(backward compatibility).
So: deleting this setting implies to adding optional: true
Most production servers don't use the source locale `en`. Even if the
default language is English, they use a local variant like `en_AU` or
`en_GB` to customise some of the translations.
However the environment is configured, the app should always fallback to
`en` if no other translation is available.
This was another large file, potentially causing a bottleneck.
All the order setup is duplicated from the other file which is a bit of shame, but I think it makes sense.
The orders file is too big and causes a bottleneck for parallelising specs.
Maybe they should be merged with the above specs, but I'm not familiar enough to know for sure.
This was actually shown in one place and represents a user-facing
change. But you weren't able to edit the field which means that only
very old enterprises would have had this field set and were not able to
change it anymore.
I searched au-prod and found the following values in the database:
- "Friday 31st January"
- "From 4pm, Monday 30 September"
- "From 5pm-7pm Monday"
- "Saturday 27 April 12noon"
- "January 31st/February 1st"
- "Saturday 1st February"
They seem specific to a certain order cycle and have no value as
fallback any more. Seems safe to remove.
It contains a fix for stimulus_reflex 3.5.0-rc4, but hasn't been released yet.
https://github.com/podia/stimulus_reflex_testing/issues/21
Now the user reflex spec works, but the products spec needs another fix. But we plan to remove it soon anyway.
Unfortunately the testing framework doesn't support the latest version of StimulusReflex.
This is a third party framework that doesn't seem to be maintained. Maybe we should convert these to ActionCable tests.
We will add a migration to sanitise all existing descriptions but before
we do that destructive action, it's good to test this in a read-only
fashion first.
- Styling(in red) for the remove button/link in view
- A remove method to the bulk_form controller
- removes elements from the Dom
- removes changed elements from the binded Array in controller
- so that menu that indicates changes disappear and blured elements
- resume to non blurring state
- Added the corresponding specs
- test with one, two variants
- test with two different products
Most of the time this doesn't get called because source_required: false.
But sometimes it [does happen](https://app.bugsnag.com/yaycode/openfoodnetwork-uk/errors/66329690f4b6380007e8a4f8)
I have a feeling that source_required? could be moved to the superclass as payment_source_class.present?. But I don't know enough about this area of the system to try it...
We can see on the respective controller spec, that having a Stripe SCA payment, with no source does not trigger the error 400, observed on the legacy checkout.
I removed it because it needed a complete rewrite and I didn't think it
was valuable enough to do so. But now that I deactivated RSpecs
validating mocks again, this spec passes and we can do that work another
time.
One spec failed due to this new behaviour. Converting the shared
examples to simple blocks solved this. But the specs could probably be
written better now. I didn't invest this time.
The KnapsackPro queue mode can't predict which specs it will run. So we
need to check on each file (top-level describe block) which type of spec
it is and if we need to compile assets for it.
Old versions of KnapsackPro would execute the `before(:suite)` hooks on
every batch, but now it's only run once. With this change, we do the
same as before.
- removes the pending
- add the click to go to the 'Order Details' page
to check for the 'shipped' status
- from enqueued to have_been_enqueued so the spec
is flaky-free
