[Cart][Behat] Logging customer's cart

[Zales0123]

Q	A
Bug fix?	no
New feature?	yes
BC breaks?	no
Deprecations?	no
Fixed tickets
License	MIT
Doc PR

Based on #4035. Again, please, don't review this until previous PR's will be merged.

[michalmarcinkowski]

@Zales0123 please rebase.

[Zales0123]

@michalmarcinkowski done ;) And awaiting review, of course.

[pamil]

Let's put : ~ at the end to be sure that context class will be interpreted as a key.

[Zales0123]

👍

[pjedrzejewski]

Maybe we can move this to separate step:

And user/customer "john@example.com" has default shipping address to "Australia"

[pjedrzejewski]

I did not review everything yet because I'd like to start a quick discussion:

1. I think we are missing one scenario here, when someone is logged in before adding to the cart. The reason is we want to verify that cart is recalculated when someone signs in and has address, but we also want to verify that it is considered at the time of adding to the cart.
2. More natural scenario (short version), maybe that's even separate feature:

Background:
    I am logged in customer
    And my default shipping address is to "Australia"

Scenario: Adding item to the cart and getting proper tax information
    When I add product ...
    Then ...

E-Mail and password are irrelevant when we are logged in already. What do you think? I know this is something we can add in separate PR but I wanted to start discussion already.

Btw. something I just realized, from scenarios it is perfectly clear what is going on and they are not coupled to UI just like we wanted, but from UI perspective we are checking the "Cart Summary View" right now. We need to remember there is "Checkout Summary View" at the end and we should ensure that correct amounts are displayed there as well. Something to think about as we progress.

[michalmarcinkowski]

1. 👍
2. IMO this is the same feature, but we could probably add a concept of default user? E.g. in this scenario I don't care about it's email or password, so I would like to read that there is a customer that is logged in. WDYT? (let's do it in a separate PR)

For checkout the "Checkout Summary View" I would say we could have a separate scenarios for verifying that "Checkout Summary View" displays the same as "Cart Summary View". This way we will not need to check it on every single scenario.

[Zales0123]

1. What do you mean by "missing scenario, when someone is logged in before adding to cart"? Wasn't all previous scenarios about already logged in customer? 😄 For example in this feature we are checking proper taxes calculation on order for logged in "john@example.com" from "United Kingdom" :)
2. 👍 for default customer, for cases like this one.

[michalmarcinkowski]

@Zales0123

1. I would say we should have this specific scenario (defined shipping address) here. In the feature you linked to there is no need to specify the customer address, it should be removed.

UPDATE

IMO in the feature you linked to we shouldn't even be logged in, as it doesn't change anything and we can validate that guest cart works properly :)

[Zales0123]

@michalmarcinkowski Agreed 👍

[michalmarcinkowski]

@Zales0123

These two steps are also irrelevant since we do not even start a checkout there.

[Zales0123]

As we can see, first feature must have some childhood disease 😄
OK, so suma summarum:

• create And I am logged in customer step with default customer
• separate step And my default shipping address is to "Australia" from And I am logged in customer
• move here step for logged in user with already defined shipping country
• remove redundant steps from here

Am I missing sth?

[michalmarcinkowski]

@Zales0123 👍

But let's merge this and do it all in a separate PRs. /cc @pjedrzejewski

[pjedrzejewski]

Agreed about separate PR. 👍 Thank you Mateusz!