-
Notifications
You must be signed in to change notification settings - Fork 142
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Testing product sets - @W-12414116@ #1003
Conversation
<Modal data-testid={'sf-product-view-modal'} size="4xl" isOpen={isOpen} onClose={onClose}> | ||
<Modal size="4xl" isOpen={isOpen} onClose={onClose}> | ||
<ModalOverlay /> | ||
<ModalContent> | ||
<ModalContent containerProps={{'data-testid': 'product-view-modal'}}> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Unlike other Chakra components, Modal
does not accept any props you pass in. So to set the test id, using containerProps
like this is one way to do it.
@@ -62,13 +62,18 @@ export const AddToCartModal = () => { | |||
const intl = useIntl() | |||
const basket = useBasket() | |||
const size = useBreakpointValue({base: 'full', lg: '2xl', xl: '4xl'}) | |||
const {currency, productItems, productSubTotal, itemAccumulatedCount} = basket | |||
const {currency, productItems, productSubTotal} = basket |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
itemAccumulatedCount
is a duplicate, when we're calculating totalQuantity
in the line below. Removing the duplicate is also aligned with what's happening in the hooks-integration branch.
behavior: 'smooth', | ||
block: 'end' | ||
}) | ||
if (ref.scrollIntoView) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Added this check because during testing, scrollIntoView
is not available function.
) | ||
} | ||
|
||
const basketWithProductSet = { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Mocking this data inside this test file. I'm not sure what our approach is. Sometimes the mocked data is in separate file, sometimes not. I guess it depends on whether the data is going to be reused?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I agree, and I would add one more consideration: sometimes I separate the mock data just because it obscures the test and makes it harder to think about if the mock data is too long.
I would vote to move this out to a <test-name>.mock.js
file and have it live in the same directory (similar naming convention keeps them adjacent in your IDE) just because it distracts from the test
test('render multi-product layout', async () => { | ||
renderWithProviders(<MockedPageWithProductSet />) | ||
|
||
await waitFor(() => { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Technically, I don't need to wait here. But I'm foreseeing the time when the hooks-integration work is merged. By then, the page-level component would use hooks to fetch data. And this asynchronous behaviour would require calling waitFor
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm thinking most things should await
/ waitFor
for that reason. Not many of our components are "dumb" enough to not need the pattern
@@ -142,6 +161,7 @@ export const TestProviders = ({ | |||
value={addToCartModal} | |||
> | |||
{children} | |||
<AddToCartModal /> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Our TestProviders were missing this modal. We had the context in there (line 160), but not the modal to render yet.
@@ -2985,6 +2985,1625 @@ export const mockedCustomerProductListsDetails = { | |||
total: 1 | |||
} | |||
|
|||
// This wishlist has more varied products | |||
export const mockedCustomerProductListsDetails2 = { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What happened to the generator function we were working on when we paired on this? Can you call out the specific bits that are different about this mock vs the other?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've moved this data into a separate file ("something.mock.js" format) that is in a more specific location than the "app/commerce-api/mock-data.js". Because the mocked data is intended to be used only for a particular test.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@vmarta I think my main question got missed here. I realize I didn't really specify this, but I'm mainly interested in "how big is the difference between the existing mock data and this new mock data". If they're only 30% similar, I can understand not using the mock generator function pattern. If they're 90% similar, I would expect we want to use the generator pattern.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@bfeister oh sorry, the new mocked data is completely different in this case.
@@ -560,18 +561,24 @@ const MOCK_PRODUCT = { | |||
c_width: 'Z' | |||
} | |||
|
|||
test('Renders AddToCartModal', () => { | |||
test('Renders AddToCartModal with multiple products', () => { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I tend to make all tests use at least some amount of await
/ async
/ waitFor
so I would make this an async
function
} | ||
] | ||
} | ||
const {getByText} = renderWithProviders( | ||
|
||
renderWithProviders( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I would await renderWithProviders(
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good. I think we could do some cleanup around items I mentioned, but no "blockers". I can approve if urgent / time is short
expect(screen.getAllByText(MOCK_PRODUCT.name)[0]).toBeInTheDocument() | ||
expect(screen.getByRole('dialog', {name: /23 items added to cart/i})).toBeInTheDocument() | ||
|
||
const numOfRowsRendered = screen.getAllByTestId('product-added').length | ||
expect(numOfRowsRendered).toEqual(MOCK_DATA.itemsAdded.length) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I remember you saying something about "awaiting too many things" which I'm still a bit confused about. If you want all of the expect
conditions to be true at the same time (usually true, I believe) then I would wrap this in waitFor(() => {})
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think I won't use the waitFor
in this case (in the product-sets feature branch). Maybe it becomes necessary within the hooks-integration work.
After the component is rendered, I expect all those expect
conditions to be true.. all at the same time, yup. Because one render of the component is enough. It won't be re-rendered by some data fetching or async operation.
}) | ||
}) | ||
|
||
test('the View Full Details button', () => { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I realize it's unlikely to matter for such a tiny component, but I might still use async
here and waitFor
just to make sure the tests don't behave differently or fail in CI / different OS runtimes
const modal = screen.getByTestId('product-view-modal') | ||
expect(modal).toBeVisible() | ||
}, | ||
// Seems like rendering the modal takes a bit more time |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Awesome find! 🎉
) | ||
} | ||
|
||
const basketWithProductSet = { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I agree, and I would add one more consideration: sometimes I separate the mock data just because it obscures the test and makes it harder to think about if the mock data is too long.
I would vote to move this out to a <test-name>.mock.js
file and have it live in the same directory (similar naming convention keeps them adjacent in your IDE) just because it distracts from the test
test('render multi-product layout', async () => { | ||
renderWithProviders(<MockedPageWithProductSet />) | ||
|
||
await waitFor(() => { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm thinking most things should await
/ waitFor
for that reason. Not many of our components are "dumb" enough to not need the pattern
global.server.use( | ||
rest.post('*/baskets/:basketId/items', (req, res, ctx) => { | ||
return res(ctx.json(basketWithProductSet)) | ||
}) | ||
) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@yunakim714 did we ever decide whether this pattern (doing setup outside of a beforeEach
) can cause race conditions. Seems like this should be in beforeEach
to be sure we avoid race conditions
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We can scope this server.use in a describe block to avoid it being used by other tests
describe('add to cart test', () => {
beforeEach(() => {
// this will apply to the tests inside this describe block
global.server.use()
})
test('something', () => {})
})
Not related to product sets, but needing CI to pass.
* Product sets: basic PLP (#878) * Show product sets on PLP (alongside other products) For now, the product sets would look the same as any other kinds of products. * Parse the product type * Add todo * Ignore linting of this line for now * Add "Starting at" label * Product sets: basic pdp (@W-12301851@) (#897) * Product set: render the children products with minimal UI * No longer showing breadcrumbs in the child items * Individual swatches now have correct `value` There was a subtle bug where the `value` of the SwatchGroup was passed down to override all the values of the children Swatch components. * 1st attempt at saving swatches state in the page url * Avoid null in the url search params * Some refactoring * Add-to-cart now works for child items * Show accordion for each child item * Add todo * Rename variable for accuracy The child product itself is actually not a set, but is considered a set product. * Optional argument * Some refactoring and fix linting * Helper hook for url search params * Clean up * Rename module for a specific use case (PDP) * Show/hide add-to-cart buttons accordingly * Modal shows product set data * Parent product * Variant's product image shows up in add-to-cart modal * Rename variables for clarity * Fix linting error * Add horizontal rules to separate the parent and child items * Tweak whitespace * Localize the Starting At label * Add todo * Product sets: wishlist (@W-12301966@) (#917) * Starting At label * 1st attempt at updating wishlist for product sets * "Select Options" → "View Options" * Some refactoring * Render add-all-to-cart button if necessary * Better way to fetch `setProducts` data * Switch to zzrf-009 temporarily * For debugging purpose I'll revert this commit later * Simpler requirement * Update what quantity meant for product set * Some refactoring * Add comment * Add-to-wishlist for the individual child items * Rename variable * Update button text on PDP * Update button text on Wishlist page * Revert "Switch to zzrf-009 temporarily" This reverts commit 484c8d8. * PDP: show wishlist button on mobile * Fix formatting issues (prettier) * PR feedback * Clearer as to which is product or variant * Make it consistent with addToWishlist signature * [Feature] Add `Add Set to Cart` Button for Product Sets (#931) * Product sets: basic PLP (#878) * Variant's product image shows up in add-to-cart modal * Get modal showing multiple items added to cart * Add scroll to after validation Co-authored-by: Vincent Marta <[email protected]> * Einstein API to support product sets (#962) * verify viewcategory and viewsearch is working * viewproduct event is sent for every child product * addtocart event fires for all children when add set to cart * parent set only sends id * Design enhancements for product sets (#975) * remove complete the set carousel * remove complete the set translations * sticky buttons on mobile view * fix modal vertical margins on desktop * move recos to modalbody and cleanup margins * add dividers between product set items in mobile view * fix margins so sticky buttons do not overlap with content * add lazy loading of product set children product images * remove height - does nothing * bump build size to 55 * add conditional for complete the set and change loading var name * lint * revert translations * Cleanup * added to cart modal is scrollable and 90% of viewport * remove dialogprops * lint * Testing product sets - @W-12414116@ (#1003) * Product tile: test product set's price label * Test product set rendering * Test callbacks are called properly * itemAccumulatedCount is redundant * Test that modal can render multiple products * Finally the add-to-cart modal shows up in testing * Clean up * Test error messages * Test lazy loading of the child products * Split a test into smaller ones * Use `describe` to group all the related tests * Update comment * Test the wishlist and its primary action buttons * Clean up * Some refactoring * Some refactoring * Linting fix * Update comment * Test `getDisplayVariationValues` * Test the usePDPSearchParams * Tweak another test to pass Not related to product sets, but needing CI to pass. * Move mocked data into their own files * Moving it to within `beforeEach` --------- Co-authored-by: Ben Chypak <[email protected]> Co-authored-by: yunakim714 <[email protected]>
We'd like to add tests to verify the functionality of product sets in our retail-react-app.
Types of Changes
Changes
How to Test-Drive This PR
Checklists
General
Accessibility Compliance
You must check off all items in one of the follow two lists:
or...
Localization