Skip to content
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

Add support for the PaymentIntent resource #606

Merged
merged 1 commit into from
Jul 9, 2018

Conversation

remi-stripe
Copy link
Contributor

@remi-stripe remi-stripe commented Jun 27, 2018

This resource is gated but stripe-mock has been updated to support the APIs so tests run as expected.

r? @brandur-stripe
cc @stripe/api-libraries
cc @michelle-stripe @jenan-stripe

paymentintent.go Outdated
ListParams `form:"*"`
}

type PaymentIntentNextSourceActionValue struct {
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We will need to do something smarter here as the shape of value changes based on type. This will be closer to what we do with BalanceTransaction and source

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from cba89ac to 92546f6 Compare June 28, 2018 21:07
paymentintent.go Outdated
switch s.Type {
case PaymentIntentNextActionAuthorizeWithURL:
fmt.Printf("%v", data)
err = json.Unmarshal(data, &s.Value.AuthorizeWithUrl)
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I know this part is incorrect because at that point data still contains the top-level info but I haven't yet figured out out I can fix it.

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from 92546f6 to 5b03e0d Compare June 28, 2018 21:08
Copy link
Contributor

@brandur-stripe brandur-stripe left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the continued epic work Remi! Left a few comments below, but looking great.

Especially whether there's a way to stub the tests easily beyond just skipping?

Basically the recommended way to do this in Go is to set up an HTTP test server that the client will interact with. If we can expect this in stripe-mock ~soon, I don't think it's worth the effort, and using t.Skip is the right option for now (we get a few guarantees by virtue that we know the tests compile, which is something at least).

ptal @remi-stripe

paymentintent.go Outdated
Created int64 `json:"created"`
Currency string `json:"currency"`
Customer *Customer `json:"customer"`
Livemode bool `json:"livemode"`
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Missing Description.

paymentintent.go Outdated
OnBehalfOf *Account `json:"on_behalf_of"`
ReceiptEmail string `json:"receipt_email"`
ReturnURL string `json:"return_url"`
SaveSourceToCustomer bool `json:"save_source_to_customer"`
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not seeing SaveSourceToCustomer in the latest documentation. Can you double-check that it's necessary?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it's a param only

client/api.go Outdated
@@ -148,6 +149,8 @@ type API struct {
PaymentSource *paymentsource.Client
// ExchangeRates is the client used to invoke /exchange_rates APIs.
ExchangeRates *exchangerate.Client
// PaymentIntents is the client used to invoke /exchange_rates APIs.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Slight fix needed in the /exchange_rates part of the comment.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

BTW, we should try to remember to alphabetize both these lists later :)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yep I can take after this PR, I did not want to mix both

paymentintent.go Outdated
ListParams `form:"*"`
}

type PaymentIntentSourceActionAuthorizeWithUrl struct {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should probably make this URL instead of Url.

paymentintent.go Outdated
// PaymentIntentSourceActionValue describes the `value` property in `next_source_action`
// The `type` in the parent should indicate which object is fleshed out.
type PaymentIntentSourceActionValue struct {
AuthorizeWithUrl *PaymentIntentSourceActionAuthorizeWithUrl `json:"-"`
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And similarly, with the name of this field: AuthorizeWithURL.


assert.Equal(t, PaymentIntentNextActionAuthorizeWithURL, intent.NextSourceAction.Type)
assert.Equal(t, "https://stripe.com", intent.NextSourceAction.Value.AuthorizeWithUrl.URL)
}
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would you mind adding a test specifically for PaymentIntentSourceAction's UnmarshalJSON? IMO, we should try to have a specific test for every type that has a custom UnmarshalJSON implementation.

It's fairly similar to ExternalAccount's unmarshaling, so I'd use the test TestExternalAccount_UnmarshalJSON in account_test.go for inspiration.

We could also then remove the next_source_action from TestPaymentIntent_UnmarshalJSON now that there's a more local test to check the same thing, but it's up to you.


func getC() Client {
return Client{stripe.GetBackend(stripe.APIBackend), stripe.Key}
}
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would you mind taking a stab at getting the new paymentintent/ package to pass Golint? Basically you'd be doing what I did in the PR that activated it for account and charge:

  1. Get Golint running locally:

    go get -u golang.org/x/lint/golint
    golint ./paymentintent
    
  2. Fix the linting problems (using the same style of comment from ./charge basically:

    paymentintent/client.go:1:1: package comment should be of the form "Package paymentintent ..."
    paymentintent/client.go:23:1: exported method Client.New should have comment or be unexported
    paymentintent/client.go:24:2: don't use underscores in Go names; var payment_intent should be paymentIntent
    paymentintent/client.go:35:1: exported method Client.Get should have comment or be unexported
    paymentintent/client.go:37:2: don't use underscores in Go names; var payment_intent should be paymentIntent
    paymentintent/client.go:48:1: exported method Client.Update should have comment or be unexported
    paymentintent/client.go:50:2: don't use underscores in Go names; var payment_intent should be paymentIntent
    paymentintent/client.go:61:1: exported method Client.Cancel should have comment or be unexported
    paymentintent/client.go:63:2: don't use underscores in Go names; var payment_intent should be paymentIntent
    paymentintent/client.go:74:1: exported method Client.Capture should have comment or be unexported
    paymentintent/client.go:76:2: don't use underscores in Go names; var payment_intent should be paymentIntent
    paymentintent/client.go:87:1: exported method Client.Confirm should have comment or be unexported
    paymentintent/client.go:89:2: don't use underscores in Go names; var payment_intent should be paymentIntent
    paymentintent/client.go:100:1: exported method Client.List should have comment or be unexported
    
  3. Add the package to the whitelist in Makefile:

    lint:
    	golint -set_exit_status ./account
    	golint -set_exit_status ./charge
    	golint -set_exit_status ./paymentintent
    

paymentintent.go Outdated
err = json.Unmarshal(data, &s.Value.AuthorizeWithUrl)
}

return err
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This UnmarshalJSON implementation is fine by me!

}

func (c Client) New(params *stripe.PaymentIntentParams) (*stripe.PaymentIntent, error) {
payment_intent := &stripe.PaymentIntent{}
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And LOL, I didn't even notice these right away, but saw them in the Linter's error messages, but these of course should be paymentIntent instead of payment_intent.

@brandur-stripe
Copy link
Contributor

brandur-stripe commented Jun 28, 2018

@remi-stripe Okay, here's a working UnmarshalJSON implementation that I came up with:

func (s *PaymentIntentSourceAction) UnmarshalJSON(data []byte) error {
	type paymentIntentSourceAction PaymentIntentSourceAction
	var v paymentIntentSourceAction
	if err := json.Unmarshal(data, &v); err != nil {
		return err
	}

	var err error
	*s = PaymentIntentSourceAction(v)
	s.Value = &PaymentIntentSourceActionValue{}

	// Unmarshal data a second time so that we can get the raw bytes for the
	// `value` field
	var rawObject map[string]*json.RawMessage
	if err := json.Unmarshal(data, &rawObject); err != nil {
		return err
	}

	switch s.Type {
	case PaymentIntentNextActionAuthorizeWithURL:
		err = json.Unmarshal(*rawObject["value"], &s.Value.AuthorizeWithUrl)
	}

	return err
}

And also, change the struct definition so that we have no JSON annotation for Value:

type PaymentIntentSourceAction struct {
	Type  PaymentIntentNextActionType     `json:"type"`
	Value *PaymentIntentSourceActionValue `json:"-"` // See custom UnmarshalJSON implementation
}

I think overall it's pretty clean, with just a couple caveats:

  1. It does have to call Unmarshal a second time. The first pass isn't doing much work though, so it's probably not too bad performance-wise.
  2. It's not ideal that we have to come up with yet another pattern for unmarshaling this late in the game :( I'm not sure there's much we can do about it here, but we it'd be kind of nice to get faster feedback from API library implementations for API review ... when there's a smell, it tends to be more noticeable here. The good news is that if we add strongly-typed structs for sources, Source's UnmarshalJSON will probably look a lot like this one's (although not exactly).

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from fe2545d to 8bba7db Compare June 28, 2018 22:18
@remi-stripe
Copy link
Contributor Author

Okay addressed all the comments and fixed the unmarshal logic thanks to Brandur. Only thing left is to decide changes to the /confirm parameters shape.

Key string
}

// New creates a payment_intent.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nit, but would you mind if we used the more human-readable "payment intent" for these comments?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

damn search and replace. Nice catch and fixed!

@brandur-stripe
Copy link
Contributor

Thanks for getting all that review feedback in so quickly!

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from 8bba7db to f37e1b4 Compare June 28, 2018 23:22
ReceiptEmail string `json:"receipt_email"`
ReturnURL string `json:"return_url"`
Shipping ShippingDetails `json:"shipping"`
Source *PaymentSource `json:"source"`
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since we only support real Sources, should this be Source directly? I think so

Copy link
Contributor

@brandur-stripe brandur-stripe Jul 5, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Arg, I just took a look at the source code and it's allows to be either a source or a card.

I really like the idea of it just being a Source as well, but we might have to check with the team for advice on how to handle this.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yep, talked to them about it and we want to support Cards moving forward so it will stay a PaymentSource unfortunately

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from f37e1b4 to 1171f3f Compare July 5, 2018 16:21
paymentintent.go Outdated
ReceiptEmail string `json:"receipt_email"`
ReturnURL string `json:"return_url"`
Shipping ShippingDetails `json:"shipping"`
Source *Source `json:"source"`
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since PaymentIntent will only support Sources and nothing else, I think it's the right call to use Source here.

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from d7746cd to d7cc37d Compare July 7, 2018 16:06
@remi-stripe remi-stripe changed the title [WIP] Add support for the PaymentIntent resource Add support for the PaymentIntent resource Jul 7, 2018
@remi-stripe remi-stripe removed their assignment Jul 7, 2018
@remi-stripe
Copy link
Contributor Author

Okay now that stripe-mock is updated the tests run and I think it is now ready to merge but needs a full review to make sure we did not miss anything

// Iter is an iterator for lists of PaymentIntents.
// The embedded Iter carries methods with it;
// see its documentation for details.
type Iter struct {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we ended up updating all these Iter comments in your other Lint PRs. Can we change this one too? (And maybe the comment on PaymentIntent() directly below too.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice catch, fixed other comments too

@brandur-stripe
Copy link
Contributor

Left a minor comment on comments, but otherwise looks great.

@remi-stripe remi-stripe force-pushed the remi-add-payment-intent branch from d7cc37d to 1f4a6cf Compare July 9, 2018 17:08
@remi-stripe
Copy link
Contributor Author

PTAL @brandur-stripe

@brandur-stripe
Copy link
Contributor

Thanks Remi!

@brandur-stripe brandur-stripe merged commit c547b00 into master Jul 9, 2018
@brandur-stripe brandur-stripe deleted the remi-add-payment-intent branch July 9, 2018 18:30
@brandur-stripe
Copy link
Contributor

Released as 36.0.0.

@brandur-stripe
Copy link
Contributor

And I just realized that I jumped the gun on this a little bit because we're probably going to tweak a few parts of this API. I think it's okay though since this is still behind a gate and it won't really be possible to start relying on it yet, so we can make those changes without breaking anyone.

@remi-stripe
Copy link
Contributor Author

@brandur-stripe My understanding was that the API was stable and changes would be versioned/punted to later on so I think it was ready to be merged. At least that's the feeling I got internally!

@brandur-stripe
Copy link
Contributor

@remi-stripe Oh great! I'd thought for some reason that some of the changes we'd talked about a few weeks ago were still WIP. Very happy to hear that they're already good to go. Thanks.

nadaismail-stripe pushed a commit that referenced this pull request Oct 18, 2024
* Improve error message for missing revised line item

* Stop connectionSuccessful, not used in latest JS code

* Documentation from video walkthrough

* Adding brian to pr template
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants