Correct way to expect/receive state with Enzyme & MockProvider - graphql

I'm testing a component that is using a graphql useLazyQuery. MockProvider is provided by the Apollo recommended library #apollo/react-testing. I want to test that a certain message is being rendered base off the length of the data that is returned from the query. I have html elements structured like this:
<div className="message" data-id={props.data ? props.data.specials.length > 0 ? 'valid' : 'invalid' : ''}>
...children
</div>
I read through Apollo's docs about testing and wrote up a test like this:
mock = {
request: {
query: GET_PRODUCTS,
variables: { zip: "91001" }
},
result: {
data: {
specials: [
{
"_id": "5ecf28c459d3781a2e99738e",
},
{
"_id": "5ecf28c459d3781a2e99738f",
}
]
}
}
}
wrapper = mount(
<MockedProvider mocks={[mock]} addTypename={false}>
<Store.Provider value={[{ loading: false, zip: null }]}>
<GetZipCode />
</Store.Provider>
</MockedProvider>
)
await wait(0)
expect(wrapper.find(message).prop('data-id')).toEqual('valid')
But I've found that the tests do not update based on the mock that I put it. I have a test about this where I'm passing this value as the mock:
mock = {
request: {
query: GET_PRODUCTS,
variables: { zip: "32005" }
},
result: {
data: { specials: [] }
}
}
...after tests
expect(wrapper.find(message).prop('data-id')).toEqual('invalid')
And for both of these tests the expected value is "" which is the initial value for my data-id prop in my html element. If I were to set the initial value to "invalid" compared to "" then the expected value in my test would output "invalid".
It seems that no matter what I passed my mock provider it doesn't wait for it to be passed. I'm using the wait package that Apollo recommends as well.

If you wanna mock using jest, you use the following approach
jest.mock("apollo-client", () => ({
__esModule: true,
useQuery: (query: any) => {
//other mocks if needed
},
useLazyQuery: jest.fn().mockReturnValue([
jest.fn(),
{
data: {
yourProperties: "Add Here",
},
loading: false,
},
]),
}));
As you see, this approach even returns the mock function that is called in the code to be tested.

I faced in similar issues while testing with useLazyQuery. I would suggest writing a custom hook on top of useLazyQuery. This will have two benefits:
No need to wrap your test instance in Mock provider. You can mock the entire module (custom hook) using jest and use mockReturnValue method to simulate different behaviours (loading, data, error)
You can easily switch between actual query or some mock data in local file system while development.
Refer this blog for implementation details and demo app.

Related

How to map different JSON objects from the fixture into specific spec test file in the cypress

I have the below Input.json as fixture and It contains two different test cases.
Input.json (Fixture folder)
[
{
"searchKeyword":"cypress"
},
{
"username":"QATesting",
"password":"testprofile"
}
]
The above data will validate two different functionality of Google. One is going to validate search engine and another one is going to validate the user login activity (This is just for sample use case which may imitate my actual requirement).
I just created the cypress runner and I just want to run the spec file by using the below runner.js file
const cypress = require('cypress')
const fixtures = require('./cypress/fixtures/Test.json')
const promises = fixtures.map(fixture => {
return cypress.run({
env: {
fixture
},
spec: './cypress/integration/test.spec.js',
});
});
I just added two different It(test cases) respectively in the below "test.spec.js" file. And one test is gonna do the search function and another one is gonna check the existing user login activity:
describe("How to map two different data set with respective test function",() =>{
const baseUrl = "https://www.google.com/";
const testData = Cypress.env('fixture')
beforeEach("",()=>{
cy.visit(baseUrl);
});
it("Test Case1: Search the keyword", function () {
cy.xpath("//input[#name='q']").type(testData.searchKeyword);
cy.xpath("//input[#value='Google Search']").click();
cy.get("//ul/li[2]").should("be.visible");
});
it("Test Case2: login to the gmail account", function(){
cy.xpath("//a[contains(text(),'Sign in')]").click();
cy.xpath("//div[contains(text(),'Use another account')]").click();
cy.xpath("#identifierId").type(testData.username);
cy.xpath("//*[contains(text(),'Next')]").click();
cy.xpath("#password").type(testData.password);
cy.xpath("#submitbtn").click();
})
});
But the second test is getting failed and the testData.username return undefined.
Is there anyway to map the specific JSON array object with specific function in the test.spec.js file?
Not sure how to map the first dataset index with first It (Test case 1) and second dataset index with second test case respectively.
One quick way is to skip if the testData does not have the required properties,
describe("How to map two different data set with respective test function",() =>{
const baseUrl = "https://www.google.com/";
const testData = Cypress.env('fixture')
beforeEach("",()=>{
cy.visit(baseUrl);
});
it("Test Case1: Search the keyword", function () {
if (!testData.searchKeyword) this.skip
cy.xpath("//input[#name='q']").type(testData.searchKeyword);
cy.xpath("//input[#value='Google Search']").click();
cy.get("//ul/li[2]").should("be.visible");
});
it("Test Case2: login to the gmail account", function() {
if (!testData.username) this.skip
cy.xpath("//a[contains(text(),'Sign in')]").click();
cy.xpath("//div[contains(text(),'Use another account')]").click();
cy.xpath("#identifierId").type(testData.username);
cy.xpath("//*[contains(text(),'Next')]").click();
cy.xpath("#password").type(testData.password);
cy.xpath("#submitbtn").click();
})
});
Tagging
You can also get into tags, adding a tag property to the testData
[
{
"tag": "search",
"searchKeyword":"cypress"
},
{
"tag": "user",
"username":"QATesting",
"password":"testprofile"
}
]
Perhaps use a library like cypress-tags, then in the runner script
const cypress = require('cypress')
const fixtures = require('./cypress/fixtures/Test.json')
const promises = fixtures.map(fixture => {
if (fixture.tag) {
process.env.CYPRESS_INCLUDE_TAGS = fixture.tag
}
return cypress.run({
env: {
fixture
},
spec: './cypress/integration/test.spec.js',
});
});
Since your fixtures data is in a array and the username and password fields are at index 1, so in order to access those you have to use:
testData[1].username
testData[1].password
In case if you don't want to use the index value, change the fixture structure to:
{
"searchKeyword": "cypress",
"username": "QATesting",
"password": "testprofile"
}
And in your test directly use:
testData.username
testData.password

GraphQL mutation in KeystoneJS: "Cannot use 'in' operator to search for 'id' in undefined"

In a KeystoneJS GraphQL project I'm trying to create a new data object (an "Article") in the 'resolveInput' hook of another, existing, data object (a "Proposal" -- when a Proposal is approved, I create an Article based on that Proposals'data).
This worked fine using the Mongoose adapter, but I've tried to do it using the built in GraphQL API, using keystone.executeQuery and I get the following error:
My Article list has a relationship with one Proposal (I'm leaving out the other fields)
fields: {
proposal: {
isUnique: false,
type: fields_1.Relationship,
ref: 'Proposal',
access: {
create: true,
read: true,
update: false
}
}
}
and I create the new Article thus (I'm omitting some code)
hooks: {
resolvedInput: async params => {
const articleCreateInput = {
title: cleanTitle,
text: cleanText,
visible: HIDDEN,
proposal: {
connect: {
id: proposalId
}
}
};
const articleCreationResult = await keystone.executeQuery(createArticle, { variables: articleCreateInput });
}
}
As far as I can see this is the correct way to do it, using connect, connecting an existing item to one you are creating.
My query is
const createArticle = `mutation createArticle($data:ArticleCreateInput) {
createArticle(data:$data) {
id
title
visible
publishDate
}
} `;
and as far as I can see I'm following the schema correctly
I'm sure I'm making an obvious mistake but at the moment I don't see it -- and I'm not sure whether that mistake is a GraphQL mistake or a KeystoneJS mistake (or both).

cypress XHR API call validation in test cases

I am exploring if I can use cypress for end-to-end testing or not in angular? I am super beginner in it.
Cypress has some server() instance for XHR.
Question:
Suppose I am testing the login page so I can write test cases for querying elements and do the validation. In this process the browser will be making some API call, will it possible to write test cases for validating what was the statusCode the API had retured? What was XHR API response etc?
of course. With cypress you can spy the requests or mock them.
I have written a quick example to show you both methods:
describe("test", () => {
it("spy", () => {
cy.server();
cy.route("POST", /.*queries.*/).as("request")
cy.visit("https://docs.cypress.io/")
.get("#search-input").type("1234567890")
.wait("#request").then(xhr => {
expect(xhr.status).to.eq(200)
})
})
it("mock", () => {
cy.server();
const obj = JSON.parse(`
{
"results": [{
"hits": [{
"hierarchy": {
"lvl2": null,
"lvl3": null,
"lvl0": "Podcasts",
"lvl1": null,
"lvl6": null,
"lvl4": null,
"lvl5": null
},
"url": "https://stackoverflow.com",
"content": "mocked",
"anchor": "sidebar",
"objectID": "238538711",
"_snippetResult": {
"content": {
"value": "mocked",
"matchLevel": "full"
}
},
"_highlightResult": {
"hierarchy": {
"lvl0": {
"value": "Podcasts",
"matchLevel": "none",
"matchedWords": []
}
},
"content": {
"value": "mocked",
"matchLevel": "full",
"fullyHighlighted": false,
"matchedWords": ["testt"]
}
}
}
]
}
]
}
`);
cy.route("POST", /.*queries.*/, obj)
cy.visit("https://docs.cypress.io/")
.get("#search-input").type("1234567890")
.get("#algolia-autocomplete-listbox-0").should("contain", "mocked")
})
})
The spy example receives the raw XHR object and thus you are able to check the status code and so on.
The mock example shows you how you can mock any ajax request.
Please note: Currently you can not spy & mock fetch requests. But as far as I know they are rewriting the network layer in order to make this possible. Let me know if you need further assistance

Why is an Array in my payload being flattened in Sinatra / Rack::Test?

I'm trying to test a small Sinatra app using rspec. I want to pass a rather complex payload and am running into issues i do not understand: my payload contains an array of hashes. When I run the actual application this will work as expected, yet when I use the post helper to run my tests, the array will contain a merged hash:
post(
"/#{bot}/webhook",
sessionId: "test-session-#{session_counter}",
result: {
contexts: [
{ some: 'fixture' },
{ name: 'generic', parameters: { facebook_sender_id: 'zuck-so-cool' } }
]
}
)
In the sinatra handler I use params to access this payload:
post '/:bot/webhook' do |bot|
do_something_with(params)
end
When I now look at the structure of params when running the test suite, I will see the following structure:
[{"some" => "fixture", "name" => "generic", "parameters" => {"facebook_sender_id" => "zuck-so-cool"}}]
which I do not really understand. Is this a syntax issue (me being a ruby noob), am I using params wrong, or is this a bug?
EDIT: So i found out this is an "issue" with the way that Rack::Test will serialize the given payload when not specifying how to (i.e. as form data). If I pass JSON and pass the correct headers it will do what I expect it to do:
post(
"/#{bot}/webhook",
{
sessionId: "test-session-#{session_counter}",
result: {
contexts: [
{ some: 'fixture' },
{ name: 'generic', parameters: { facebook_sender_id: 'zuck-so-cool' } }
]
}
}.to_json,
{ 'HTTP_ACCEPT' => 'application/json', 'CONTENT_TYPE' => 'application/json' }
)
Still I am unsure of this is an issue with the passed data structure not being possible to be serialized into form data or if this is a bug in the way that Rack::Test serializes data.
Looking at the relevant portion of the specs it looks like this is is expected behavior.

Type error with internationalization feature of sails.js based on i18n

I'm trying to use the internationalization feature of sails based on i18n.
In my controller it works well. However, I would like to setup this in my model definition.
Please see the code below:
module.exports = {
attributes: {
name:{
type:'string',
required:true,
displayName: sails.__("test")
},
....
Unfortunately it does not work. I have the error below:
displayName: sails.__("test")
^
TypeError: Object [a Sails app] has no method '__'
Would you have an idea?
Any help will be very much appreciated.
Thanks,
displayName: sails.__("test")
You are trying to invoke the internationalization function statically; that is, you're seeing the error because you're running that function the moment your .js file is require()d by node.js, and before sails has finished loading.
There are two ways you can go about solving this problem.
1. Translate the value on each query
If you'd like to store the original value of displayName, and instead internationalize it each time you query for the model, you can override toJSON().
Instead of writing custom code for every controller action that uses a particular model (including the "out of the box" blueprints), you can manipulate outgoing records by simply overriding the default toJSON function in your model.
For example:
attributes: {
name:{
type:'string',
required:true,
},
getDisplayName: function () {
return sails.__(this.name);
},
toJSON: function () {
var obj = this.toObject();
obj.displayName = sails.__(this.name);
return obj;
},
...
}
2. Translate the value before create
You can use the Waterline Lifecycle Callbacks to translate the value to a particular language before the model is saved to the databas
Sails exposes a handful of lifecycle callbacks on models that are called automatically before or after certain actions. For example, we sometimes use lifecycle callbacks for automatically encrypting a password before creating or updating an Account model.
attributes: {
name:{
type:'string',
required:true,
},
displayName: {
type: 'string'
},
...
},
beforeCreate: function (model, next) {
model.displayName = sails.__(model.name);
next();
}
This internationalized the value of displayName will now be set on your model before it is inserted into the database.
Let me know how this works out for you.
Your solution is interesting. However, my wish would be to have a display name for each properties.
module.exports = {
attributes: {
name:{
type:'string',
required:true,
displayName: "Your great name"
},
adress:{
type:'string',
required:true,
displayName: "Where do you live?"
},
....
So is there a simple or clean solution to apply sails.__( foreach properties display name of the attribute?
Thanks,

Resources