// didn't add node definitions because this problem occurs on initial data fetch
// user type
const userType = new GraphQLObjectType({
name: 'User',
fields: () => ({
id: globalIdField('User'),
email: { type: GraphQLString },
posts: {
type: postConnection,
args: connectionArgs,
// getUserPosts() function is below
resolve: (user, args) => connectionFromArray(getUserPosts(user.id), args),
},
}),
interfaces: [nodeInterface],
})
// post type
const postType = new GraphQLObjectType({
name: 'Post',
fields: () => ({
id: globalIdField('Post'),
title: { type: GraphQLString },
content: { type: GraphQLString },
}),
interfaces: [nodeInterface],
})
// connection type
const {connectionType: postConnection} =
connectionDefinitions({name: 'Post', nodeType: postType})
// Mongoose query on other file
exports.getUserPosts = (userid) => {
return new Promise((resolve, reject) => {
Post.find({'author': userid}).exec((err, res) => {
err ? reject(err) : resolve(res)
})
})
}
I get the following warning in browser console:
Server request for query App failed for the following reasons:
Cannot read property 'length' of undefined
_posts40BFVD:posts(first:10) {
^^^
That's the only information I got, there's no more errors or references. What could be the reason?
This code is from relay-starter-kit, I only replaced all the Widget code with Post. Everything is almost the same as in starter, therefore I think the cause is somewhere around the database code.
But I can't see the problem because getUserPosts() returns same structure: array of objects..
What was the problem?
resolve: (user, args) => connectionFromArray(getUserPosts(user.id), args)
getUserPosts() returned a promise (blocking code would probably be a bad idea) but there was no callback. What happened in my opinion was that connectionFromArray() continued executing the code but it didn't have the data from getUserPosts() yet, which caused the whole system to fail.
One possible solution
I use Babel and JavaScript "future features" anyway, therefore I decided to use async and await. But there still was a problem and getUserPosts() returned an empty array.
Then I discovered that if another function is called with await in a async function, that another function has to be async as well, otherwise all await-s fail. Here's my final solution that works:
// async function that makes db query
exports.getUserPosts = async (userId) => {
try {
// db query
const posts = await Post.find({author: args}).exec()
return posts
} catch (err) {
return err
}
}
// and resolve method in Schema, also async
resolve: async (user, args) => {
const posts = await getUserPosts(user._id)
return connectionFromArray(posts, args)
}
Im not still sure though if it's the best way. My logic tells me that I should use async as much as possible in Node but Im far from Node expert. I'll update the question when I know more about it.
I would be glad to know if there's a better or even a recommended way to deal with this database query situation using Relay.
The problem is with the resolve function of posts field.
resolve: (user, args) => connectionFromArray(getUserPosts(user.id), args),
First, getUserPosts is an asynchronous function which returns a promise. You have to take that into account while writing resolve function. Second, user.id is a global ID field generated for you by graphql-relay module's globalIdField helper function. That ID should be converted to local ID.
With relay-starter-kit, you can use async and await. The code will look like:
resolve: async (user, args) => {
let userId = fromGlobalId(user.id).id;
// convert this userId string to Mongo ID if needed
// userId = mongoose.Types.ObjectId(userId).
const posts = await getUserPosts(userId);
return connectionFromArray(posts, args),
},
In your async version of getUserPosts, the err object is returned in case of error.
exports.getUserPosts = async (userId) => {
try {
// db query
const posts = await Post.find({author: args}).exec()
return posts
} catch (err) {
return err
}
}
A good practice is to either re-throw the error or return an empty array.
graphql-relay provides a connectionFromPromisedArray function which waits until the promise resolves. resolve: (user, args) => connectionFromPromisedArray(getUserPosts(user.id), args), which probably is the most recommended/easiest way when dealing with promised connections.
If your user.id is a global ID field then you have to extract the real DB ID from the base64 string const { type, id } = fromGlobalId(user.id);.
or
resolve: (user, args) => connectionFromPromisedArray(getUserPosts(fromGlobalId(user.id).id), args),
In a app I work on which is using Relay and Mongoose I found it better to write my own connection implementation to handle pagination and filtering on the DB level rather than on the app level. I took a lot of inspiration from graffiti-mongoose when I wrote it.
Related
I'm facing an issue with typeorm.
In order to init my database for my repository tests, I wanted to create a sql file with several inserts, and execute the file directly with the entity manager.
It seams that it allows only one query at the time.
Somebody has an idea how to digest directly an sql file in one execution?
Here is an example I'm trying to do.
(it works well with entities and saving them, but for more complicated tests, I'd rather use an sql file).
I don't want to use migration globally because I want all the tests to be independant.
describe('Lead time repository', () => {
let app: INestApplication;
let leadTimeRepository: LeadTimeRepository;
function populateDB() {
const initScript = fs.readFileSync(path.join(__dirname, 'populate.sql'), 'utf-8');
console.log(initScript);
leadTimeRepository.manager.query(initScript);
}
beforeAll(async () => {
const module = await Test.createTestingModule({
imports: [DatabaseModule, LeadTimeModule],
}).compile();
app = module.createNestApplication();
await app.init();
leadTimeRepository = app.get<LeadTimeRepository>(LeadTimeRepository);
populateDB();
});
afterAll(async () => {
await app.close();
});
describe('findLeadTimeForEvent', () => {
it('Should do something.', () => {
console.log('should do something');
});
});
});
Thanks by advance.
I've been playing around with stripe and would like to learn how to get ephemeral keys in the following way:
Back-end:
//Stripe API requirement for payment
exports.stripeEphemeralKey = functions.https.onCall((data, context) => {
const uid = context.auth.uid;
//Get values
admin.database().ref().child("users").child(uid)
.on("value", (snapshot) =>{
//Get user data
let user = snapshot.val()
//Log data
console.log("Create ephemeral key for:")
console.log(user)
//Create ephemeral key
stripe.ephemeralKeys.create(
{customer: user.customerid },
{stripe_version: '2018-11-08'}
)
.then((key) => {
console.log(key)
console.log("Succesful path. Ephemeral created.")
return "Testing"
})
.catch((err) => {
console.log("Unsuccesful path. Ephemeral not created.")
console.log(err)
return {
valid: false,
data: "Error creating Stripe key"
}
})
})
})
Client-side:
functions.httpsCallable("stripeEphemeralKey").call(["text": "Testing"]) { (result, error) in
print(result?.data)
}
I have tested this code by replacing the body of the stripeEphemeralKey with a simple "Testing" string and that returns just fine. But with the code above I just get Optional() back.
For testing I added lots of console logs. Firebase logs show the execution path gets to the "Succesful path. Ephemeral created." log, and furthermore I can actually see the ephemeral key I get back from stripe.
So, what is the proper correct way to get the ephemeral key in Swift for iOS using the onCall Firebase function?
The backend does what it should, but I can't seem to get the answer back.
Thank you.
The backend does not actually do what it should do. You're doing at least two things wrong here.
First, your callable function needs to return a promise that resolves with the value that you want to send to the client. Right now, your function callback isn't returning anything at all, which means the client won't receive anything. You have return values inside promise handlers, but you need a top-level return statement.
Second, you're using on() to read data from Realtime Database, which attaches a listener that persists until it's removed. This is almost certainly never what you want to do in a Cloud Function. Instead, use once() to get a single snapshot of the data you want to read, and act on that.
For my own reference, and those who might find this helpful:
//Stripe API requirement for payment
exports.stripeEphemeralKey = functions.https.onCall((data, context) => {
const uid = context.auth.uid;
return admin.database().ref().child("users").child(uid)
.once("value", (snapshot) =>{
console.log(snapshot.val() )
})
.then( (snap) => {
const customer = snap.val()
//Log data
console.log("Create ephemeral key for:")
console.log(customer)
//Create ephemeral key
return stripe.ephemeralKeys.create(
{customer: customer.customerid },
{stripe_version: '2018-11-08'}
)
.then((key) => {
console.log(key)
console.log("Succesful path. Ephemeral created.")
return {
valid: true,
data: key
}
})
.catch((err) => {
console.log("Unsuccesful path. Ephemeral not created.")
console.log(err)
return {
valid: false,
data: "Error creating Stripe key"
}
})
})
.catch( err => {
console.log("Unsuccesful path. Ephemeral not created.")
console.log(err)
return {
valid: false,
data: "Error gettting customerid"
}
})
})
The key seems to be to chain the initial database request with .then(), and do our our work chaining the returns uninterrupted as we use functions that return promises. In particular, placing my work-code inside the callback on the original admin.database().ref().once() function did not work for me.
I am new to this kind of programming, so someone who knows about this might the why better.
I have a call back function which is getting data from an external API and depends on a data check I have tried for a slot elicitation inside callback but looks like elicitation is not working inside the callback. Please find the code snippet below,
GetCustomerDetails().then(response => {
var serializedcustomerDetails = convert.xml2json(response.data, {
compact: true,
spaces: 2
});
var customerDetails = JSON.parse(serializedcustomerDetails);
let filteredCustomerDetails = _.filter(customerDetails.CustomerInfo.CustomerDetails, function (o) {
return o.CustomerName._text.includes(customerName);
})
if (filteredCustomerDetails.length == 1) {
callback(elicitSlot(outputSessionAttributes, intentRequest.currentIntent.name,
intentRequest.currentIntent.slots, 'CustomerCode', {
contentType: 'PlainText',
content: `Do you mean ${filteredCustomerDetails[0].CustomerName._text} of ${filteredCustomerDetails[0].SpecialityName._text} department?`
}));
return;
}
}).catch(error => {
console.log(`${error}`)
})
This is my first Awnser on stack so please bear with me.
I have come accross the same problem in a recent project and there are a few things that you can check.
How long does the API call take?
If your API call takes a long time it will be worth checking the timeout settings on your Lambda function. AWS Console -> Lambda -> Your Function -> Basic settings -> Timeout.
Does your Lambda function finish before the API call is done?
I fixed this issue by building a node module to handle my business logic, the module has a function called getNextSlot it returns as a Promise. Inside this function I check the incoming event and figure out which slot I need to elicit next, part of my flow is to call an API endpoint that takes around 10 seconds to complete.
I use the request-promise package to make the api call, this node module makes sure that the lambda function keeps running while the call is running.
exports.getData = function (url, data) {
var pr = require("request-promise");
var options = {
method: 'POST',
url: 'api.example',
qs: {},
headers:
{
'Content-Type': 'application/json'
},
body: {
"example": data
},
json: true,
timeout: 60000
};
return pr(options);
}
In my main code I call this function as:
apiModule.getData("test", "data")
.then(function (data) {
//Execute callback
})
.catch(function (error) {
console.log(error);
reject(error);
});
This solved the issue for me anyways.
Thanks,
I use Google Cloud Functions to create an API endpoint for my users to interact with the Realtime Database.
The problem I have is that I'm not sure how the code works. I have a helper function doSomething that I need to call only once, but I have a suspicion that there are cases where it can be called twice or possibly more (when multiple users call the API at the same time and the update operation hasn't been processed by the DB yet). Is it possible? Does it mean I need to use a transaction method? Thank you!
DB structure
{
somePath: {
someSubPath: null
}
}
Google Cloud Functions code
const functions = require('firebase-functions')
const admin = require('firebase-admin')
const cors = require('cors')({origin: true});
admin.initializeApp(functions.config().firebase)
// API ENDPOINT
exports.test = functions.https.onRequest((req, res) => {
cors(req, res, () => {
admin.database().ref('/somePath/someSubPath').once('value')
.then(snapshot => {
const value = snapshot.val()
if (value) return res.status(400).send({ message: 'doSomethingAlreadyCalled' })
doSomething()
const updates = { '/somePath/someSubPath': true }
return admin.database().ref().update(updates)
.then(() => res.status(200).send({ message: 'OK' }))
})
.catch(error => res.status(400).send({ message: 'updateError' }))
})
})
// HELPERS
const doSomething = () => {
// needs to be called only once
}
I believe you were downvoted due to the above pseudocode not making complete sense and there being no log or output of what your code is actually doing in your question. Not having a complete picture makes it hard for us to help you.
Just Going from your structure in the question, your actual code could be calling twice due to function hoisting. Whenever I have this issue, I’ll go back to the api documentation and try to restructure my code from rereading.
HTH
I use my postgres database query to determine my next action. And I need to wait for the results before I can execute the next line of code. Now my conn.query returns a Future but I can't manage to get it async when I place my code in another function.
main() {
// get the database connection string from the settings.ini in the project root folder
db = getdb();
geturl().then((String url) => print(url));
}
Future geturl() {
connect(db).then((conn) {
conn.query("select trim(url) from crawler.crawls where content IS NULL").toList()
.then((result) { return result[0].toString(); })
.catchError((err) => print('Query error: $err'))
.whenComplete(() {
conn.close();
});
});
}
I just want geturl() to wait for the returned value but whatever I do; it fires immediately. Can anyone point me a of a piece of the docs that explains what I am missing here?
You're not actually returning a Future in geturl currently. You have to actually return the Futures that you use:
Future geturl() {
return connect(db).then((conn) {
return conn.query("select trim(url) from crawler.crawls where content IS NULL").toList()
.then((result) { return result[0].toString(); })
.catchError((err) => print('Query error: $err'))
.whenComplete(() {
conn.close();
});
});
}
To elaborate on John's comment, here's how you'd implement this using async/await. (The async/await feature was added in Dart 1.9)
main() async {
try {
var url = await getUrl();
print(url);
} on Exception catch (ex) {
print('Query error: $ex');
}
}
Future getUrl() async {
// get the database connection string from the settings.ini in the project root folder
db = getdb();
var conn = await connect(db);
try {
var sql = "select trim(url) from crawler.crawls where content IS NULL";
var result = await conn.query(sql).toList();
return result[0].toString();
} finally {
conn.close();
}
}
I prefer, in scenarios with multiple-chained futures (hopefully soon a thing of the past once await comes out), to use a Completer. It works like this:
Future geturl() {
final c = new Completer(); // declare a completer.
connect(db).then((conn) {
conn.query("select trim(url) from crawler.crawls where content IS NULL").toList()
.then((result) {
c.complete(result[0].toString()); // use the completer to return the result instead
})
.catchError((err) => print('Query error: $err'))
.whenComplete(() {
conn.close();
});
});
return c.future; // return the future to the completer instead
}
To answer your 'where are the docs' question: https://www.dartlang.org/docs/tutorials/futures/
You said that you were trying to get your geturl() function to 'wait for the returned value'. A function that returns a Future (as in the example in the previous answer) will execute and return immediately, it will not wait. In fact that is precisely what Futures are for, to avoid code doing nothing or 'blocking' while waiting for data to arrive or an external process to finish.
The key thing to understand is that when the interpreter gets to a call to then() or 'catchError()' on a Future, it does not execute the code inside, it puts it aside to be executed later when the future 'completes', and then just keeps right on executing any following code.
In other words, when using Futures in Dart you are setting up chunks of code that will be executed non-linearly.