Multiple async calls with single await
Installation
npm i craftx -S
yarn add craftx
Import
You can use any of these methods to use craftx in your application
import { fn, json } from "craftx";
import craftx from "craftx";
const { fn, json } = require("craftx");
const craftx = require("craftx");
Introduction
Craftx is a powerful javaScript library that let you craft or call asynchronous functions with single await at the end. This library comes with two built-in functionalities that covers all the basis:
- When you create a JSON, you are guaranteed to have all the fields resolved instead of pending promises.
- When you access function parameter, you are guaranteed to have all the parameters resolved instead of pending promises.
1. JSON
TL;DR
What do you think output would be?
const calcAge = async () => 65;
(async () => {
console.log({
age: calcAge(),
});
})();
How do we solve this problem? Either by adding a lot of awaits for each promise, or just by simply:
const { json } = require("craftx");
const calcAge = async () => 65;
(async () => {
console.log(
await json({
age: calcAge(),
})
);
})();
Try it on RunKit
Read in detail
The Problem
Let's say you need to create a JSON
{
id: 1,
name: 'John F. Kennedy',
age: 45,
}
This is good as long as you are not using promises, but if you want to use promises like:
const queryName = () => Promise.resolve('John F. Kennedy');
const calculateAge = () => Promise.resolve(45);
{
id: 1,
name: queryName(),
age: calculateAge(),
}
this will result potentially unresolved promises.
and if you do this:
{
id: 1,
name: await queryName(),
age: await calculateAge(),
}
the calls are now sequential and defeats the purpose of non-blocking event based I/O that Node.js is known for.
Solution
To handle this situation without wrapping the promise resolution in a separate function, you can use this utility package to handle this situation
const { json } = require("craftx");
await json({
id: 1,
name: queryName(),
age: calculateAge(),
});
This will resolve only after all the promises are resolved. Resulting in:
{
id: 1,
name: 'John F Kennedy',
age: 45,
}
Max depth
Default object depth supported by craftx is 64
.
2. Async Function
TL;DR
What do you think output would be?
const getAge = async () => 123;
const getDetail = async (age) => `The age is ${age}`;
(async () => {
const result = await getDetail(getAge());
console.log(result);
})();
How do we solve this problem?
const { fn } = require("craftx");
const getAge = async () => 123;
const getDetail = fn(async (age) => `The age is ${age}`);
(async () => {
const result = await getDetail(getAge());
console.log(result);
})();
Try it on RunKit
Read in detail
Problem
There are times when you want to use promise values without awaiting (i.e, automatically once the promise is fulfilled). This utility helps you achieve this goal using native promise mechanism.
We will walk you through an example and provide explanation where necessary.
Let's say you have various utility functions to query the database.
const queryCompanyInfo_ = async () => ({
username: "@amrayn",
});
const queryAccountInfo_ = async (company) => ({
company,
created: "19-02-2020",
});
Notice the queryAccountInfo_
takes company
parameter (promise) that will be provided by queryCompanyInfo_
. But you don't know whether this promise is fulfilled or not. If you use Promise.all
directly (without this library) you won't be able to provide this (resolved) company object to queryAccountInfo_
.
Promise.all([
queryCompanyInfo_(),
queryAccountInfo_(),
]).then(([userInfo, accountInfo]) => {
console.log(accountInfo);
});
A possible solution is to await for the promises first:
const companyInfo = await queryCompanyInfo_();
const accountInfo = queryAccountInfo_(companyInfo);
This has 2 basic problems.
- You're not making use of parallelism here. Which defeats the purpose of Promises to some extent.
- The code is very soon going to be messy and unreadable.
Solution
craftx allows you to "craft" a function that will help you achieve your goal without worrying about any of the above problem.
const { fn } = require("craftx");
const queryCompanyInfo = fn(queryCompanyInfo_);
const queryAccountInfo = fn(queryAccountInfo_);
const finalJson = await queryAccountInfo(queryCompanyInfo());
This will result in:
{
company: {
username: '@amrayn',
},
created: '19-02-2020',
}
Misc
Convert param based function
fn()
is good when the function takes normal parameters, e.g, myfn(param1, param2)
but this will not work if we have json based parameters, e.g, myfn({ param1: 1, param2: Promise.resolve(2) })
For this fnjson()
was added.
Simply pass in function in to this function.
const getName = ({ firstName, lastName }) => `${firstName} ${lastName}`;
const getNameSync = fnjson(getName);
Now if you pass in JSON with unresolved promises to the function, it would be correctly passed in to getName
function
Get Object Value
If you have a function that returns an object, and you want to grab just one specific value from the object, you can use built-in get
function to do that.
const { get } = require("craftx");
const getProfile = async (uid) => ({
name: "John",
age: 45,
father: {
name: "Peter",
},
});
(async () => {
console.log(await get(getProfile(), "father.name"));
console.log(await get(getProfile(), "mother.name", "Steph"));
console.log(await get(getProfile(), "brother.name"));
})();
Synopsis: get(object, path, defaultValue, options)
. The options
is passed through to fn()
internally.
NOTE: This function uses lodash.get to resolve the JSON path.
Options
If the first parameter is an object for the fn()
, that object is used for setting up the options.
For example:
const getNumb = fn(() => 123, {
startTime: res.startTime,
endTime: res.endTime,
});
You can also override the options for a crafted function later.
getNumb.setOptions({
startTime: res.startTime,
endTime: res.endTime,
name: "getNumb",
description: "Get number",
debug: false,
});
Following are the possible options
Option | Description |
---|
name | An identity for the function. Defaults to <function>.name - IT MUST NOT CONTAIN SPACE |
description | A description for the function |
startTime | Function for server timing - (name, description) => {} - the name and description is passed back to this function |
endTime | Function for server timing - (name) => {} - the name is passed back to this function |
debug | Boolean value to tell craftx whether debug logging is enabled or not. It will use a global logger.debug() object. If no such object exists, it will use console.debug() |
Bulk Export (Advanced)
Converting existing exports to crafted functions is easy, either using fn
for each function which can be cumbersome depending on number of functions; or you can simply convert the whole object using a helper function fnExport
.
Let's say you have:
const function1 = () => {};
const function2 = () => {};
module.exports = {
function1,
function2,
};
Just use fnExport
when exporting
const { fnExport } = require("craftx");
const function1 = () => {};
const function2 = () => {};
module.exports = fnExport({
function1,
function2,
});
Alternatively, you can do it when importing like in example of /examples/json.js
. Doing it multiple times does not harm.
License
Copyright (c) 2020-present @abumq (Majid Q.)
https://github.com/abumq/craftx
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.