craftx

Multiple async calls with single await

Installation

npm i craftx -S
# or
yarn add craftx

Import

You can use any of these methods to use craftx in your application

// using import
import { fn, json } from "craftx";

import craftx from "craftx";

// using require

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; // note the async

(async () => {
  console.log({
    age: calcAge(),
  });
})();

// output:
// {age: Promise}
// Promise {<fulfilled>: undefined}

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({
      // notice using craftx' json() function
      age: calcAge(),
    })
  );
})();

// output:
// {age: 65}

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);
})();

// output:
// The age is [object Promise]
// Promise {<fulfilled>: undefined}

How do we solve this problem?

const { fn } = require("craftx"); // or you can import using const craftx = require('craftx') and use craftx.fn(...)

const getAge = async () => 123;
const getDetail = fn(async (age) => `The age is ${age}`); // notice the wrap around "fn"

(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_(), // notice we cannot provide "company" here
]).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"); // or you can import { fn }

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")); // output: Peter
  console.log(await get(getProfile(), "mother.name", "Steph")); // output: Steph
  console.log(await get(getProfile(), "brother.name")); // output: undefined
})();

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({
  // hint: server-timing
  startTime: res.startTime,
  endTime: res.endTime,

  name: "getNumb",
  description: "Get number",

  debug: false, // to enable craftx debugging
});

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.

Changelog

3.2.4

• No change to API (ghost release)
• Using Github actions to publish package and run tests