request-group
v1.2.6
Published
Handle multiple request async
Downloads
9
Maintainers
Readme
Request Group
Request Group is a module for primarily Node.js environments and written in TypeScript. The entire purpose is to handle batch request of multiple kinds in a friendly and agnostic way. Included in this module are two basic classes called FileRequest and WebRequest to get the ball rolling in your project. These class's are only basic and may not satisfy all your needs. If you are looking to expand or develop your own request type it would be wise to start by extending the BaseRequest or BaseWebRequest. If you want to go deeper you can implement the Request interface entirely on your own. This module will of course work in a regular Node.js enviorment if that is your preference, but all examples and the preferred way to use this module is to use it with TypeScript. All *.ds.ts files are included when building
Installing
npm install request-group
Building
git clone https://github.com/GabrieleNunez/request-group.git
cd request-group
npm install
npm run build
Creating a request-group
Fortunately creating a request-group is extremely simple and straight forward. If you are using TypeScript then RequestGroup only requires one type supplied and two parameters.
import { RequestGroup, WebRequest } from 'request-group';
// if you dont supply the queue size then the default is 4
const requestQueueSize: number = 4;
// if you don't supply the queue interval then the default is 2500 or 2.5 seconds
// the queue interval is given in millseconds since we use setTimeout
const requestQueueInterval: number = 2500;
// create a new Request Group with a page engine type of string
// the page engine is simply the output of the Request. This could be a class or a primitive.
// For example if we are using Cheerio this **should** be RequestGroup<CheerioStatic>
// If we are using puppeteer this **should** be RequestGroup<puppeteer.Page>
let foo:RequestGroup<string> = new RequestGroup<string>(requestQueueSize, requestQueueInterval);
// OR you can also create a Request Group by using request.createManager(...) like so
let webRequest: WebRequest = new WebRequest('https://github.com/GabrieleNunez/request-group');
let bar = webRequest.createManager(requestQueueSize, requestQueueInterval);
// both ways will work just fine and give you a Request Group
// if you want to hook into the request group and find out ASAP when a request is completed
// hook in a callback that way we can read and manipulate the page data
requestGroup.setRequestComplete((request: Request<CheerioStatic>): Promise<void> => {
console.log("You can manipulate the returned request here".);
return new Promise((resolve): void => {
console.log("You can also manipulate it here");
console.log("This request has completed");
resolve();
});
});
// after you create your Request Group just queue up the request
foo.queue(webRequest);
// when you are ready to run the queue
foo.run().then(() => {
console.log('Queue completed');
});
/**
* If you wanted to use await
* await foo.run();
* console.log('Queue completed');
*/
The PageEngine Type explained
RequestGroup<PageEngine>
The PageEngine is the output you are expecting from the request to return.
If you are using the built in request classes then the output is always going to be string.
To use Puppeteer and Cheerio as examples, if you wanted to create request to extend and return the output of those then the PageEngine **should** be puppeteer.Page or CheerioStatic respectively
How to use ( Extending )
Extending BaseWebRequest and using Cheerio as a page engine
cheerio_request.ts
// File: cheerio_request.ts
import { BaseWebRequest, Request } from 'request-group';
import * as cheerio from 'cheerio';
import * as moment from 'moment';
import * as URL from 'url';
import * as request from 'request';
export class CheerioRequest extends BaseWebRequest<CheerioStatic> {
protected userAgent: string | null;
public constructor(url: string, userAgent: string | null = null) {
super(url);
this.userAgent = userAgent;
}
/**
* Properly dispose of any resources that we need to get rid of
*/
public dispose(): Promise<void> {
return new Promise((resolve): void => {
resolve();
});
}
/**
* Make a request out to the interweb world
*/
public run(): Promise<Request<CheerioStatic>> {
let urlTarget = URL.parse(this.requestUrl);
let urlHost: string | undefined = urlTarget.hostname;
return new Promise((resolve, reject): void => {
if (urlHost === undefined) {
reject();
} else {
let headers = {
'User-Agent': this.userAgent,
Connection: 'keep-alive',
Accept: '*/*',
Host: urlHost as string,
};
if (this.requestCookie && this.requestCookie.trim().length > 0) {
headers['Cookie'] = this.requestCookie;
}
console.log('Requesting: ' + this.requestUrl);
request(
{
method: this.requestMethod,
uri: this.requestUrl,
headers: headers,
jar: typeof headers['Cookie'] == 'undefined' ? true : false,
strictSSL: false,
},
(error, response, body): void => {
this.momentPing = moment();
this.momentDone = moment();
this.requestCompleted = true;
this.momentDuration = moment.duration(this.momentInitiated.diff(this.momentDone));
this.requestErrors.push(error);
this.pageData = cheerio.load(body);
console.log('Cheerio Parse complete: ' + this.requestUrl);
resolve(this);
},
);
}
});
}
}
export default CheerioRequest;
index.ts
// index.ts
import CheerioRequest from './request_type/cheerio_request';
import { RequestGroup, Request } from 'request-group';
function cheerioExample(): Promise<void> {
return new Promise(
async (resolve): Promise<void> => {
// create our Request Group object
let requestGroup: RequestGroup<CheerioStatic> = new RequestGroup<CheerioStatic>(2, 1000);
// hook in a callback that way we can read and manipulate the page data
requestGroup.setRequestComplete(
(request: Request<CheerioStatic>): Promise<void> => {
return new Promise((requestResolve): void => {
let $: CheerioStatic = request.getPage();
console.log(request.getMetadata<string>('request-name') + ' Looping through messages');
console.log(
request.getMetadata<string>('request-name') +
' Total Messages: ' +
$('table.files td.message').length,
);
$('table.files td.message').each((index: number, element: CheerioElement): void => {
console.log(request.getMetadata<string>('request-name') + ' Iteration: ' + index);
let txt: string = $(element)
.text()
.trim();
if (txt.length > 0) {
console.log(request.getMetadata<string>('request-name') + ': ' + txt);
}
});
requestResolve();
});
},
);
// these are the things we want to crawl
let urls: string[] = [
'https://github.com/GabrieleNunez/request-group',
'https://github.com/GabrieleNunez/bronco',
'https://github.com/GabrieleNunez/thecoconutcoder.com',
'https://github.com/GabrieleNunez/webcam.js',
];
// loop through our urls and then add them into the queue
for (var i = 0; i < urls.length; i++) {
console.log('Adding: ' + urls[i]);
let cheerioRequest: CheerioRequest = new CheerioRequest(urls[i]);
let requestName: string | undefined = urls[i].split('/').pop();
// just in case sanity check
if (requestName === undefined) {
requestName = 'unknown-' + i;
}
cheerioRequest.setMetadata<string>('request-name', requestName as string);
cheerioRequest.setMetadata<number>('request-index', i);
// queue up the request we just made
requestGroup.queue(cheerioRequest);
}
console.log('Letting request queue run');
await requestGroup.run();
console.log('This request queue has completed');
resolve();
},
);
}
// demonstrate using cheerio for request
cheerioExample().then((): void => {
console.log('Completed');
});
How to use (Built in File)
index.ts
import { RequestGroup, Request, FileRequest } from 'request-group';
function fileExample(): Promise<void> {
return new Promise(
async (resolve): Promise<void> => {
// create our Request Group object
let requestGroup: RequestGroup<string> = new RequestGroup<string>(2, 2000);
requestGroup.setRequestComplete(
(request: Request<string>): Promise<void> => {
return new Promise((requestResolve): void => {
let fileRequest: FileRequest = request as FileRequest;
console.log('File read');
let fileContents: string = request.getPage();
console.log('- - -');
console.log(request.getMetadata<string>('name'));
console.log(
'Access: ' +
fileRequest.getFileAccessTime().format('dddd, MMMM Do YYYY, h:mm:ss a') +
'\tModified: ' +
fileRequest.getFileModifiedTime().format('dddd, MMMM Do YYYY, h:mm:ss a') +
'\tCreated: ' +
fileRequest.getFileCreatedTime().format('dddd, MMMM Do YYYY, h:mm:ss a'),
);
console.log('- - -');
console.log(fileContents);
console.log('- - -');
requestResolve();
});
},
);
let filePaths: string[] = [
'tsconfig.json',
'package.json',
'.eslintrc.js',
'.prettierrc.js',
'.eslintignore',
];
for (var i = 0; i < filePaths.length; i++) {
let request: FileRequest = new FileRequest(filePaths[i]);
let fileName: string | undefined = filePaths[i].split('/').pop();
if (fileName === undefined) {
fileName = 'unknown-' + i;
}
request.setMetadata<string>('name', fileName);
requestGroup.queue(request);
}
console.log('Letting file request run');
await requestGroup.run();
console.log('Request completed!');
resolve();
},
);
}
// demonstrate files using the included file request
fileExample().then((): void => {
console.log('Completed');
});
Found a problem? Have a suggestion? Want a feature?
I am looking to colloborate more with other developers. All issues will be addresses ASAP and pull request's welcomed. Let's make something great
Development Dependencies
- @types/node ^12.0.10
- @types/request ^2.48.1
- @typescript-eslint/eslint-plugin ^ 1.11.0
- @typescript-eslint/parser ^ 1.11.1
- eslint ^6.0.1
- eslint-config-prettier ^ 6.0.0
- eslint-plugin-prettier ^ 3.1.0
- prettier ^ 1.18.2
- ts-node ^ 8.3.0
- typescript ^ 3.5.2
Library Dependencies
- moment ^ 2.24.0
- request ^ 2.88.0
Keywords
request scraping manager http file utility browser async asynchronous web eslint typescript prettier moment batch group
ヾ(-_- )ゞ
〈( ^.^)ノ
ƪ(ƪ꒪ৱ৹)ᵒʰ˵˵