phantomjscloud
v3.5.5
Published
Official PhantomJs Cloud Client API Library for Node.js
Downloads
1,112
Maintainers
Readme
PhantomJsCloud
This library provides a simple and high quality mechanism for interacting with Chrome as a scaleable cloud service. It leverages PhantomJsCloud.com (A Headless Browser Service) to allow multiple calls in parallel at high performance.
Chrome by default.
- You may also use the obsolete PhantomJs Browser instead of the default (Chrome Latest, v74+, via Puppeteer) by setting the
backend="phantomjs"
parameter (see options, below).
Requirements
- Internet access.
- Tested on Nodejs
10.15.2
, but should work with Nodejs >=6.x
Optional: Get an ApiKey by creating an account at PhantomJsCloud.com
Platforms
- This works in Node.js and Browsers (via Browserify or Webpack).
- Also tested to work using node via Amazon Lambda and Google Cloud Functions.
Basic Usage
- Install:
npm install phantomjscloud --save
- Use:
var phantomJsCloud = require("phantomjscloud");
var browser = new phantomJsCloud.BrowserApi();
browser.requestSingle({ url: "http://www.example.com", renderType: "plainText" }, (err, userResponse) => {
//can use a callback like this example, or a Promise (see the Typescript example below)
if (err != null) {
throw err;
}
console.log(JSON.stringify(userResponse.content));
});
or a slightly more complex example using async/await via Typescript:
//typings will automatically be loaded
import phantomJsCloud = require("phantomjscloud");
let apiKey = undefined;//leave undefined to use a demo key. get a free key at https://Dashboard.PhantomJsCloud.com
let browser = new phantomJsCloud.BrowserApi(apiKey);
async function doWork(){
//the page you wish to render
let pageRequest: phantomJsCloud.ioDatatypes.IPageRequest = { url: "http://www.example.com", renderType:"plainText" };
//make a request to render a single page, returning the plain-text contents of the page.
const userResponse = await browser.requestSingle(pageRequest);
console.log(JSON.stringify({
//the content of the page you requested
content: userResponse.content,
//metadata about your request, such as billing
meta: userResponse.meta,
//the status code of your request
statusCode: userResponse.statusCode
}, null, "\t")); //pretty-print
}
doWork();
Options
phantomJsCloud.BrowserApi()
var browser = new phantomJsCloud.BrowserApi(apiKeyOrOptions?) : BrowserApi;
Constructing the browserApi, and optionally for setting default configuration options.
apiKeyOrOptions
: Optional. If set, can be either anapiKey:string
or anoptions
object with the parameters{apiKey, endpointOrigin, suppressDemoKeyWarning, requestOptions, retryOptions }
.apiKey:string
Optional. If not set, the default "demo" key is used with the public cloud endpoint. If you use the default demo key, you get 100 Pages/Day. If you sign up for an account at PhantomJsCloud.com you will get 500 Pages/Day free.endpointOrigin:string
Optional. Used if you subscribe to a Private Cloud + SLA. Defaults to the PhantomJs Cloud Public Endpoint.suppressDemoKeyWarning
Optional. set to true to not show a warning for using demo keys.requestOptions
Optional. override HTTP request options, default to undefined (use defaults). Takes the following parameters:timeout
Optional. default timeout for the network request is 65000 (65 seconds)
retryOptions
Optional. override network failure retry options, default to undefined (use defaults) Takes the following parameters:timeout
Optional. assumes the network request timed out if it takes longer than this. default is 66000 (66 seconds)max_tries
Optional. maximum number of attempts to try the operation. default is 3interval
Optional. initial wait time between retry attempts in milliseconds(default 1000)backoff
Optional. if specified, increase interval by this factor between attemptsmax_interval
Optional. if specified, maximum amount that interval can increase to
RETURNS
: ABrowserApi
object that is used to make the requests to the PhantomJs Cloud.
browser.requestSingle()
browser.requestSingle(request, customOptions?, callback?) : Promise<IUserResponse>
For making a single request.
request
: Either be aIPageRequest
orIUserRequest
object. See https://phantomjscloud.com/docs/http-api/ for full details. The request default values can be seen here.customOptions
: Optional. can override the options set in theBrowserApi
class constructor.callback
: Optional. For people who don't use promises. If you use this, the function should have the signature(err: Error, result: IUserResponse) => void
RETURNS
: A Promise returning aIUserResponse
.
browser.requestBatch()
browser.requestBatch(requests, customOptions?, callback? ) : Promise<IUserResponse>[]
Submit multiple requests at the same time, and get an array of promises back.
requests
: An array. Each element should be either be aIPageRequest
orIUserRequest
object.customOptions
: Optional. can override the options set in theBrowserApi
class constructor.callback
: Optional. For people who don't use promises. If you use this, the function should have the signature(err: Error, item: {request, result}) => void
This will be called once for each request sent.RETURNS
: An array of Promises. Use a construct likebluebird.all()
to wait for all to finish if you wish.
Typescript Typings
If you use Visual Studio or VSCode the IntelliSense will automatically load when you: import phantomjscloud = require("phantomjscloud");
You do not need to load anything from the DefinitelyTyped nor Typings projects.
Examples
Here are some basic examples. Look at the phantomjscloud-node-examples project on github for more.
Capture Amazon.com as a PDF
Basic Javascript
var pageRequest = { url: "https://amazon.com", renderType: "pdf" };
console.log("about to request page from PhantomJs Cloud. request =", JSON.stringify(pageRequest, null, "\t"));
browser.requestSingle(pageRequest, (err, userResponse) => {
if (userResponse.statusCode != 200) {
throw new Error("invalid status code" + userResponse.statusCode);
}
fs.writeFile(userResponse.content.name, userResponse.content.data,
{
encoding: userResponse.content.encoding,
}, (err) => {
console.log("captured page written to " + userResponse.content.name);
});
});
Typescript with Promises
//the page you wish to render
let pageRequest: phantomJsCloud.ioDatatypes.IPageRequest = { url: "https://amazon.com", renderType: "pdf" };
console.log("about to request page from PhantomJs Cloud. request =", JSON.stringify(pageRequest, null, "\t"));
browser.requestSingle(pageRequest)
.then((userResponse) => {
if (userResponse.statusCode != 200) {
throw new Error("invalid status code" + userResponse.statusCode);
}
fs.writeFile(userResponse.content.name, userResponse.content.data,
{
encoding: userResponse.content.encoding,
}, (err) => {
console.log("captured page written to " + userResponse.content.name);
});
});
All Parameters
Shows using all parameters in a request, capturing the page as a .jpg
image. Most the parameters used are the defaults, but you can see a list of the most up-to-date default values here.
Typescript with Promises
//the page you wish to render
let userRequest: phantomJsCloud.ioDatatypes.IUserRequest = {
pages:[
{
"url": "http://example.com",
"content": null,
"urlSettings": {
"operation": "GET",
"encoding": "utf8",
"headers": {},
"data": null
},
renderType: 'jpg',
outputAsJson: false,
requestSettings: {
ignoreImages: false,
disableJavascript: false,
userAgent: 'Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/534.34 (KHTML, like Gecko) Safari/534.34 PhantomJS/2.0.0 (PhantomJsCloud.com/2.0.1)',
xssAuditingEnabled: false,
webSecurityEnabled: false,
resourceWait: 15000,
resourceTimeout: 35000,
maxWait: 35000,
ioWait: 2000,
waitInterval: 1000,
stopOnError: false,
resourceModifier: [],
customHeaders: {},
clearCache: false,
clearCookies: false,
cookies: [],
deleteCookies: [],
},
suppressJson: [
'events.value.resourceRequest.headers',
'events.value.resourceResponse.headers',
'frameData.content',
'frameData.childFrames',
],
renderSettings: {
quality: 70,
viewport: {
height: 1280,
width: 1280,
},
zoomFactor: 1,
passThroughHeaders: false,
emulateMedia: 'screen',
omitBackground: false,
passThroughStatusCode: false,
},
scripts: {
pageNavigated: [],
load: [],
domReady: [],
loadFinished: [],
},
scriptSettings: {
stopOnError: false,
async: false,
},
}
],
proxy:false
};
console.log("about to request page from PhantomJs Cloud. request =", JSON.stringify(userRequest, null, "\t"));
browser.requestSingle(userRequest)
.then((userResponse) => {
if (userResponse.statusCode != 200) {
throw new Error("invalid status code" + userResponse.statusCode);
}
fs.writeFile(userResponse.content.name, userResponse.content.data,
{
encoding: userResponse.content.encoding,
}, (err) => {
console.log("captured page written to " + userResponse.content.name);
});
});
Technical details
Internally this library will pool all requests and execute in a FIFO fashion. The number of parallel requests increases automatically: We gracefully ramp-up the rate of requests to match PhantomJsCloud's autoscale capacity.
Roadmap
- web crawler
- batch api framework (store results to S3, Google Cloud Storage, etc)
- page automation helpers (button clicking, etc. you can do this already, but you need to know a lot to do it...)
Contributing
This is the official, reference API for PhantomJsCloud.com You can help out by writing API Client Libraries for other languages (lots of requests for PHP and Python!)
Building
If you want to build this yourself:
- install
npm install mocha -g
- install dependencies
npm install
- Install vscode and run the project (see
.vscode/launch.json
for actual execution args)