Use headless chrome to intercept image request data - node.js

I have a use case that needs to use Headless Chrome Network (https://chromedevtools.github.io/devtools-protocol/tot/Network/) to intercept all images requests and find out the image size before saving it (basically discard small images such as icons).
However, I am unable to figure out a way to load the image data in memory before saving it. I need to load it in Img object to get width and height. The Network.getResponseBody is taking requestId which I don't have access in Network.requestIntercepted. Also Network.loadingFinished always gives me "0" in encodedDataLength variable. I have no idea why. So my questions are:
How to intercept all responses from jpg/png request and get the image data? Without saving the file via URL string to the disk and load back.
BEST: how to get image dimension from header response? Then I don't have to read the data into memory at all.
My code is below:
const chromeLauncher = require('chrome-launcher');
const CDP = require('chrome-remote-interface');
const file = require('fs');
(async function() {
async function launchChrome() {
return await chromeLauncher.launch({
chromeFlags: [
'--disable-gpu',
'--headless'
]
});
}
const chrome = await launchChrome();
const protocol = await CDP({
port: chrome.port
});
const {
DOM,
Network,
Page,
Emulation,
Runtime
} = protocol;
await Promise.all([Network.enable(), Page.enable(), Runtime.enable(), DOM.enable()]);
await Network.setRequestInterceptionEnabled({enabled: true});
Network.requestIntercepted(({interceptionId, request, resourceType}) => {
if ((request.url.indexOf('.jpg') >= 0) || (request.url.indexOf('.png') >= 0)) {
console.log(JSON.stringify(request));
console.log(resourceType);
if (request.url.indexOf("/unspecified.jpg") >= 0) {
console.log("FOUND unspecified.jpg");
console.log(JSON.stringify(interceptionId));
// console.log(JSON.stringify(Network.getResponseBody(interceptionId)));
}
}
Network.continueInterceptedRequest({interceptionId});
});
Network.loadingFinished(({requestId, timestamp, encodedDataLength}) => {
console.log(requestId);
console.log(timestamp);
console.log(encodedDataLength);
});
Page.navigate({
url: 'https://www.yahoo.com/'
});
Page.loadEventFired(async() => {
protocol.close();
chrome.kill();
});
})();

This should get you 90% of the way there. It gets the body of each image request. You'd still need to base64decode, check size and save etc...
const CDP = require('chrome-remote-interface');
const sizeThreshold = 1024;
async function run() {
try {
var client = await CDP();
const { Network, Page } = client;
// enable events
await Promise.all([Network.enable(), Page.enable()]);
// commands
const _url = "https://google.co.za";
let _pics = [];
Network.responseReceived(async ({requestId, response}) => {
let url = response ? response.url : null;
if ((url.indexOf('.jpg') >= 0) || (url.indexOf('.png') >= 0)) {
const {body, base64Encoded} = await Network.getResponseBody({ requestId }); // throws promise error returning null/undefined so can't destructure. Must be different in inspect shell to app?
_pics.push({ url, body, base64Encoded });
console.log(url, body, base64Encoded);
}
});
await Page.navigate({ url: _url });
await sleep(5000);
// TODO: process _pics - base64Encoded, check body.length > sizeThreshold, save etc...
} catch (err) {
if (err.message && err.message === "No inspectable targets") {
console.error("Either chrome isn't running or you already have another app connected to chrome - e.g. `chrome-remote-interface inspect`")
} else {
console.error(err);
}
} finally {
if (client) {
await client.close();
}
}
}
function sleep(miliseconds = 1000) {
if (miliseconds == 0)
return Promise.resolve();
return new Promise(resolve => setTimeout(() => resolve(), miliseconds))
}
run();

Related

FETCH for uploading folder with multiple files DOES NOT ABORT when I close the Tab

I'm trying to implement a multiple file loader in my web app, i.e. I want to upload to a given path in my local machine a folder with multiple files in it.
I'm using Svelte for the frontend part and Express for the backend part.
So far I've done the following. Client side:
async function load_report() {
let first_input = document.getElementById("folder");
let files_list = Array.from(first_input.files);
let local_files_paths = files_list.map((f) => f.webkitRelativePath);
const payload = new FormData();
for (let i = 0; i < files_list.length; i++) {
payload.append("files", files_list[i]);
payload.append("local_files_paths", local_files_paths[i]);
}
try {
isUploading = true; // to disable the buttons for other uploading operations
const response = await fetch("/upload_report", {
//signal: signal,
method: "POST", // or "PUT"
body: payload,
// No content-type! With FormData obect, Fetch API sets this automatically.
// Doing so manually can lead to an error
});
isUploading = false;
let fetch_msg = await response.json();
if (!response.ok) {
upload_report_error = `${fetch_msg["title"]}: ${fetch_msg["message"]}`;
alert.addAlertToStore(
"danger",
fetch_msg["title"],
fetch_msg["message"]
);
console.error(fetch_msg["message"]);
} else {
alert.addAlertToStore(
"success",
fetch_msg["title"],
fetch_msg["message"]
);
}
} catch (e) {
console.error(e);
} finally {
//$alert = `Todo '${name}' has been added`;
first_input.value = null;
}
}
and also in my onMount() function:
onMount(() => {
window.onbeforeunload = function () {
if (isUploading) {
return "You are uploading! CHILL OUT!";
}
};
Server-side, the endpoint is treated in this way:
app.post("/upload_report", async(req, res) => {
if (!req.files) {
return res.status(400).send({ title: 'upload_folder_error', message: 'No files were uploaded.'});
}
let date;
try{
date = req.body.local_files_paths[0].match(/([0-9_-]+)\//)[1];
}catch(e){
return res.status(500).send({ title: 'upload_folder_error', message: e.message });
}
if(fs.existsSync(`./app/${date}/folder`)){
return res.status(500).send({ title: 'upload_folder_error', message: 'this folder already exists' });
}
for(let i= 0; i< req.body.local_files_paths.length; i++){
let _file = req.files.files[i];
//move photo to uploads directory
_file.mv(`*some_path*`);
}
console.log('DONE');
return res.status(200).send({ title: 'upload_folder_success', message: 'Folder correctly uploaded' });
});
My problem is that when I try to close the tab, a pop up alert appears telling me If I really want to close the tab, and I can either confirm or cancel the operation. But, when I confirm the operation, I would expect the fetch to be consequentially aborted (correct me if I'm wrong) but instead it goes on server side (i.e. I can see my server logging "DONE") and the files are correctly uploaded... what I'm missing in that?
PS. I also tried to use the AbortController() and calling abort inside the onbeforeunload event, but that seemed to me to be wrong.
Could you please help me to figure it out? Thanks!

Puppeteer create PDF files from HTML data hangs Windows 10 system

I created an App that processes students results by extracting data from multiple excel workbooks. The problem is that using Puppeteer to generate the PDF files, throws the system into a loop till it hangs the system.
Actually, I have tested same codes below using PhantomJs which is bundled as pdf-creator-node, and was able to generate 150 PDF files comfortably in 3 minutes. The only challenge I dumped PhantomJs is that all the styling in the CSS file was not included, even when I inserted it as an inline style in the header, suing replace function of JS. Another, is that PhantomJs is no longer in active development. I searched the web, and found out that only Puppeteer is the valid solution with active development and support too.
I tried using page.close() at the end of pdfCreator() which is in a loop, and browser.close() at the end of pdfGenerator(). What I am doing wrong?
Here below are the codes in the server.js and PdfGenerator.js files, with a sample of the ERROR, and screenshot of my Task Manager after the system crawled out of hanging state. For HTML generation, I used Mustache. I excluded some lines of codes in server.js because the total character count was over 60k.
server.js
// [codes were removed here]
if(getCode == 'compute-result') {
// declare variable
let setData = null;
let setTitle = 'Results Computation...';
let setArgs = getArgs;
// dataFromFile = ReadFile(pathCodeTextFile);
// setArgs = Number(dataFromFile);
setCode = 'compute-result';
let setView = [];
let setNext = true;
let countTerms = [];
// if(getArg > 0) {
// Final Result computation
const getJson = ReadFile(pathJsonResults);
// const getCtrl = ReadFile(pathJsonCtrl);
const getResultObject = JSON.parse(getJson);
getResult = getResultObject;
const totalResults = getResult.firstTerm.length + getResult.secondTerm.length + getResult.thirdTerm.length;
if(setView.length < 1 && getResult != null) {
setData = 'PDFs for Students Results initiating...';
setView.unshift('Reading saved data...');
client.emit('query', {data: setData, title: setTitle, code: setCode, next: setNext, args: null, view: JSON.stringify(setView)});
}
Sleep(2000).then(() => {
if(getResult != null) {
setData = 'Students Results will be ready in a moment';
client.emit('query', {data: setData, title: setTitle, code: setCode, next: setNext, args: setArgs, view: JSON.stringify(setView)});
}
const wacthFiles = (file, className, termName, sessionName, completed, pdfList) => {
try {
if(typeof file == 'string' && !FileExists(pathJsonPdfList)) {
if(pdfList.length < 2){
setData = 'Saving PDFs to downladable files...';
}
if(className != null && termName != null && sessionName != null) {
setTitle = `${pdfList.length} Result PDF${pdfList.length > 1?'s':''}...`;
setView.unshift(file);
if(!countTerms.includes(termName)) {
countTerms.push(termName)
}
// setCode = -1000 - pdfList.length;
// console.log('PDF PROGRESS: ', `${pdfList.length} Result PDF${pdfList.length > 1?'s':''}... ${setCode}`);
// when all PDFs are created
if(completed) {
setTitle = setTitle.replace('...', ' [completed]');
setData = 'Result Download button is Active. You may click it now.';
setView.unshift('=== PDF GENERATION COMPLETED ===');
setView.unshift(`A total of ${pdfList.length} students' Results were generated`);
WriteFile(pathJsonPdfList, JSON.stringify(pdfList));
// set donwload button active
setCode = Number(codeTextFilePdfCompleted);
setNext = false;
getResult = null;
let termString = countTerms.toString();
termString = ReplaceAll(termString, '-term', '');
termString = ReplaceAll(termString, ',', '-');
const addTxt = `${className} _${termString} Term${countTerms.length>1?'s':''} (${sessionName})`;
WriteFile(pathCodeTextFile, addTxt);
// console.log('======== PDF GENERATION ENDS ================');
} else {
setCode = -1 * pdfList.length;
}
client.emit('query', {data: setData, title: setTitle, code: setCode, next: setNext, args: setArgs, view: JSON.stringify(setView)});
}
}
} catch (error) {
console.log('ERROR ON WATCHER: ', error);
}
}
if(!FileExists(pathJsonPdfList) && getResult !== null) {
PdfGenerator(getResult, wacthFiles);
}
// Watcher(pathWatchResults, setCode, wacthDir, 10000);
});
// }
}
}
} catch (error) {
})
client.on('disconnect', () => {
console.log('SERVER: Disconnected');
});
server.listen(portApi, () =>{
console.log('Server listens on port 8881')
});
// serve static files
app.use(express.static(pathPublic));
// [codes were removed here]
PdfGenerator.js
The problem lies in these functions: PdfGenerator & createPdf
'use strict';
process.setMaxListeners(Infinity) // fix for Puppeteer MaxListenerExceededWarning
const Puppeteer = require('puppeteer')
const {HtmlGenerator} = require('../components/HtmlGenerator')
const {WriteFile, FileExists, RandomNumber, RoundNumber, IsNumberFraction, ReadFile} = require('../components/Functions')
if (process.env.NODE_ENV !== 'production') {
require('dotenv').config();
}
const pathFirstTermResults = process.env.DIR_FIRST_TERM_RESULTS;
const pathSecondTermResults = process.env.DIR_SECOND_TERM_RESULTS;
const pathThirdTermResults = process.env.DIR_THIRD_TERM_RESULTS;
const publicDir = process.env.DIR_PUBLIC;
const cssFile = process.env.PATH_CSS_FILENAME;
const pathCssRaw = __dirname + '\\' + publicDir + '\\' + cssFile;
const pathCss = pathCssRaw.replace(`\\uploads`, '');
const tagCssReplace = process.env.TAG_CSS_REPLACE;
let jsonDir = process.env.PATH_JSON;
jsonDir = jsonDir.split('/').pop();
let htmlDir = process.env.DIR_HTML;
htmlDir = __dirname + '\\' + htmlDir.split('/').pop();
const htmlType1 = htmlDir + '\\' + process.env.HTML_TYPE1;
const htmlType2 = htmlDir + '\\' + process.env.HTML_TYPE2;
const htmlType3 = htmlDir + '\\' + process.env.HTML_TYPE3;
const pathJsonPdfList = './' + jsonDir + '/' + process.env.JSON_PDF_LIST_FILENAME;
const pathJsonPdfContent = __dirname + '\\' + jsonDir + '\\' + process.env.JSON_PDF_CONTENT;
const firstTermDir = 'first-term';
const secondTermDir = 'second-term';
const thirdTermDir = 'third-term';
let cumulativeFirstTermTotalList = {};
let cumulativeSecondTermTotalList = {};
let firstTermOnce = true;
let secondTermOnce = true;
let thirdTermOnce = true;
let isActive = false;
const getPath = (p, f) => {
let dir = pathFirstTermResults;
switch (p) {
case firstTermDir:
dir = pathFirstTermResults;
break;
case secondTermDir:
dir = pathSecondTermResults;
break;
case thirdTermDir:
dir = pathThirdTermResults;
break;
default:
break;
}
return dir + f
}
const resolution = {
x: 1920,
y: 1080
}
const args = [
'--disable-gpu',
`--window-size=${resolution.x},${resolution.y}`,
'--no-sandbox',
]
const createPdf = (page, content, templateType, filename, className, term, sessionName, isProcessActive, pdfFileList, cb) => {
let path, document, options;
path = getPath(term, filename);
if(path != null) {
let options = {
path: path,
format: 'A4',
printBackground: true,
margin: {
left: '0px',
top: '0px',
right: '0px',
bottom: '0px'
}
}
let templateData = '';
switch (templateType) {
case '1':
templateData = ReadFile(htmlType1);
break;
case '2':
templateData = ReadFile(htmlType2);
break;
case '3':
templateData = ReadFile(htmlType3);
break;
default:
templateData = ReadFile(htmlType1);
break;
}
(async() => {
const html = HtmlGenerator(content, templateData);
if(html != undefined && html !== '' && html != null) {
// create PDF file
cb(filename, className, term, sessionName, isProcessActive, pdfFileList);
// get style from .css & replace
const css = ReadFile(pathCss);
await page.setContent(html, { waitUntil: 'networkidle0'});
await page.addStyleTag(css);
await page.pdf(options);
page.close();
}
})()
}
}
const pdfGenerator = (json, cb) => {
let data = {};
let pdfFileList = [];
if(typeof json == 'string') {
data = JSON.parse(json)
} else {
data = json;
}
try {
// declare defaults
let filename = 'Student' + '.pdf';
let termName = firstTermDir;
const templateType = data.keys.templateType;
const session = data.classInfo.Session;
const sessionName = session.replace('/', '-');
const students = data.students;
const className = data.classInfo.Class_Name;
const recordFirstTerm = data.firstTerm;
const recordSecondTerm = data.secondTerm;
const recordThirdTerm = data.thirdTerm;
let pdfCreatedList = [];
let isReset = false;
let totalResultsExpected = Object.keys(recordFirstTerm).length + Object.keys(recordSecondTerm).length + Object.keys(recordThirdTerm).length;
let totalResultsCount = 0;
let jsonForPdf = {};
let record = {};
let sRecord, path, id, fName, lName;
// get each student
let logEndOnce = true;
let logBeforeOnce = true;
logBeforeOnce && console.log('============== *** ================');
logBeforeOnce && console.log('======== PDF GENERATION BEGINS ================');
const computeResult = (page, setTerm, setRecord, setReset) => {
const termName = setTerm;
const record = setRecord;
let isReset = setReset;
logBeforeOnce && console.log(`====== ${termName} RESULTS BEGINS ======`);
for(let elem of students){
id = elem.id;
fName = elem.firstName;
lName = elem.lastName;
filename = `${lName} ${fName} _${termName} ${sessionName}.pdf`;
// sRecord = record.filter(function (entry) { return entry[id] !== undefined; });
sRecord = record[id];
path = getPath(termName, filename);
// create pdf
if(!FileExists(path) && !FileExists(pathJsonPdfList)){
// generate final JSON for the student
// isReset = (pdfCreatedList.includes(id))? false: true;
jsonForPdf = finalJson(elem, sRecord, data, termName);
(pdfFileList.length < 1) && WriteFile(pathJsonPdfContent, JSON.stringify(jsonForPdf));
pdfFileList.push({
'term': termName,
'file': filename
});
totalResultsCount = pdfFileList.length;
const pdfDate = new Date();
console.log(`${filename} (${totalResultsCount}/${totalResultsExpected}) at ${pdfDate.getHours()}hr${pdfDate.getHours()>1?'s':''} - ${pdfDate.getMinutes()}min${pdfDate.getMinutes()>1?'s':''} - ${pdfDate.getSeconds()}sec${pdfDate.getSeconds()>1?'s':''}`);
isActive = (totalResultsExpected === totalResultsCount)? true: false;
logEndOnce = false;
// cb(filename, className, termName, sessionName, isActive, pdfFileList);
// WriteFile(path, null);
isReset = true;
createPdf(page, jsonForPdf, templateType, filename, className, termName, sessionName, isActive, pdfFileList, cb);
}
}
logBeforeOnce && console.log(`====== ${termName} RESULTS ENDS ======`);
}
// get each student result for First Term
const computeFirstTerm = (p) => {
return new Promise((resolve) => {
if(data.keys.firstTerm === '1') {
termName = firstTermDir;
record = recordFirstTerm;
pdfCreatedList = [];
isReset = false;
computeResult(p, termName, record, isReset)
}
resolve()
})
}
// get each student result for Second Term
const computeSecondTerm = (p) => {
return new Promise((resolve) => {
if(data.keys.secondTerm === '1') {
termName = secondTermDir;
record = recordSecondTerm;
pdfCreatedList = [];
isReset = false;
computeResult(p, termName, record, isReset)
}
resolve()
})
}
// get each student result for Third Term
const computeThirdTerm = (p) => {
return new Promise((resolve) => {
if(data.keys.thirdTerm === '1') {
termName = thirdTermDir;
record = recordThirdTerm;
pdfCreatedList = [];
isReset = false;
computeResult(p, termName, record, isReset)
}
resolve()
})
}
(async () => {
browser = await Puppeteer.launch({
headless: true,
handleSIGINT: false,
args: args,
});
const page = await browser.newPage();
await page.setViewport({
width: resolution.x,
height: resolution.y,
})
await computeFirstTerm(page);
await computeSecondTerm(page);
await computeThirdTerm(page);
browser.close()
})()
if(totalResultsExpected === totalResultsCount && totalResultsCount !== 0 && !logEndOnce) {
logEndOnce = true;
logBeforeOnce = false;
console.log('======== PDF GENERATION ENDS ================');
}
} catch (error) {
console.log('==== ERROR IN PDF GENERATION: ', error)
}
}
module.exports = {
PdfGenerator: pdfGenerator
}
ERROR
info Visit https://yarnpkg.com/en/docs/cli/run for documentation about this command.
lerna ERR! yarn run start stderr:
<--- Last few GCs --->
[9884:000002D68A73C6B0] 1665171 ms: Scavenge 44.1 (45.8) -> 43.2 (45.8) MB, 223.9 / 0.0 ms (average mu = 0.956, current mu = 0.952) allocation failure
[9884:000002D68A73C6B0] 1684089 ms: Scavenge 44.1 (45.8) -> 43.3 (45.8) MB, 587.3 / 0.0 ms (average mu = 0.956, current mu = 0.952) allocation failure
[9884:000002D68A73C6B0] 1749901 ms: Scavenge 44.2 (45.8) -> 43.3 (45.8) MB, 5099.0 / 0.0 ms (average mu = 0.956, current mu = 0.952) allocation failure
<--- JS stacktrace --->
FATAL ERROR: Committing semi space failed. Allocation failed - JavaScript heap out of memory
1: 00007FF6ED61013F
2: 00007FF6ED59F396
3: 00007FF6ED5A024D
4: 00007FF6EDED19EE
5: 00007FF6EDEBBECD
6: 00007FF6EDD5F61C
7: 00007FF6EDD6933F
8: 00007FF6EDD5BF19
9: 00007FF6EDD5A0D0
10: 00007FF6EDD7EA06
11: 00007FF6EDAB1CD5
12: 00007FF6EDF5F3E1
13: 00007FF6EDF602E9
14: 000002D68C4EF69E
error Command failed with exit code 134.
Screenshot of Task Manager, Chromium running multiple instances of over 50.
I appreciate any help. I hope this can be resolved to give me a smooth PDF generation.
Thank you.
Example solution (limiting parallel browsers)
I created you a PdfPrinter class which you can integrate into your setup. It allows you to limit the amount of parallel pdf generation jobs and allows setting a limit and manages opening/closing the browser for you. The PdfPrinter class is also highly coupled and needed some modification for using it as a general queue. Logicwise this can be modified to be a general queue.
You can try to integrate that into your code. This is a fully working test example with simplified pdfs (without the part of getting the actual data from the excel..)
As far as I understood your code, you do not need to pass the page around all your functions. First create your html + css and then use the pdfPrinter and let it handle page creation + browser launching..
(I like to code stuff like this so I went straight ahead..)
var puppeteer = require('puppeteer')
const defaultPrinterOptions = {
format: 'A4',
printBackground: true,
margin: {
left: '0px',
top: '0px',
right: '0px',
bottom: '0px'
}
}
class PdfPrinter {
maxBrowsers = 2
enqueuedPrintJobs = []
failedJobs = []
browserInstances = 0
// max browser instances in parallel
constructor(maxBrowsers) {
this.maxBrowsers = maxBrowsers
}
/**
*
* #param {*} html the html content to print
* #param {*} css to apply to the page
* #param {*} printOptions options passed to puppeteer
*/
// enqueues a print but the exact end moment cannot be known..
enqueuePrint = (html, css, path, done) => {
// merge custom options with defaultOptions..
const printOptions = {
...defaultPrinterOptions,
// add the path to the options.
path: path
}
// create a function which can be stored in an array
// it will later be grabbed by startPrinter() OR at the time any
// brwoser freed up..
// the function needs to be passed the actual used browser instance!
this.enqueuedPrintJobs.push(async(browser) => {
// catch the error which may be produced when printing something..
try {
// print the document
await this.print(browser, html, css, printOptions)
} catch (err) {
console.error('error when printing document..CLosing browser and starting a new job!!', printOptions.path)
console.error(err)
// store someting so you now what failed and coudl be retried or something..
this.failedJobs.push({ html, css, path: printOptions.path })
// puppeteer can run into erros too!!
// so close the browser and launch a new one!
await this.closeBrowser(browser)
browser = await this.launchBrowser()
}
// after the print, call done() so the promise is resovled in the right moment when
// this particular print has ended.!
done()
// start the next job right now if there are any left.
const job = this.enqueuedPrintJobs.shift()
if (!job) {
console.log('No print jobs available anymore. CLosing this browser instance.. Remaining browsers now:', this.maxBrowsers - this.browserInstances + 1)
await this.closeBrowser(browser)
return
}
// job is actually this function itself! It will be executed
// and automatically grab a new job after completion :)
// we pass the same browser instance to the next job!.
await job(browser)
})
// whenever a print job added make sure to start the printer
// this starts new browser instances if the limit is not exceeded resp. if no browser is instantiated yet,
// and does nothing if maximum browser count is reached..
this.tryStartPrinter()
}
// same as enqueuePrint except it wraps it in a promise so we can now the
// exact end moment and await it..
enqueuePrintPromise(html, css, path) {
return new Promise((resolve, reject) => {
try {
this.enqueuePrint(html, css, path, resolve)
} catch (err) {
console.error('unexpected error when setting up print job..', err)
reject(err)
}
})
}
// If browser instance limit is not reached will isntantiate a new one and run a print job with it.
// a print job will automatically grab a next job with the created browser if there are any left.
tryStartPrinter = async() => {
// Max browser count in use OR no jobs left.
if (this.browserInstances >= this.maxBrowsers || this.enqueuedPrintJobs.length === 0) {
return
}
// browser instances available!
// create a new one
console.log('launching new browser. Available after launch:', this.maxBrowsers - this.browserInstances - 1)
const browser = await this.launchBrowser()
// run job
const job = this.enqueuedPrintJobs.shift()
await job(browser)
}
closeBrowser = async(browser) => {
// decrement browsers in use!
// important to call before closing browser!!
this.browserInstances--
await browser.close()
}
launchBrowser = async() => {
// increment browsers in use!
// important to increase before actualy launching (async stuff..)
this.browserInstances++
// this code you have to adjust according your enviromnemt..
const browser = await puppeteer.launch({ headless: true })
return browser
}
// The actual print function which creates a pdf.
print = async(browser, html, css, printOptions) => {
console.log('Converting page to pdf. path:', printOptions.path)
// Run pdf creation in seperate page.
const page = await browser.newPage()
await page.setContent(html, { waitUntil: 'networkidle0' });
await page.addStyleTag({ content: css });
await page.pdf(printOptions);
await page.close();
}
}
// testing the PDFPrinter with some jobs.
// make sure to run the printer in an `async` function so u can
// use await...
const testPrinterQueue = async() => {
// config
const maxOpenedBrowsers = 5 // amount of browser instances which are allowed to be opened in parallel
const testJobCount = 100 // amount of test pdf jobs to be created
const destDir = 'C:\\somepath' // the directory to store the pdfs in..
// create sample jobs for testing...
const jobs = []
for (let i = 0; i < testJobCount; i++) {
jobs.push({
html: `<h1>job number [${i}]</h1>`,
css: 'h1 { background-color: red; }',
path: require('path').join(destDir, `pdf_${i}.pdf`)
})
}
// track time
const label = 'printed a total of ' + testJobCount + ' pdfs!'
console.time(label)
// run the actual pdf generation..
const printer = new PdfPrinter(maxOpenedBrowsers)
const jobProms = []
for (let job of jobs) {
// run jobs in parallel. Each job wil be runned async and return a Promise therefor
jobProms.push(
printer.enqueuePrintPromise(job.html, job.css, job.path)
)
}
console.log('All jobs enqueued!! Wating for finish now.')
// helper function which awaits all the print jobs, resp. an array of promises.
await Promise.all(jobProms)
console.timeEnd(label)
// failed jobs::
console.log('jobs failed:', printer.failedJobs)
// as file:
await require('fs').promises.writeFile('failed-jobs.json', JSON.stringify(printer.failedJobs))
}
testPrinterQueue().then(() => {
console.log('done with everyting..')
}).catch(err => {
console.error('unexpected error occured while printing all pages...', err)
})
You only need to adjust the destDir / openedBrowsers and testJobCount vars in the beginning of testPrinterQueue() for getting this to work.
What caused the problem in your code
Let's have a look at this piece
(async () => {
browser = await Puppeteer.launch({
headless: true,
handleSIGINT: false,
args: args,
});
const page = await browser.newPage();
await page.setViewport({
width: resolution.x,
height: resolution.y,
})
await computeFirstTerm(page);
await computeSecondTerm(page);
await computeThirdTerm(page);
browser.close()
})()
You created an anonymous function which is executed immediatly. Within the function all the statements are correctly awaited using await. But if you run this whole piece within a synchronious part of your application, the whole function will start immediatly but NOT been awaited before running next code.
Checkout this example:
//utility
function wait(ms) {
return new Promise(resolve => {
setTimeout(resolve, ms)
})
}
const AsyncFunction = async() => {
console.log('Async named function started')
// simulate execution time of 2 seconds
await wait(2000)
console.log('Async named function ended')
};
function SyncFunction() {
console.log('sync function started')
// example of async function execution within a sync function..
AsyncFunction();
// what you have done in your code:
(async() => {
console.log('Async anonymus function started')
await wait(3000)
console.log('Async anonymus function ended')
})()
// what
console.log('sync function ended.')
}
SyncFunction()
console.log('done')
Note the output:
Async named function started
Async anonymus function started
sync function ended. // => sync function already ended
done // sync function ended and code continues execution.
Async named function ended
Async anonymus function ended
To correctly await your async stuff you need to put your whole application in async scope:
//utility
function wait(ms) {
return new Promise(resolve => {
setTimeout(resolve, ms)
})
}
const AsyncFunction = async() => {
console.log('Async named function started')
// simulate execution time of 2 seconds
await wait(2000)
console.log('Async named function ended')
};
// this is now async!!
async function SyncFunction() {
console.log('sync function started')
// example of async function execution within a sync function..
await AsyncFunction();
// what you have done in your code:
await (async() => {
console.log('Async anonymus function started')
await wait(3000)
console.log('Async anonymus function ended')
})()
// what
console.log('sync function ended.')
}
SyncFunction().then(() => {
console.log('done')
}).catch(err => {
console.error('unexpected error occured..')
})
This output is what we want
sync function started
Async named function started
Async named function ended
Async anonymus function started
Async anonymus function ended
sync function ended.
done
Hope this helps you understand.
Feel free to leave a comment.

fs.writeFile crashes node app after writing first json file

I'm trying to crawl several web pages to check broken links and writing the results of the links to a json files, however, after the first file is completed the app crashes with no error popping up...
I'm using Puppeteer to crawl, Bluebird to run each link concurrently and fs to write the files.
WHAT IVE TRIED:
switching file type to '.txt' or '.php', this works but I need to create another loop outside the current workflow to convert the files from '.txt' to '.json'. Renaming the file right after writing to it also causes the app to crash.
using try catch statements for fs.writeFile but it never throws an error
the entire app outside of express, this worked at some point but i trying to use it within the framework
const express = require('express');
const router = express.Router();
const puppeteer = require('puppeteer');
const bluebird = require("bluebird");
const fs = require('fs');
router.get('/', function(req, res, next) {
(async () => {
// Our (multiple) URLs.
const urls = ['https://www.testing.com/allergy-test/', 'https://www.testing.com/genetic-testing/'];
const withBrowser = async (fn) => {
const browser = await puppeteer.launch();
try {
return await fn(browser);
} finally {
await browser.close();
}
}
const withPage = (browser) => async (fn) => {
const page = await browser.newPage();
// Turns request interceptor on.
await page.setRequestInterception(true);
// Ignore all the asset requests, just get the document.
page.on('request', request => {
if (request.resourceType() === 'document' ) {
request.continue();
} else {
request.abort();
}
});
try {
return await fn(page);
} finally {
await page.close();
}
}
const results = await withBrowser(async (browser) => {
return bluebird.map(urls, async (url) => {
return withPage(browser)(async (page) => {
await page.goto(url, {
waitUntil: 'domcontentloaded',
timeout: 0 // Removes timeout.
});
// Search for urls we want to "crawl".
const hrefs = await page.$$eval('a[href^="https://www.testing.com/"]', as => as.map(a => a.href));
// Predefine our arrays.
let links = [];
let redirect = [];
// Loops through each /goto/ url on page
for (const href of Object.entries(hrefs)) {
response = await page.goto(href[1], {
waitUntil: 'domcontentloaded',
timeout: 0 // Remove timeout.
});
const chain = response.request().redirectChain();
const link = {
'source_url': href[1],
'status': response.status(),
'final_url': response.url(),
'redirect_count': chain.length,
};
// Loops through the redirect chain for each href.
for ( const ch of chain) {
redirect = {
status: ch.response().status(),
url: ch.url(),
};
}
// Push all info of target link into links
links.push(link);
}
// JSONify the data.
const linksJson = JSON.stringify(links);
fileName = url.replace('https://www.testing.com/', '');
fileName = fileName.replace(/[^a-zA-Z0-9\-]/g, '');
// Write data to file in /tmp directory.
fs.writeFile(`./tmp/${fileName}.json`, linksJson, (err) => {
if (err) {
return console.log(err);
}
});
});
}, {concurrency: 4}); // How many pages to run at a time.
});
})();
});
module.exports = router;
UPDATE:
So there is nothing wrong with my code... I realized nodemon was stopping the process after each file was saved. Since nodemon would detect a "file change" it kept restarting my server after the first item

Click anywhere on page using Puppeteer

Currently I'm using Puppeteer to fetch cookies & headers from a page, however it's using a bot prevention system which is only bypassed when clicking on the page; I don't want to keep this sequential so it's "detectable"
How can I have my Puppeteer click anywhere on the page at random? regardless of wether it clicks a link, button etc..
I've currently got this code
const getCookies = async (state) => {
try {
state.browser = await launch_browser(state);
state.context = await state.browser.createIncognitoBrowserContext();
state.page = await state.context.newPage();
await state.page.authenticate({
username: proxies.username(),
password: proxies.password(),
});
await state.page.setViewport(functions.get_viewport());
state.page.on('response', response => handle_response(response, state));
await state.page.goto('https://www.website.com', {
waitUntil: 'networkidle0',
});
await state.page.waitFor('.unlockLink a', {
timeout: 5000
});
await state.page.click('.unlockLink a');
await state.page.waitFor('input[id="nondevice"]', {
timeout: 5000
});
state.publicIpv4Address = await state.page.evaluate(() => {
return sessionStorage.getItem("publicIpv4Address");
});
state.csrfToken = await state.page.evaluate(() => {
return sessionStorage.getItem("csrf-token");
});
//I NEED TO CLICK HERE! CAN BE WHITESPACE, LINK, IMAGE
state.browser_cookies = await state.page.cookies();
state.browser.close();
for (const cookie of state.browser_cookies) {
if(cookie.name === "dtPC") {
state.dtpc = cookie.value;
}
await state.jar.setCookie(
`${cookie.name}=${cookie.value}`,
'https://www.website.com'
)
}
return state;
} catch(error) {
if(state.browser) {
state.browser.close();
}
throw new Error(error);
}
};
The simplest way I can think of out of my head to choose a random element from DOM would be probably something like using querySelectorAll() which will return you an array of all <div>s in your document (or choose any other element, like <p> or anything else), then you can easily use click() on random one from the result, for example:
await page.evaluate(() => {
const allDivs = document.querySelectorAll('.left-sidebar-toggle');
const randomElement = allDivs[Math.floor(Math.random() * allDivs.length)];
randomElement.click();
});

Do node js worker never times out?

I have an iteration that can take up to hours to complete.
Example:
do{
//this is an api action
let response = await fetch_some_data;
// other database action
await perform_operation();
next = response.next;
}while(next);
I am assuming that the operation doesn't times out. But I don't know it exactly.
Any kind of explanation of nodejs satisfying this condition is highly appreciated. Thanks.
Update:
The actual development code is as under:
const Shopify = require('shopify-api-node');
const shopServices = require('../../../../services/shop_services/shop');
const { create } = require('../../../../controllers/products/Products');
exports.initiate = async (redis_client) => {
redis_client.lpop(['sync'], async function (err, reply) {
if (reply === null) {
console.log("Queue Empty");
return true;
}
let data = JSON.parse(reply),
shopservices = new shopServices(data),
shop_data = await shopservices.get()
.catch(error => {
console.log(error);
});
const shopify = new Shopify({
shopName: shop_data.name,
accessToken: shop_data.access_token,
apiVersion: '2020-04',
autoLimit: false,
timeout: 60 * 1000
});
let params = { limit: 250 };
do {
try {
let response = await shopify.product.list(params);
if (await create(response, shop_data)) {
console.log(`${data.current}`);
};
data.current += data.offset;
params = response.nextPageParameters;
} catch (error) {
console.log("here");
console.log(error);
params = false;
};
} while (params);
});
}
Everything is working fine till now. I am just making sure that the execution will ever happen in node or not. This function is call by a cron every minute, and data for processing is provided by queue data.

Resources