pdf.js

// NB: I'd normally deploy this using CloudFormation
/*
    The lambda needs the following configuration:
    
    - A bunch of memory probably 2GB
    - Better to have a long timeout just in case
    - Must include the following later:
    - nodejs10.x (or a newer layer - but haven't tested)
    - BUCKET_NAME environment variable
    - Permissions to access BUCKET_NAME
*/

const chromium = require('chrome-aws-lambda');
const puppeteer = chromium.puppeteer;
const { S3 } = require('aws-sdk');

const s3 = new S3();

exports.handler = async event => {
    console.log(event.url);
    const buffer = await toPdf(event.url);
    console.log('got buffer');
    await s3
        .putObject({
            Bucket: process.env.BUCKET_NAME,
            Key: 'image.png', // todo: determine the key dynamically, based on the provided URL
            Body: buffer
        })
        .promise();
    console.log('wrote to s3');
    // TODO implement
    const response = {
        statusCode: 200,
        body: JSON.stringify('Hello from Lambda!')
    };
    return response;
};

('use strict');
async function toPdf(url) {
    let browser;
    let buffer;
    let error;
    try {
        browser = await puppeteer.launch({
            args: chromium.args,
            defaultViewport: chromium.defaultViewport,
            executablePath: await chromium.executablePath,
            headless: true
        });
        const page = await browser.newPage();
        await page.goto(url, { waitUntil: 'networkidle0' });
        buffer = await page.pdf({ format: 'A4' });
    } catch (err) {
        error = err;
    } finally {
        if (browser) {
            await browser.close();
        }
    }
    if (error) {
        throw error;
    }
    if (!buffer) {
        throw new Error('Oh no! Did not get the pdf!');
    }
    return buffer;
}