Merge pull request #11 from PredicSis/tp-puppeteer

[master] Move from PhantomJS to Puppeteer

Dockerfile

@@ -1,14 +1,16 @@
-FROM mhart/alpine-node:5.10
-MAINTAINER Gregoire MORPAIN <gm@predicsis.com>
+FROM node:8
+MAINTAINER Gregoire MORPAIN <gm@predicsis.com>, mdouchement
 
 ENV NODE_ENV production
+ENV DOCKER_EXECUTION 1
 
 RUN mkdir -p /usr/src/app
 WORKDIR /usr/src/app
 
+ARG chromium_libs='gconf-service libasound2 libatk1.0-0 libc6 libcairo2 libcups2 libdbus-1-3 libexpat1 libfontconfig1 libgcc1 libgconf-2-4 libgdk-pixbuf2.0-0 libglib2.0-0 libgtk-3-0 libnspr4 libpango-1.0-0 libpangocairo-1.0-0 libstdc++6 libx11-6 libx11-xcb1 libxcb1 libxcomposite1 libxcursor1 libxdamage1 libxext6 libxfixes3 libxi6 libxrandr2 libxrender1 libxss1 libxtst6 ca-certificates fonts-liberation libappindicator1 libnss3 lsb-release xdg-utils'
+RUN apt-get update && apt-get install -y git ${chromium_libs} --no-install-recommends && rm -rf /var/lib/apt/lists/*
+
 COPY package.json /usr/src/app/
-RUN apk update && apk add git
-RUN curl -Ls "https://github.com/dustinblackman/phantomized/releases/download/2.1.1/dockerized-phantomjs.tar.gz" | tar xz -C /
 RUN npm install
 COPY . /usr/src/app/
 

README.md

@@ -1,29 +1,20 @@
 # html2pdf
 
-PDF screenshot API based on PhantomJS
+PDF screenshot API based on Puppeteer.
 
 ##### Node version
-version >= `4.0.0`
+version >= `8.0.0`
 
 ## Usage
 
 ```bash
 $ NODE_ENV=production npm install
 $ NODE_ENV=production PORT=4005 npm start
-```
 
-```
-Open http://localhost:4005/export?url=http://www.google.com&filename=google.pdf&format=A4&orientation=portrait
+$ DEBUG=1 NODE_ENV=production PORT=4005 npm start
 ```
 
-### Usage with SPA
 
-```js
-// on page_load
-window.phantomWaitForCallback = true;
-
-// when you are done fetching your api and rendering the page
-if (typeof window.callPhantom === 'function') {
- window.callPhantom({ event: 'loaded' });
-}
+```
+Open http://localhost:4005/export?url=http://www.google.com&filename=google.pdf&format=A4&orientation=portrait
 ```

index.js

@@ -1,26 +1,107 @@
-var express = require('express');
-var generatePDF = require('./pdf/export');
+var express = require('express')
+var fs = require('fs')
+var tmp = require('tmp')
+const puppeteer = require('puppeteer')
 
-var app = express();
-app.set('port', process.env.PORT || 4005);
+// Global
+var debug = (process.env.DEBUG === '1')
+tmp.setGracefulCleanup()
 
-app.get('/export', function (req, res) {
-  console.log('generating pdf with params :', req.query);
+var browserOpts = {}
+if (process.env.DOCKER_EXECUTION === '1') {
+  browserOpts = {
+    executablePath: process.env.CHROMIUM_BIN,
+    args: ['--no-sandbox', '--headless', '--disable-gpu', '--disable-dev-shm-usage']
+  }
+}
+
+// PDF
+
+async function _generatePDF(url, orientation, format, tmpFile) {
+  const browser = await puppeteer.launch(browserOpts)
+  try {
+    const page = await browser.newPage()
+
+    // From the launched Chromium console. Useful for debug.
+    if (debug) {
+      page.on('console', msg => {
+        console.log('--CONSOLE------------------')
+        for (let i = 0; i < msg.args.length; ++i)
+        console.log(`${i}: ${msg.args[i]}`)
+        console.log('___________________________')
+      })
+    }
+
+    page.on('pageerror', errmsg => {
+      console.log('[pageerror] ' + errmsg)
+
+      throw new Error(errmsg)
+    })
+
+    page.on('error', err => {
+      console.log('[ERROR] ' + err.message)
+      console.log(err.stack)
+
+      throw err
+    })
+
+    await page.goto(url, {waitUntil: 'networkidle2'})
+    await new Promise(resolve => setTimeout(resolve, 1000)) // wait for CSS animations
+
+    // https://github.com/GoogleChrome/puppeteer/blob/master/docs/api.md#pagepdfoptions
+    await page.pdf({
+      path: tmpFile,
+      format: format,
+      landscape: orientation === 'landscape'
+    })
+  } finally {
+    browser.close()
+  }
+}
+
+var generatePDF = (url, orientation, format) => {
+  return new Promise((resolve, reject) => {
+    tmp.file((err, tmpFile) => {
+      if (err) {
+        return reject(err)
+      }
+
+      _generatePDF(url, orientation, format, tmpFile)
+        .then((data) => {
+          fs.access(tmpFile, fs.F_OK, (err) => {
+            if (err) {
+              return reject(err)
+            }
+            return resolve(fs.createReadStream(tmpFile))
+          })
+        })
+        .catch(err => reject(err))
+    })
+  })
+}
+
+// Server
+
+var app = express()
+app.set('port', process.env.PORT || 4005)
+
+app.get('/export', (req, res) => {
+  console.log('generating pdf with params :', req.query)
   if (!req.query || !req.query.url) {
-    return res.status(404).end('missing url param');
+    return res.status(404).end('missing url param')
   }
   generatePDF(req.query.url, req.query.orientation, req.query.format)
-    .then(function(stream) {
-      res.attachment(req.query.filename);
-      stream.pipe(res);
-      console.log('generated pdf with params :');
-      console.log(req.query);
+    .then(stream => {
+      res.attachment(req.query.filename)
+      stream.pipe(res)
+      console.log('generated pdf with params :')
+      console.log(req.query)
     })
-    .catch(function(err) {
-      console.error(err);
-      res.status(500).json(err);
-    });
-});
-
-app.listen(app.get('port'));
-console.info('listening on port: ' + app.get('port'));
+    .catch(err => {
+      console.error(err)
+      res.status(500).json(err)
+    })
+})
+
+app.listen(app.get('port'))
+console.info('listening on port: ' + app.get('port'))