Prototype readability server

master
Tanner Collin 5 years ago
parent a814411c12
commit c0607b3fb6
  1. 48
      readerserver/main.js
  2. 13
      readerserver/package.json
  3. 1013
      readerserver/yarn.lock

@ -0,0 +1,48 @@
const express = require('express');
const app = express();
const port = 3000;
const request = require('request');
const JSDOM = require('jsdom').JSDOM;
const createDOMPurify = require('dompurify');
const Readability = require('readability');
app.use(express.urlencoded());
app.get('/', (req, res) => {
res.send('POST a URL to urlencoded \'url\' to parse.');
});
const requestCallback = (url, res) => (error, response, body) => {
if (!error && response.statusCode == 200) {
console.log('Response OK.');
const doc = new JSDOM('', {url: url});
const DOMPurify = createDOMPurify(doc.window);
const clean = DOMPurify.sanitize(body);
const cleanDoc = new JSDOM(clean, {url: url});
const reader = new Readability(cleanDoc.window.document);
const article = reader.parse();
res.send(article);
} else {
console.log('Response error:', error ? error.toString() : response.statusCode);
res.sendStatus(response ? response.statusCode : 404);
}
};
app.post('/', (req, res) => {
const url = req.body.url;
const requestOptions = {
url: url,
headers: {'User-Agent': 'Googlebot/2.1 (+http://www.google.com/bot.html)'},
};
console.log('Parse request for:', url);
request(requestOptions, requestCallback(url, res));
});
app.listen(port, () => {
console.log(`Example app listening on port ${port}!`);
});

@ -0,0 +1,13 @@
{
"name": "readerserver",
"version": "1.0.0",
"main": "main.js",
"license": "MIT",
"dependencies": {
"dompurify": "^1.0.11",
"express": "^4.17.1",
"jsdom": "^15.1.1",
"readability": "https://github.com/mozilla/readability",
"request": "^2.88.0"
}
}

File diff suppressed because it is too large Load Diff
Loading…
Cancel
Save