DeadCrawl.js

SEO+javascript == Undead (An Express.js middleware)

Renders the page with Zombie.js and saves to file.

Install

npm install dead-crawl

Example

var app = require('express')();
var deadCrawl = require('dead-crawl').deadCrawl;

app.use(deadCrawl());
app.use(routes);
app.use(express.static(...));

You may alter your hashbang delimiter by passing it in as an option.

app.use(deadCrawl({hashbang: "#"}));

Will rebuild the url with that vs. the default #!.
eg. http://example.com?_escaped_fragment_=/js/page => http://example.com/#/js/page

null or '' will omit the hashbang entirely.

To change the destination root of the saved html files, supply a destRoot option.

app.use(deadCrawl({destRoot: __dirname+'/public/crawled'}));

The default is destRoot is .;

To run process before the html gets written to file you can supply a beforeWrite option with an Array of functions.

function waitForMetaDescription(browser, _, next) {
  var i = 0;
  var waiting = setInterval(function() {
    var desc = browser.query('meta[name="description"]')
      .attributes
      .content
      ._nodeValue;

    i++;
    if (!!desc || i > 10) {
      next();
    }
  }, 100);
}

function removeNgApp(browser, next) {
  next(browser.html().replace(/\sng\-app="\w+"/, ''));
}

var opts = {
  destRoot: __dirname+'/public/crawls',
  beforeWriter: [
    waitForMetaDescription,
    removeNgApp
  ]
};

app.use(deadCrawl(opts));

The first function will always be provided with 3 arguments: browser, null, next.

The last function (before the writer) needs to next any processed html (or string) to be written to file. Else browser.html() will be written to file.

Information on using next to pass arguments can be viewed at WalkingDead.js.

Once a page has been crawled, any further requests (?_escaped_fragment_=) will send back the saved html file. If your page is dynamic you will have to ensure you sweep out the cached files when needed.

Notes

http://www.yearofmoo.com/2012/11/angularjs-and-seo.html

License

MIT