How can I run patch monkey in async mode?

[Long Nguyen]

I want to implement simple html scraper into ql.io. My idea is when you write query like code below, ql.io will fetch the page, create browser like environment (JSDom), include jQuery, execute query and fetch result. But my problem is JSDom run in async mode but ql.io 's patch monkey run in sync mode, so ql.io return result before JSDom finish running. In example below, result always is blank array. Any chance to fix that?

https://github.com/ql-io/ql.io/blob/3db25d6a673ec1cb47ad718390c94d485d51a1a5/modules/engine/lib/engine/http.request.js#L414

Query:

select * from www.scraper
  where url = "http://abcdef.com/xyz"

      and title = '[{"selector":"ul li a.title"}]'

      and url = '[{"selector":"ul li a.title", "prop": "href"}]';

www.scraper.ql

-- definition of www.scraper table

create table www.scraper

  on select get from "{url}"

  using patch 'www.scraper.js'

  resultset 'results'

www.scraper.js

var jsdom = require('jsdom');
exports['parse response'] = function(args) {
    var result = [];
    // get body

    var body = '';

    _.each(args.body, function(buf) {

         body += buf.toString('UTF-8');

    });

    // create JSDom environment, include jQuery
    jsdom.env({
         html: body,
         scripts: ['http://ajax.googleapis.com/ajax/libs/jquery/1.8.3/jquery.min.js'],
         done: function(err, window) {
            // run CSS query using jQuery
            // jQuery(selector).each(function() { result.push($(element).text()) });
         }

    });

    // return result as new response
    return {
       type: 'application/json',
       content: JSON.stringify(result)
    }
}

[shimonchayim]

Hi Nguyen,

Right now it doesn't but something we are thinking about. Please do file an issue.

- Cylus