http://axe.g0v.tw/level/2 crawer result test in http://jsonlint.com/

axe2.js

var http = require("http");
var cheerio = require("cheerio");

/**
 * Utility function that downloads a URL and invokes
 * callback with the data.
 */
function download(url, callback) {
  http.get(url, function(res) {
    var data = "";
    res.on('data', function (chunk) {
      data += chunk;
    });
    res.on("end", function() {
      callback(data);
    });
  }).on("error", function() {
    callback(null);
  });
}


// DO THE JOB
var url = "http://axe-level-1.herokuapp.com/lv2/";

var resultJson = [];
var column_title = ["town", "village", "name"];

download(url, function(data) {
  if (data) {
    // console.log(data);
    var $ = cheerio.load(data);
    $("a").each(function(i, e) {
      var href = $(e).attr("href");
      href = 'http://axe-level-1.herokuapp.com/lv2/' + href;
      console.log('parse ' + href);
      parse(href);
    });
    // console.log(JSON.stringify(resultJson));
  }
});

function parse(path) {
  download(path, function(data) {
    if (data) {
      // console.log(data);
      var $ = cheerio.load(data);
      $("tr").each(function(i, e) {
      	var columns = $(e).find("td");
      	// column holder
      	var person = {};
      	columns.each(function(j, e) {
		      person[column_title[j]] = $(e).text();
      	});
      	resultJson.push(person);
      });
    }
    console.log("=======");
    console.log(JSON.stringify(resultJson));
  });
}