Created
November 23, 2015 10:40
-
-
Save eknowles/f5da0f90d784a1603a81 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
var request = require('request'); | |
var cheerio = require('cheerio'); | |
var csv = require('ya-csv'); | |
var fs = require('fs'); | |
// npm i ya-csv cheerio request | |
url = 'http://www.bbc.co.uk/corporate2/insidethebbc/managementstructure/biographies'; | |
var writer = csv.createCsvStreamWriter(fs.createWriteStream('salaries.csv')); | |
writer.writeRecord(['name', 'title', 'salary', 'remuneration', 'link']); | |
request(url, function (error, response, html) { | |
if (!error) { | |
var $ = cheerio.load(html); | |
var people = []; | |
$('li', '.text').filter(function () { | |
var data = $(this); | |
var person = {name: '', title: '', salary: 0, remuneration: 0, link: ''}; | |
var link = data.find('a'); | |
person.name = data.text().split(',')[0]; | |
person.title = data.text().replace(person.name + ', ', ''); | |
person.link = link.attr('href'); | |
if (person.name === 'John Shield') { | |
person.link = 'http://www.bbc.co.uk/aboutthebbc/insidethebbc/managementstructure/biographies/shield_john.html'; | |
} | |
if (person.link.substring(0, 20) !== 'http://www.bbc.co.uk') { | |
person.link = 'http://www.bbc.co.uk' + person.link; | |
} | |
people.push(person); | |
}); | |
people.forEach(function (person, index, array) { | |
request(person.link, function (error, response, html) { | |
if (!error) { | |
var $ = cheerio.load(html); | |
var salary = $('#heading-salary-and-total-remuneration').next().children().contents(); | |
salary.each(function (i, elem) { | |
var d = elem.data; | |
var salary_re = /Salary\: £([0-9]*)\,([0-9]*)/; | |
var remuneration_re = /Total remuneration\: £([0-9]*)\,([0-9]*)/; | |
if (salary_re.test(d)) { | |
person.salary = parseInt(d.replace(salary_re, '$1$2')); | |
} | |
else if (remuneration_re.test(d)) { | |
person.remuneration = parseInt(d.replace(remuneration_re, '$1$2')); | |
} | |
}); | |
writer.writeRecord([person.name, person.title, person.salary, person.remuneration, person.link]); | |
} | |
}); | |
}); | |
} | |
}); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment