Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- /*
- * To change this license header, choose License Headers in Project Properties.
- * To change this template file, choose Tools | Templates
- * and open the template in the editor.
- */
- package jrex;
- import java.io.BufferedWriter;
- import java.io.File;
- import java.io.IOException;
- import java.nio.file.Files;
- import java.nio.file.Paths;
- import java.util.ArrayList;
- import java.util.List;
- import java.util.Map;
- import java.util.Spliterator;
- import java.util.concurrent.TimeUnit;
- import org.apache.commons.csv.CSVFormat;
- import org.apache.commons.csv.CSVPrinter;
- import org.apache.commons.csv.CSVRecord;
- import org.jsoup.Jsoup;
- import org.jsoup.nodes.Document;
- import org.openqa.selenium.By;
- import org.openqa.selenium.InvalidArgumentException;
- import org.openqa.selenium.WebDriver;
- import org.openqa.selenium.WebElement;
- import org.openqa.selenium.chrome.ChromeDriver;
- import org.openqa.selenium.support.ui.Wait;
- /**
- *
- * @author jonat
- */
- public class Jrex {
- /**
- * @param args the command line arguments
- */
- //TODO
- // Count # of requests
- // run during specific hours.
- private static final String companylinks = "./sample.csv";
- public static void main(String[] args) throws InterruptedException, IOException, InvalidArgumentException {
- //Set chrome webdriver
- System.setProperty("webdriver.chrome.driver", "C:\\Users\\jonat\\Documents\\Libraries\\chrome\\chromedriver.exe");
- // Open Browser
- WebDriver drive = new ChromeDriver();
- drive.get("https://www.linkedin.com/login");
- //Login to LinkedIn
- WebElement loginEmail = drive.findElement(By.id("username"));
- loginEmail.click();
- loginEmail.sendKeys("My- Username")
- WebElement loginPassword = drive.findElement(By.id("password"));
- loginPassword.click();
- loginPassword.sendKeys("My Password");
- WebElement loginButton = drive.findElement(By.className("login__form_action_container"));
- loginButton.click();
- drive.manage().timeouts().implicitlyWait(15, TimeUnit.SECONDS);
- //Waiting like a human
- double randomNumber = smallArms.getRandomNumber();
- System.out.println(randomNumber);
- //Thread.sleep((long) randomNumber);
- //drive.get(company_link_csv_reader.main().get(0));
- String urls = company_link_csv_reader.main().get(0);
- System.out.println(urls);
- System.out.println(urls.equals("www.linkedin.com/company/rex-recruiting/"));
- drive.navigate().to(urls);
- /*
- for(int j = 0; j < company_link_csv_reader.main().size();){
- String comps = company_link_csv_reader.main().get(j);
- drive.get(comps);
- System.out.println(comps);
- Thread.sleep((long) randomNumber);
- System.out.println(randomNumber);
- j++;*/
- //Thread.sleep((long) randomNumber);
- //drive.wait();
- }
- //drive.manage().timeouts().implicitlyWait(15, TimeUnit.SECONDS);
- //String pageSource = drive.getPageSource();
- //System.out.println(pageSource);
- //File input = new File("/tmp/input.html");
- //Document doc = Jsoup.parse(input, "UTF-8", "C:\\Users\\jonat\\Documents\\NetBeansProjects\\jrex\\companypage.html");
- //Document doc = Jsoup.parse(pageSource);
- //Grab company about page content
- //Grab Connected contact link
- //Grab company website link
- // Move to jobs
- //Grab Job content
- //Cycle through jobs
- //Cycle through job pages until no more
- //iterate through company list
- //Write content to csv
- }
Advertisement
Add Comment
Please, Sign In to add comment