java/project/DoubanStrategy.java

package com.yyt.moviecrawler.strategy;

import com.yyt.moviecrawler.model.Movie;
import com.yyt.moviecrawler.model.DoubanMovie;
import org.openqa.selenium.By;
import org.openqa.selenium.WebDriver;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.chrome.ChromeDriver;

import java.util.ArrayList;
import java.util.List;

public class DoubanStrategy implements CrawlerStrategy {
    @Override
    public List<Movie> crawl(int limit) {
        WebDriver driver = new ChromeDriver();
        driver.get("https://movie.douban.com/top250");
        List<Movie> movies = new ArrayList<>();
        try {
            Thread.sleep(3000);
            List<WebElement> elements = driver.findElements(By.cssSelector(".item"));
            for (int i = 0; i < Math.min(limit, elements.size()); i++) {
                WebElement el = elements.get(i);
                String title = el.findElement(By.cssSelector(".title")).getText();
                double score = Double.parseDouble(el.findElement(By.cssSelector(".rating_num")).getText());
                String type = el.findElement(By.cssSelector(".bd p")).getText().split("/")[1].trim();
                String author = el.findElement(By.cssSelector(".bd p")).getText().split("/")[0].trim();

                movies.add(new DoubanMovie(title, score, type, author));
            }
        } catch (Exception e) {
            e.printStackTrace();
        } finally {
            driver.quit();
        }
        return movies;
    }
}