前言

今天我们来讲讲如何使用.NET开源（MIT License）的轻量、灵活、高性能、跨平台的分布式网络爬虫框架DotnetSpider来快速实现网页数据抓取功能。

“

注意：为了自身安全请在国家法律允许范围内开发网页爬虫功能。

网页数据抓取需求

本文我们以抓取博客园10天推荐排行榜第一页的文章标题、文章简介和文章地址为示例，并把抓取下来的数据保存到对应的txt文本中。

• 请求地址：https://www.cnblogs.com/aggsite/topdiggs

创建控制台应用

创建名为DotnetSpiderExercise的控制台应用。

安装DotnetSpider NuGet包

NuGet包管理器搜索：DotnetSpider

添加Serilog日志组件

NuGet包管理器搜索：Serilog.AspNetCore

添加RecommendedRankingModel

namespace DotnetSpiderExercise
{
    public class RecommendedRankingModel
    {
        /// <summary>
        /// 文章标题
        /// </summary>
        public string ArticleTitle { get; set; }

        /// <summary>
        /// 文章简介
        /// </summary>
        public string ArticleSummary { get; set; }

        /// <summary>
        /// 文章地址
        /// </summary>
        public string ArticleUrl { get; set; }
    }
}

添加RecommendedRankingSpider

网页数据抓取的业务逻辑都在这里面。

using DotnetSpider.DataFlow.Parser;
using DotnetSpider.DataFlow;
using DotnetSpider.Downloader;
using DotnetSpider.Http;
using DotnetSpider.Scheduler.Component;
using DotnetSpider.Selector;
using DotnetSpider;
using Microsoft.Extensions.Logging;
using Microsoft.Extensions.Options;
using Serilog;
using DotnetSpider.Scheduler;
using Microsoft.Extensions.Hosting;
using System.Reflection;

namespace DotnetSpiderExercise
{
    public class RecommendedRankingSpider : Spider
    {
        public RecommendedRankingSpider(IOptions<SpiderOptions> options,
            DependenceServices services,
            ILogger<Spider> logger) : base(options, services, logger)
        {
        }

        public static async Task RunAsync()
        {
            var builder = Builder.CreateDefaultBuilder<RecommendedRankingSpider>();
            builder.UseSerilog();
            builder.UseDownloader<HttpClientDownloader>();
            builder.UseQueueDistinctBfsScheduler<HashSetDuplicateRemover>();
            await builder.Build().RunAsync();
        }

        protected override async Task InitializeAsync(CancellationToken stoppingToken = default)
        {
            //添加自定义解析
            AddDataFlow(new Parser());
            //使用控制台存储器
            AddDataFlow(new ConsoleStorage());
            //添加采集请求:博客园10天推荐排行榜
            await AddRequestsAsync(new Request("https://www.cnblogs.com/aggsite/topdiggs")
            {
                //请求超时10秒
                Timeout = 10000
            });
        }

        class Parser : DataParser
        {
            public override Task InitializeAsync()
            {
                return Task.CompletedTask;
            }

            protected override Task ParseAsync(DataFlowContext context)
            {
                var recommendedRankingList = new List<RecommendedRankingModel>();
                // 网页数据解析
                var number = 1;
                var recommendedList = context.Selectable.SelectList(Selectors.XPath(".//article[@class='post-item']"));
                foreach (var news in recommendedList)
                {
                    var articleTitle = news.Select(Selectors.XPath(".//a[@class='post-item-title']"))?.Value;
                    var articleSummary = news.Select(Selectors.XPath(".//p[@class='post-item-summary']"))?.Value?.Replace("\n", "").Replace(" ", "");
                    var articleUrl = news.Select(Selectors.XPath(".//a[@class='post-item-title']/@href"))?.Value;

                    Console.WriteLine($"第{number}篇文章 标题：{articleTitle}");

                    recommendedRankingList.Add(new RecommendedRankingModel
                    {
                        ArticleTitle = articleTitle,
                        ArticleSummary = articleSummary,
                        ArticleUrl = articleUrl
                    });

                    number++;
                }

                using (StreamWriter sw = new StreamWriter("RecommendedRanking.txt"))
                {
                    foreach (RecommendedRankingModel model in recommendedRankingList)
                    {
                        string line = $"文章标题：{model.ArticleTitle}\r\n文章简介：{model.ArticleSummary}\r\n文章地址：{model.ArticleUrl}";
                        sw.WriteLine(line + "\r\n ========================================================================================== \r\n");
                    }
                }
                return Task.CompletedTask;
            }
        }
    }
}

Program执行数据抓取

namespace DotnetSpiderExercise
{
    public class Program
    {
        static async Task Main(string[] args)
        {
            Console.WriteLine("网页数据抓取开始...");

            await RecommendedRankingSpider.RunAsync();

            Console.WriteLine("网页数据抓取完成...");
        }
    }
}

抓取数据和页面数据对比

抓取数据

页面数据

项目源码地址

• GitHub源码地址：https://github.com/dotnetcore/DotnetSpider
• GitHub wiki：https://github.com/dotnetcore/DotnetSpider/wiki
• 本文示例源码：https://github.com/YSGStudyHards/DotNetExercises/tree/master/DotnetSpiderExercise