C#网络爬虫-阿里云开发者社区

C#网络爬虫

2016-07-23 908

版权

本文内容由阿里云实名注册用户自发贡献，版权归原作者所有，阿里云开发者社区不拥有其著作权，亦不承担相应法律责任。具体规则请查看《阿里云开发者社区用户服务协议》和《阿里云开发者社区知识产权保护指引》。如果您发现本社区中有涉嫌抄袭的内容，填写侵权投诉表单进行举报，一经查实，本社区将立刻删除涉嫌侵权内容。

简介： 公司编辑妹子需要爬取网页内容，叫我帮忙做了一简单的爬取工具这是爬取网页内容，像是这对大家来说都是不难得，但是在这里有一些小改动，代码献上，大家参考 1 private string GetHttpWebRequest(string url) 2 { 3 ...

公司编辑妹子需要爬取网页内容，叫我帮忙做了一简单的爬取工具

这是爬取网页内容，像是这对大家来说都是不难得，但是在这里有一些小改动，代码献上，大家参考

 1 private string GetHttpWebRequest(string url)  
 2        {  
 3            HttpWebResponse result;  
 4            string strHTML = string.Empty;  
 5            try  
 6            {  
 7                Uri uri = new Uri(url);  
 8                WebRequest webReq = WebRequest.Create(uri);  
 9                WebResponse webRes = webReq.GetResponse();  
10   
11                HttpWebRequest myReq = (HttpWebRequest)webReq;  
12                myReq.UserAgent = "User-Agent:Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; .NET CLR 1.0.3705";  
13                myReq.Accept = "*/*";  
14                myReq.KeepAlive = true;  
15                myReq.Headers.Add("Accept-Language", "zh-cn,en-us;q=0.5");  
16                result = (HttpWebResponse)myReq.GetResponse();  
17                Stream receviceStream = result.GetResponseStream();  
18                StreamReader readerOfStream = new StreamReader(receviceStream, System.Text.Encoding.GetEncoding("utf-8"));  
19                strHTML = readerOfStream.ReadToEnd();  
20                readerOfStream.Close();  
21                receviceStream.Close();  
22                result.Close();  
23            }  
24            catch  
25            {  
26                Uri uri = new Uri(url);  
27                WebRequest webReq = WebRequest.Create(uri);  
28                HttpWebRequest myReq = (HttpWebRequest)webReq;  
29                myReq.UserAgent = "User-Agent:Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; .NET CLR 1.0.3705";  
30                myReq.Accept = "*/*";  
31                myReq.KeepAlive = true;  
32                myReq.Headers.Add("Accept-Language", "zh-cn,en-us;q=0.5");  
33                //result = (HttpWebResponse)myReq.GetResponse();  
34                try  
35                {  
36                    result = (HttpWebResponse)myReq.GetResponse();  
37                }  
38                catch (WebException ex)  
39                {  
40                    result = (HttpWebResponse)ex.Response;  
41                }  
42                Stream receviceStream = result.GetResponseStream();  
43                StreamReader readerOfStream = new StreamReader(receviceStream, System.Text.Encoding.GetEncoding("gb2312"));  
44                strHTML = readerOfStream.ReadToEnd();  
45                readerOfStream.Close();  
46                receviceStream.Close();  
47                result.Close();  
48            }  
49            return strHTML;  
50        }

　　这是根据url爬取网页远吗，有一些小改动，很多网页有不同的编码格式，甚至有些网站做了反爬取的防范，这个方法经过能够改动也能爬去

以下是爬取网页所有的网址链接

/// <summary>  
        /// 提取HTML代码中的网址  
        /// </summary>  
        /// <param name="htmlCode"></param>  
        /// <returns></returns>  
        private static List<string> GetHyperLinks(string htmlCode, string url)  
        {  
            ArrayList al = new ArrayList();  
            bool IsGenxin = false;  
            StringBuilder weburlSB = new StringBuilder();//SQL  
            StringBuilder linkSb = new StringBuilder();//展示数据  
            List<string> Weburllistzx = new List<string>();//新增  
            List<string> Weburllist = new List<string>();//旧的  
            string ProductionContent = htmlCode;  
            Regex reg = new Regex(@"http(s)?://([\w-]+\.)+[\w-]+/?");  
            string wangzhanyuming = reg.Match(url, 0).Value;  
            MatchCollection mc = Regex.Matches(ProductionContent.Replace("href=\"/", "href=\"" + wangzhanyuming).Replace("href='/", "href='" + wangzhanyuming).Replace("href=/", "href=" + wangzhanyuming).Replace("href=\"./", "href=\"" + wangzhanyuming), @"<[aA][^>]* href=[^>]*>", RegexOptions.Singleline);  
            int Index = 1;  
            foreach (Match m in mc)  
            {  
                MatchCollection mc1 = Regex.Matches(m.Value, @"[a-zA-z]+://[^\s]*", RegexOptions.Singleline);  
                if (mc1.Count > 0)  
                {  
                    foreach (Match m1 in mc1)  
                    {  
                        string linkurlstr = string.Empty;  
                        linkurlstr = m1.Value.Replace("\"", "").Replace("'", "").Replace(">", "").Replace(";", "");  
                        weburlSB.Append("$-$");  
                        weburlSB.Append(linkurlstr);  
                        weburlSB.Append("$_$");  
                        if (!Weburllist.Contains(linkurlstr) && !Weburllistzx.Contains(linkurlstr))  
                        {  
                            IsGenxin = true;  
                            Weburllistzx.Add(linkurlstr);  
                            linkSb.AppendFormat("{0}<br/>", linkurlstr);  
                        }  
                    }  
                }  
                else  
                {  
                    if (m.Value.IndexOf("javascript") == -1)  
                    {  
                        string amstr = string.Empty;  
                        string wangzhanxiangduilujin = string.Empty;  
                        wangzhanxiangduilujin = url.Substring(0, url.LastIndexOf("/") + 1);  
                        amstr = m.Value.Replace("href=\"", "href=\"" + wangzhanxiangduilujin).Replace("href='", "href='" + wangzhanxiangduilujin);  
                        MatchCollection mc11 = Regex.Matches(amstr, @"[a-zA-z]+://[^\s]*", RegexOptions.Singleline);  
                        foreach (Match m1 in mc11)  
                        {  
                            string linkurlstr = string.Empty;  
                            linkurlstr = m1.Value.Replace("\"", "").Replace("'", "").Replace(">", "").Replace(";", "");  
                            weburlSB.Append("$-$");  
                            weburlSB.Append(linkurlstr);  
                            weburlSB.Append("$_$");  
                            if (!Weburllist.Contains(linkurlstr) && !Weburllistzx.Contains(linkurlstr))  
                            {  
                                IsGenxin = true;  
                                Weburllistzx.Add(linkurlstr);  
                                linkSb.AppendFormat("{0}<br/>", linkurlstr);  
                            }  
                        }  
                    }  
                }  
                Index++;  
            }  
            return Weburllistzx;  
        }

这块的技术其实就是简单的使用了正则去匹配！接下来献上获取标题，以及存储到xml文件的方法

 1 /// <summary>  
 2         /// // 把网址写入xml文件  
 3         /// </summary>  
 4         /// <param name="strURL"></param>  
 5         /// <param name="alHyperLinks"></param>  
 6         private static void WriteToXml(string strURL, List<string> alHyperLinks)  
 7         {  
 8             XmlTextWriter writer = new XmlTextWriter(@"D:\HyperLinks.xml", Encoding.UTF8);  
 9             writer.Formatting = Formatting.Indented;  
10             writer.WriteStartDocument(false);  
11             writer.WriteDocType("HyperLinks", null, "urls.dtd", null);  
12             writer.WriteComment("提取自" + strURL + "的超链接");  
13             writer.WriteStartElement("HyperLinks");  
14             writer.WriteStartElement("HyperLinks", null);  
15             writer.WriteAttributeString("DateTime", DateTime.Now.ToString());  
16             foreach (string str in alHyperLinks)  
17             {  
18                 string title = GetDomain(str);  
19                 string body = str;  
20                 writer.WriteElementString(title, null, body);  
21             }  
22             writer.WriteEndElement();  
23             writer.WriteEndElement();  
24             writer.Flush();  
25             writer.Close();  
26         }  
27         /// <summary>  
28         /// 获取网址的域名后缀  
29         /// </summary>  
30         /// <param name="strURL"></param>  
31         /// <returns></returns>  
32         private static string GetDomain(string strURL)  
33         {  
34             string retVal;  
35             string strRegex = @"(\.com/|\.net/|\.cn/|\.org/|\.gov/)";  
36             Regex r = new Regex(strRegex, RegexOptions.IgnoreCase);  
37             Match m = r.Match(strURL);  
38             retVal = m.ToString();  
39             strRegex = @"\.|/$";  
40             retVal = Regex.Replace(retVal, strRegex, "").ToString();  
41             if (retVal == "")  
42                 retVal = "other";  
43             return retVal;  
44         }  
45 /// <summary>  
46         /// 获取标题  
47         /// </summary>  
48         /// <param name="html"></param>  
49         /// <returns></returns>  
50         private static string GetTitle(string html)  
51         {  
52             string titleFilter = @"<title>[\s\S]*?</title>";  
53             string h1Filter = @"<h1.*?>.*?</h1>";  
54             string clearFilter = @"<.*?>";  
55   
56             string title = "";  
57             Match match = Regex.Match(html, titleFilter, RegexOptions.IgnoreCase);  
58             if (match.Success)  
59             {  
60                 title = Regex.Replace(match.Groups[0].Value, clearFilter, "");  
61             }  
62   
63             // 正文的标题一般在h1中，比title中的标题更干净  
64             match = Regex.Match(html, h1Filter, RegexOptions.IgnoreCase);  
65             if (match.Success)  
66             {  
67                 string h1 = Regex.Replace(match.Groups[0].Value, clearFilter, "");  
68                 if (!String.IsNullOrEmpty(h1) && title.StartsWith(h1))  
69                 {  
70                     title = h1;  
71                 }  
72             }  
73             return title;  
74         }