C#/.NET获取网页中全部图片Img标签

最新推荐文章于 2024-03-04 09:52:00 发布

老陈聊架构

最新推荐文章于 2024-03-04 09:52:00 发布

阅读量2k

点赞数

分类专栏： # 『Winform_C/S架构开发』文章标签： C# 图片标签

本文链接：https://blog.csdn.net/qq_34202873/article/details/88753158

版权

『Winform_C/S架构开发』专栏收录该内容

34 篇文章 7 订阅

订阅专栏

C#/.NET获取网页中全部图片链接

方法函数1：获取HTML中的Img标签

/// <summary> 
/// 取得HTML中所有图片的 URL。 
/// </summary> 
/// <param name="sHtmlText">HTML代码</param> 
/// <returns>图片的URL列表</returns> 
private string[] GetHtmlImageUrlList(string sHtmlText)
{
	// 定义正则表达式用来匹配 img 标签 
	Regex regImg = new Regex(@"<img\b[^<>]*?\bsrc[\s\t\r\n]*=[\s\t\r\n]*[""']?[\s\t\r\n]*(?<imgUrl>[^\s\t\r\n""'<>]*)[^<>]*?/?[\s\t\r\n]*>", RegexOptions.IgnoreCase);

	// 搜索匹配的字符串 
	MatchCollection matches = regImg.Matches(sHtmlText);
	int i = 0;
	string[] sUrlList = new string[matches.Count];

	// 取得匹配项列表 
	foreach (Match match in matches)
		sUrlList[i++] = match.Groups["imgUrl"].Value;
	return sUrlList;
}

方法函数2：下载网页

public static string DownloadHtml(string url)
{
    string html = string.Empty;
    try
    {
        var request = WebRequest.Create(url) as HttpWebRequest;
        request.Timeout = RequestModels.Timeout*1000;//设置30s的超时
        request.UserAgent = RequestModels.UserAgent;
        //request.ContentType = "application/x-www-form-urlencoded; charset=UTF-8";
        request.ContentType = RequestModels.ContentType;
        request.Headers.Add("Cookie", @RequestModels.Cookie);
        //Encoding enc = Encoding.GetEncoding("GB2312"); // 如果是乱码就改成 utf-8 / GB2312

        using (var response = request.GetResponse() as HttpWebResponse)
        {
            if (response.StatusCode != HttpStatusCode.OK)
            {
                //Log.Error(string.Format("抓取{0}地址返回失败,response.StatusCode为{1}", url, response.StatusCode));
                return url;
            }
            try
            {

                StreamReader sr = new StreamReader(response.GetResponseStream(), Encoding.UTF8);//Encoding.GetEncoding("GB2312"));// 
                html = sr.ReadToEnd();
                sr.Close();
            }
            catch (Exception ex)
            {
                Log.Error(string.Format("DownloadHtml抓取{0}保存失败", url), ex);
                html = null;
            }
        }
    }
    catch (System.Net.WebException ex)
    {
        if (ex.Message.Equals("远程服务器返回错误: (306)。"))
        {
            //logger.Error("远程服务器返回错误: (306)。", ex);
            //Console.WriteLine("DownloadHtml url={0}结果为空,306", url);
            return null;
        }
    }
    catch (Exception ex)
    {
        //logger.Error("异常", ex);
        //Log.Error(string.Format("DownloadHtml抓取{0}出现异常", url), ex);
        html = null;
    }
    return html;
}