网页抓取小工具

最近在线看电子书,由于篇幅太长,而且找不到下载地址,于是写了个小工具,将电子书下载到本地。

整体思路:

1、抓取出目录中各章节的名称及URL

2、遍历章节URL,获取具体内容

3、将章节URL进行分包,交给多线程处理

4、将处理完的内容重新整理,按章节名称排序

5、将内容写入TXT文件

首先抓取导航页面的内容,通过WebRequest对象获取网页内容

/// <summary>
/// 通过链接地址获取HTML内容
/// </summary>
/// <param name="url"></param>
/// <returns></returns>
private static string GetHtml(string url)
{
    string html = "";
    try
    {
        WebRequest request = WebRequest.Create(url);
        request.Credentials = CredentialCache.DefaultCredentials;
        HttpWebResponse response = (HttpWebResponse)request.GetResponse();
        Stream stream = response.GetResponseStream();
        StreamReader reader = new StreamReader(stream, encoding);

        html = reader.ReadToEnd();
        reader.Close();
        stream.Close();
        response.Close();
    }
    catch
    {
                
    }

    return html;
}

通过正则获取章节地址及名称

/// <summary>
/// 获取所有链接地址
/// </summary>
/// <param name="html"></param>
private static Dictionary<string, string> GetAllUrl(string html)
{
    string titlePattern = @"第(?<index>\d+)节";
    Dictionary<string, string> dictRet = new Dictionary<string, string>();
    string pattern = @"<a[^>]*?href=(['""]?)(?<url>[^'""\s>]+)\1[^>]*>(?<text>(?:(?!</?a\b).)*)</a>";
    Regex reg = new Regex(pattern, RegexOptions.IgnoreCase);
    MatchCollection mc = reg.Matches(html);
    foreach (Match m in mc)
    {
        //将TITLE解析成第XXX节
        string title = NoHTML(m.Groups["text"].Value).Replace("\n\r", "").Replace(" ", "") ;
        Match mTitle = Regex.Match(title, titlePattern);
        if (!mTitle.Success)
        {
            continue;
        }

        title = string.Format("第{0}节", mTitle.Groups["index"].Value.PadLeft(3, '0'));

        string url = m.Groups["url"].Value;
        url = url.StartsWith("http://") ? url : string.Format("{0}/{1}", webUrl.TrimEnd('/'), url);
        if (!dictRet.ContainsKey(url))
        {
            dictRet.Add(url, title);
        }
    }

    return dictRet;
}

按线程数对地址进行分包处理

/// <summary>
/// 按线程数对URL进行分包处理
/// </summary>
/// <param name="dictUrls"></param>
/// <returns></returns>
private static Dictionary<int, Dictionary<string, string>> SplitUrl(Dictionary<string, string> dictUrls)
{
    Dictionary<int, Dictionary<string, string>> dictRet = new Dictionary<int, Dictionary<string, string>>();
    int count = dictUrls.Count;
    int splitCount = count / threadCount;   //每包的大小
    int keyIndex = 0;
    int calCount = 0;
    foreach (string keyUrl in dictUrls.Keys)
    {
        if (calCount == splitCount && keyIndex < threadCount - 1)
        {
            keyIndex++;
            calCount = 0;
        }

        if (!dictRet.ContainsKey(keyIndex))
        {
            dictRet.Add(keyIndex, new Dictionary<string, string>());
        }

        dictRet[keyIndex].Add(keyUrl, dictUrls[keyUrl]);
        calCount++;
    }

    return dictRet;
}

开启线程,获取内容,存入字典

static void Main(string[] args)
{
    Dictionary<int, Dictionary<string, string>> dictSplitUrls = SplitUrl(dictUrls);
    for (int i = 0; i < threadCount; i++)
    {
        Dictionary<string, string> dictRun = dictSplitUrls[i];
        int index = i;
        ThreadStart ts = delegate { Run(dictRun, index); };
        new Thread(ts).Start();
    }
}

/// <summary>
/// 运行内容解析
/// </summary>
/// <param name="dictUrls"></param>
private static void Run(Dictionary<string, string> dictUrls, int threadId)
{
    foreach (string keyUrl in dictUrls.Keys)
    {
        if (!dictContent.ContainsKey(dictUrls[keyUrl]))
        {
            dictContent.Add(dictUrls[keyUrl], "");
        }

        Console.WriteLine("Thread {0},deal with {1}:{2}...", threadId, dictUrls[keyUrl], keyUrl);
        string content = GetHtml(keyUrl);
        if (content.Length > 0)
        {
            content = AnalyseContent(content);
        }

        if (content.Length > 0)
        {
            dictContent[dictUrls[keyUrl]] = content;
            Write(content, string.Format("{0}_{1}.txt", systemName, dictUrls[keyUrl]));
            Console.WriteLine("success");
        }
        else
        {
            Console.WriteLine("failed");
        }

        lock (lockObj)
        {
            dealedCount++;
        }
    }

    if (dealedCount == totalCount)
    {
        WriteToTxt();
    }
}

最后将所有内容重新整理,输出到文本

/// <summary>
/// 输出到文本文件
/// </summary>
private static void WriteToTxt()
{
    Console.WriteLine("获取内容完成,生成TXT");

    //字典按章节排序
    StringBuilder sbTxt = new StringBuilder();
    List<KeyValuePair<string, string>> lstOrder = dictContent.OrderBy(t => t.Key).ToList();
    StringBuilder sbContent = new StringBuilder();
    foreach (KeyValuePair<string, string> item in lstOrder)
    {
        Console.WriteLine("生成{0}", item.Key);
        sbContent.AppendFormat("{0} {1}", item.Key, item.Value);
    }

    Write(sbContent.ToString(), string.Format("{0}.txt", systemName));
    Console.WriteLine("生成TXT成功");
}

 

Demo

posted @ 2012-06-06 23:43  flysoul  阅读(1395)  评论(3编辑  收藏