python消找出img中的src标签_获取文本中所有的<img>标签的位置，获取所有img标签的src...

public static int[] GetImagePos(string str)

{

str = str.Replace("$", " ");

str = Regex.Replace(str, @"(?!)<.>", "");

str = Regex.Replace(str, @"]*>", "$");

int startPos = 0;

int foundPos = -1;

int count = 0;

List foundItems = new List();

do

{

foundPos = str.IndexOf("$", startPos);

if (foundPos > -1)

{

startPos = foundPos + 1;

count++;

foundItems.Add(foundPos);

}

} while (foundPos > -1 && startPos < str.Length);

return ((int[])foundItems.ToArray());

}

///

/// 获取Img的路径

///

/// Html字符串文本

/// 以数组形式返回图片路径

public static string[] GetHtmlImageUrlList(string htmlText)

{

Regex regImg = new Regex(@"]*?\bsrc[\s\t\r\n]*=[\s\t\r\n]*[""']?[\s\t\r\n]*(?[^\s\t\r\n""'<>]*)[^<>]*?/?[\s\t\r\n]*>", RegexOptions.IgnoreCase);

//新建一个matches的MatchCollection对象 保存 匹配对象个数(img标签)

MatchCollection matches = regImg.Matches(htmlText);

int i = 0;

string[] sUrlList = new string[matches.Count];

//遍历所有的img标签对象

foreach (Match match in matches)

{

//获取所有Img的路径src,并保存到数组中

sUrlList[i++] = match.Groups["imgUrl"].Value;

}

return sUrlList;

}