c＃删除<a>字符串</a>中的特定<a>标签（仅电子邮件）</a>

Question

So I get a string like this from an external method:

var myString = "<p>Lorem &sect; 5 ipsum</p>\r\n<p><p>E-Mail: <a href=\"email@domain.com\">email@domain.com</a></p>\r\n<p>Lorem ipsum dolor sit amet</p><p><a href=\"http://www.adress.com\">name</a></p>\r\n";

I want to replace all e-mail addresses (no other links) with plain text. So afterwards my link should look something like this:

var myClearedString = "<p>Lorem &sect; 5 ipsum</p>\r\n<p><p>E-Mail: email@domain.com</p>\r\n<p>Lorem ipsum dolor sit amet</p><p><a href=\"http://www.adress.com\">name</a></p>\r\n"

There could be 1 to n occurrences in the string. I already searched stackoverflow, but the only thing related was this question: Replace mailto-links

In my opinion it would be the best way to convert the string into XML and search for it. Unfortunately it seems that some chars in my string are causing troubles (i assume it might be \\n or \\r).

Answer 1

You should look into Html AgilityPack for this. I'm sure that there are many Regular expressions that could get you most of the way but parsing HTML using Regex is generally a bad idea. See https://stackoverflow.com/a/1732454/880642 for some reasons why.

Agility pack will safely parse the document for you and let you traverse it to find the links that meet your criteria.

var htmlDocument = new HtmlDocument();
htmlDocument.LoadHtml(htmlPage);
var links = htmlDocument.DocumentNode.SelectNodes("//a[@href]");
foreach (var node in links)
{
    HtmlAttribute attribute = node.Attributes["href"];
    if(IsEmail(attribute.Value))
         node.ParentNode.RemoveChild(node, true); //<-- keepGrandChildren
}
var newhtml = htmlDocument.DocumentNode.OuterHtml;

You can probably use a regex to verify that the attribute value is an email or any number of .Net functions to see whether a string is an email. I'm surprised that these aren't mailto: links but you have to work with the data that you have.

Answer 2

I'll probably be hung for suggesting this, but you could use regular expressions.

Start with including the necessary dependency:

using System.Text.RegularExpressions;

Then we need to figure out the regular expression that will identify the sub strings that match your criteria. There are several sites that offer regular expression testing. Just search for "regular expression tester".

This will get every anchor tag, and create 3 groups:

(<a[^>]+>)(.*?)(<\/a>)

Now we need to get all the matches and replace them with the plain text value.

We can use the Regex.Replace method to complete the task:

string newValue = Regex.Replace(test, @"(<a[^>]+>)(.*?)(<\/a>)", (m) => 
{
    return m.Groups[2].Value;
});

This snippet is running the lambda expression for every instance matched. Then returns the value from the second group (being the content of the tag).

Answer 3

这将是Regex和Replace的很好用

Regex.Replace(myString, @"(<a.*?>)", "").Replace("</a>","")

Answer 4

You can use:

Regex.Replace(source, "<a.*?>", string.Empty);

or if you want replace more times you can use compiled regex:

Regex removeRegex = new Regex("<a.*?>", RegexOptions.Compiled);

and use it as below :

removeRegex.Replace(source, string.Empty);