RegExp for but not for

Question

Please help to rewrite all links without "target" attribute.

For example, the text is:

<a href="google.com" onclick="alert('Hello!!')">My Link 1</a>
<a href="my.com" class="some-class">My Link 2</a>
<a href="dot.net" target="_parent" class="some-class">My Link 3</a>
<a href="find.me" class="some-class">My Link 4</a>

The text is needed to got:

<a href="google.com" onclick="alert('Hello!!')" target="_blank">My Link 1</a>
<a href="my.com" class="some-class" target="_blank">My Link 2</a>
<a href="dot.net" target="_parent" class="some-class">My Link 3</a>
<a href="find.me" class="some-class" target="_blank">My Link 4</a>

3rd link is untouched, other links have attribute "target" now.

Please help to compose Regular Expression correct. I tried this:

Regex.Replace(text, "<(a)([^>]+)(((?! target=).)*$)([^>]+)>", "<$1 target=\"_parent\" $2 $3>");

but it's not working.

"Html Agility Pack" is undesirable.

Don't [parse html with Regex](http://stackoverflow.com/a/1732454/1895201) — DGibbs, Mar 04 '13 at 12:29
It's impossible to understand what author wanted to tell there. I still want to parse string with Regex. — user1820034, Mar 04 '13 at 12:37
Hmm, do you want some other solution that can make it without Regex? — Pawan Nogariya, Mar 04 '13 at 12:38
@user1820034 Are you familiar with satire? What's wrong with HTML agility pack? It was built for these sort of tasks. Why are you pre-emptively rejecting the solution to your problem? — DGibbs, Mar 04 '13 at 12:41
It should be small and simple solution. It's not desirable to install any packages or big classes. — user1820034, Mar 04 '13 at 12:41
I don't like HTML agility pack either, but I DON't recommend regex on parsing html. I have an app that parses HTML using SGMLReader and converting HTML to XML and there I would use XMLTextReader to parse it. — jomsk1e, Mar 04 '13 at 21:49

score 4 · Accepted Answer · answered Mar 04 '13 at 14:25

4

This should work as desired:

Regex.Replace(text, "<a(((?!target=).)*)\">", "<a$1\" target=\"_parent\">")

A small amount of assumption is required that every opening anchor tag you close must have the " character just before closing the opening tag with the > character.

i.e. <a......">My link</a>

answered Mar 04 '13 at 14:25

Chirag Bhatia - chirag64

4,430
3
26
35

You can drop the small assumption by replacing the match for any character `.` by the match for any character except a closing bracket `>`, like so: `[^>]`. The full regex becomes: `])*)>` – Jules Colle Dec 11 '21 at 16:00

score 1 · Answer 2 · edited Dec 27 '16 at 18:32

1

Solution for you:

Regex _r = new Regex("<a (.+?)>");
foreach (Match m in _r.Matches(text))
{
    string Link = m.Groups[0].Value;
    if (!Link.Contains("target"))
        text = text.Replace(Link, string.Format("{0} target=\"_parent\">", Link.Substring(0, Link.Length - 1)));
}

edited Dec 27 '16 at 18:32

Termininja

6,620
12
48
49

answered Mar 04 '13 at 12:52

01010100101001010

19
2

Thanks, but I've tried `Regex.Replace(text, "<(a)([^>]+)>", "<$1 target=\"_parent\" $2>")` already, but it also grabs all links with "target" attribute. I want to exclude them. – user1820034 Mar 04 '13 at 12:55
There are a lot of links with "target" and without "target" attributes in text. So `if(!text.Contains(...))` does not work. – user1820034 Mar 04 '13 at 13:01
Thank you for help. That's good, but I don't want to remove "target='something'". In your example, if "target='something'" exists, it's replaced with "target='_parent'", but it should be "target='something'". – user1820034 Mar 04 '13 at 13:11

score 0 · Answer 3 · answered Mar 04 '13 at 12:50

0

May be something easier like this ? :

if (false == text.Contains("target="))
{
   Regex.Replace(text, "<(a)([^>]+)(((?! target=).)*$)([^>]+)>", "<$1 target=\"_parent\" $2 $3>");
}

answered Mar 04 '13 at 12:50

Xaruth

4,034
3
19
26

It does not work because variable "text" contains a lot of html tags with or without "target" attribute. – user1820034 Mar 04 '13 at 13:14

RegExp for but not for

3 Answers3