1

I need to replace some text in C# using RegEx:

string strSText = "<P>Bulleted list</P><UL><P><LI>Bullet 1</LI><P></P><P>
<LI>Bullet 2</LI><P></P><P><LI>Bullet 3</LI><P></UL>"

Basically I need to get rid of the

"<P>"

tag(s) introduced between

"<UL><P><LI>", 
"</LI><P></P><P><LI>" and
"</LI><P></UL>"

I also need to ignore any spaces between these tags when performing the removal.

So

"</LI><P></P><P><LI>", "</LI>    <P></P><P><LI>", "</LI><P></P><P>   <LI>" or 
"</LI> <P> </P> <P> <LI>"

must all be replaced with

"</LI><LI>"

I tried using the following RegEx match for this purpose:

strSText = Regex.Replace(strSText, "<UL>.*<LI>", "<UL><LI>", RegexOptions.IgnoreCase);
strSText = Regex.Replace(strSText, "</LI>.*<LI>", "</LI><LI>", 
RegexOptions.IgnoreCase);
strSText = Regex.Replace(strSText, "</LI>.*</UL>", "</LI></UL>", 
RegexOptions.IgnoreCase);

But it performs a "greedy" match and results in:

"<P>Bulleted list</P><UL><LI>Bullet 3</LI></UL>"

I then tried using "lazy" match:

strSText = Regex.Replace(strSText, "<UL>.*?<LI>", "<UL><LI>", RegexOptions.IgnoreCase);
strSText = Regex.Replace(strSText, "</LI>.*?<LI>", "</LI><LI>", 
RegexOptions.IgnoreCase);
strSText = Regex.Replace(strSText, "</LI>.*?</UL>", "</LI></UL>", 
RegexOptions.IgnoreCase);

and this results in:

"<P>Bulleted list</P><UL><LI>Bullet 1</LI></UL>"

But I want the following result, which preserves all other data:

"<P>Bulleted list</P><UL><LI>Bullet 1</LI><LI>Bullet 2</LI><LI>Bullet 3</LI></UL>"
kaya3
  • 47,440
  • 4
  • 68
  • 97
user2751130
  • 67
  • 9
  • 7
    Don't use regular expressions for parsing HTML. [What is the best way to parse html in C#?](http://stackoverflow.com/questions/56107/what-is-the-best-way-to-parse-html-in-c) – Jonathon Reinhart Sep 11 '13 at 07:59
  • Would something like `strSText.Replace("

    • ", "
      • ");` etc... work?
     – DGibbs Sep 11 '13 at 08:17

2 Answers2

1

The following regexp matches one or more <P> or </P> tags:

(?:</?P>\s*)+

So if you place that between the other tags you have, you can get rid of them, i.e.

strSText = Regex.Replace(strSText, @"<UL>\s*(?:</?P>\s*)+<LI>", "<UL><LI>", RegexOptions.IgnoreCase);
strSText = Regex.Replace(strSText, @"</LI>\s*(?:</?P>\s*)+<LI>", "</LI><LI>", RegexOptions.IgnoreCase);
strSText = Regex.Replace(strSText, @"</LI>\s*(?:</?P>\s*)+</UL>", "</LI></UL>", RegexOptions.IgnoreCase);
Ed Chapel
  • 6,842
  • 3
  • 30
  • 44
krisku
  • 3,916
  • 1
  • 18
  • 10
1

Not really an answer to your question, but more of a comment to Jonathon: Parse HTML with HTMLAgilityPack

Christian HC
  • 56
  • 3