Parse multiline BBCode with C# Regex

Question

I'm working on a C# class to parse BBCodes out of text pulled from a database for PHPBB posts. On the PHPBB there is a custom BBCode added which looks like this:

[deck={TEXT1}]{TEXT2}[/deck]

Which, sitting in the database, looks like this:

[deck=FirstText:13giljne]Large Multiline Text[/deck:13giljne]

I'm attempting to replace that using a Regex in C#. My C# looks like this:

string text = "[deck=FirstText:13giljne]Large Multiline Text[/deck:13giljne]";
string replace = "my replacement string";
string pattern = @"\[deck=((.|\n)*?)(?:\s*)\]((.|\n)*?)\[/deck(?:\s*)\]";
RegexOptions options = RegexOptions.Compiled | RegexOptions.IgnoreCase | RegexOptions.Singleline;
Regex regex = new Regex(pattern, options);
string result = regex.Replace(text, replace);

I'm pretty sure it all just comes down to my Regex pattern being wrong. Which comes as no surprise to me, since Regex isn't exactly my strong suit.

Thanks in advance. Any help is greatly appreciated.

EDIT: Since some people found it unclear, I'll add larger examples.

Source text:

[deck=Foo:13giljne]
    Item #1
    Item #2
    Item #3
    Item #4
[/deck:13giljne]

Desired result:

<span>Foo</span>
<div>
    Item #1
    Item #2
    Item #3
    Item #4
</div>

Hopefully this gives a clearer picture of what I'm trying to do.

Answer 1

I think your regex shows that you need to match "First Text" and "Large Multiline Text".

\[deck=([^\:]+?):(?:[^\]]+)\]([^\[]+?)\[/deck:(?:[^\]]+)\]

This should help and it's very close to yours.

Answer 2

If you're new to regular expressions, you might try matching a little at a time so that you're sure your string will match. For example, given the string:

string text = "[deck=FirstText:13giljne]Large Multiline Text[/deck:13giljne]";

Write an expression that matches the first part:

string firstPart = "\[deck=[^\]]+\]";

The [^\]]+ says, "match everything that isn't a ] character".

Verify that it matches:

Match m = Regex.Match(text, firstPart);

Then tack on the second part:

string firstAndSecond = firstPart + "[^\[]*";

And test that.

Once that's working, you can add the last part:

string search = firstAndSecond + "\[/deck[^\]]\]";

The final regular expression would be (\[deck=[^\]]+\])([^\[]+)(\[/deck[^\]]+\].

I grouped the individual parts to make it easier to see them. You can remove the groups if you want or make them non-capturing.

EDIT:

I see from your edit that you want to capture the FirstText, and the three groups:

string search = "(\[deck=([^:]+):[^\]]+\])([^\[]+)(\[/deck[^\]]+\]";

The replacement string, then, would be something like:

string replace = "<span>$2</span>\n<div>$3</div>";

Answer 3

Check STML Parser on GitHub or Nuget. It doesn't use RegEx but it is much more efficient and faster.

Answer 4

If you want to replace "[deck=FirstText:13giljne]Large Multiline Text[/deck:13giljne]", don't use regex. Use Replace.

string result= text.Replace("[deck=FirstText:13giljne]Large Multiline Text[/deck:13giljne]", replace);

Regexes are usually used where the string is not fully known, but its structure is known.