This is a pretty nice regexp for breaking down BBCode that I wrote a few yrs ago and just found it again in my old PC

REGEX

# This regular expression is written by Kubo2
# It is designed to syntactily match BB Tags in
# raw BB Code.
#
# It is written specifically for dh Project BB parser core - DH\BB\Parser.
#
# Author: Kubo2 <[email protected]>
# Date: 11/17/2014
#---------------------------------------------------------------------------------------------
#
#
# match paired tags 
# like [tag name some attrs]text content .*[/tag name]
\[ # opening brace
   # tag name and attrs
   #
   (?<tag>
      [a-zA-Z0-9]+ (?# tag name consists just of alphanumeric characters <del>or it is a breaker</del>)
   )
   (?<attrs>
      [^\]]*  # and does not contain any ] brace character
   )
] # closing brace
(?<contents>
   (?# tag contents - nested text)
   (?# anything that DOES NOT contain end-tag)
   .+?
   (?<!\[/\k<tag>])
)
# stuff below is the endtag pattern
\[/\k<tag>]
#
# ---- OR ----
   |
#
# match single-root tags
# with no end-tag
# [inline tag name attrs]
\[
   (?<tag>
      (?# tag name)
      !          |
      \#\d+  |
      \w+
   )
]
#
# MUST NOT be followed by the closing tag
# but if that is, implies the first

TEST.bb

Hi, this is some text, this is [b]bold text[/b]. Here is the fu[!]cking motherfuc[!]ker, and it behaves like a breaker.
New line in bb code [#2] suck
Fuck [b]this
one[/b]!
here then the horizontal line appears [hr]
Here is [url=http://example.com]link to example.com[/url] an some additional stuff
and fuck it all [url=hi]why does this suck?[/url]