@(dsquery computer -limit 0 | %{[regex]::Replace($_, “`”CN=([^,]+),.*$”, ‘$1 ‘)})

This uses the .Net Regex object’s Replace method. I won’t repeat what is said above. But the regular expression is “`”CN=([^,]+),.*$”.

Breaking it into pieces we have
`” An escaped quote, so this matches the quote
CN= A literal string, so this matches the string CN=
(…) The stuff between the parens is a CAPTURED and will be stored in the $1
[^,] Match anything but the comma
+ Do the previous match (the [^,]) one or more times.
, A literal string, so this matches the comma
. Match any character
* Do the previous match (the any character) zero or more times
$ Match the end of the string.

So, what this regex does is throw away the “CN= at the beginning of the string and also everything after the comma. The stuff between the CN= and the comma is stored in the capture.

The Replace then takes the incoming string and replaces it with what was captured in the $1.

Just look at some of the comments at URL below where people have been comparing Oracle conferences with SQL conferences.

http://www.freelists.org/post/oracle-l/Slightly-OT-SQL-Server-conference-vs-Oracle-conferences