PowerShell 中的正则表达式从 Active Directory 中的 Managedby 属性获取城市名称

Question

Can anyone help me with this. I need to derive a City name from the "managedby" attribute in Active Directory which looks like this:

CN=Marley\, Bob,OU=Users,OU=PARIS,DC=Domain,DC=com

So I need to take everything out and be left with "PARIS"

I really don't know enough about Regex but assume its going to involve using -replace in some way. I have tried following some examples on the web but I just get lost. I can remove all special characters using:

'CN=Marley\, Bob,OU=Users,OU=PARIS,DC=Domain,DC=com' -replace '[\W]', ''

But I have no idea how to clean that up further.

Any help would be greatly appreciated

Answer 1

Actually you don't need regex for that. If the structure of the distinguished name is always the same you can use nested -split s... like this:

(('CN=Marley\, Bob,OU=Users,OU=PARIS,DC=Domain,DC=com' -split '=')[3] -split ',')[0]

or this:

(('CN=Marley\, Bob,OU=Users,OU=PARIS,DC=Domain,DC=com' -split ',')[-3] -split '=')[1]

I'd recommend the second version because this way you avoid confusion you can have with commas in the CN part of the distinguished name. ;-)

If you like to do it with regex anyway you can use look-arounds to extract what's between the users OU and the domain like this:

'CN=Marley\, Bob,OU=Users,OU=PARIS,DC=Domain,DC=com' -match '(?<=Users,OU=).+(?=,DC=DOmain)'
$Matches[0]

Answer 2

The following is a -replace -based solution that assumes that the city name follows the last ,OU= in the input string (though it wouldn't be hard to make the regex more specific).

It also supports city names with escaped , characters ( \, ), such as PARIS\, Texas .

$str = 'CN=Marley\, Bob,OU=Users,OU=PARIS\, Texas,DC=Domain,DC=com' 
# -> 'PARIS, Texas'
$str -replace '.+,OU=(.+?),DC=.+', '$1' -replace '\\,', ','

• .+,OU= greedily matches one or more ( + ) arbitrary characters ( . ) up to the last ,OU= substring in the input string.

• (.+?) matches on or more subsequent characters non-greedily ( +? ), via a capture group (capturing subexpression, (...) ).

• ,DC=.+ matches the next occurrence of substring ,DC followed by whatever is left in the string ( .+ ).

• Note that this means that the regex matches the entire string, so that the value of the substitution expression, $1 , is the only thing returned:

  • $1 refers to the value of the 1st capture group, which contains the city name.

• The second -replace operation unescapes the \, , ie turns it into , - note how the literal \ to replace had to be escaped as \\ in the regex.