How do I read a txt file and keep only IP addresses based on the first 2 or 3 octets of the IP?

Question

Hello,
I have a text file and each line contains random text followed by an IP address as follows.
some text....172.30.25.30
some text.....172.30.85.10
some text..172.30.25.35
some text.......172.30.85.11
some text....172.30.15.1
some text...172.30.15.2
some text.......172.10.1.1
some text...172.20.4.2
some text..172.10.1.2
some text.....172.20.5.1
I'd like to create an output file which has only one entry for each unique entry in the file where either the first 2 or 3 octets are unique as follows:
Output File
172.30.25
172.30.85
172.30.15
172.10.1
172.20
Any suggestions are appreciated!
Thanks for your help! SdeDot

Answer 1

Thanks mjolinor.  Works great!
Two questions.
1. Could you plz suggest how this could be modified so this code would read the file in or accept it from the pipeline instead of wrapping the (@' around the data?
2. Could you plz briefly describe some of the details of the code so I can further research and understand.
Thanks for your help.
Thanks for your help! SdeDot
1. It already reads in the file.  The (@' .. '@) bits are just there to create a file using your test data to demonstrate that it works.
2.  Not user what kind of "details" you want.  There really isn't much there, and get-help on the cmdlets used should provide information on what's going on with them in that script.
[string](0..33|%{[char][int](46+("686552495351636652556262185355647068516270555358646562655775 0645570").substring(($_*2),2))})-replace " "