sort out the required data

UNIX for Advanced & Expert Users
1

Hi All,

I have a file 1.txt which has the duplicate dns entries as shown:

Name: 000f9fbc6738.net.in|Addresses: 10.241.66.169, 10.84.2.222,212.241.66.170
Name: 001371e8ed3e.net.in|Addresses: 10.241.65.153, 10.84.1.101
Name: 00e06f5bd42a.net.in|Addresses: 10.72.19.218, 172.18.72.250

**
**

Now I have another file 2.txt which shows the current active dns IP entry as shown:

000f9fbc6738|10.84.2.222|27|
001371e8ed3e|10.84.1.101|27|
00e06f5bd42a|10.72.19.218|27|

**
**

Note: here |27| means it is an active IP with respect to that mac.

Now i would like to sort this data i.e., I want the macaddress and IP which is inactive by comparing 1.txt and 2.txt file and the final output should look like the one as shown below:

Name: 000f9fbc6738.net.in|Addresses: 10.241.66.169,212.241.66.170
Name: 001371e8ed3e.net.in|Addresses: 10.241.65.153
Name: 00e06f5bd42a.net.in|Addresses: 172.18.72.250

**
**

Your suggestion/help is greatly appreciated.

Thanks
imas.

2

Try this:

sed 's!\(.*\)|\(.*\)|\(.*\)|!s/\2//g!'  file2 >temp_file
sed -f temp_file file1
3 
awk -F"|" 'FNR==NR && $(NF-1) == "27"{active[$1]=$2; next}
{
 old=$0
 gsub(/Name: |\.net.*/,"")
 gsub(active[$0],"",old)
 print old  
}' file2 file1
4

Hi Dennis

Thanks a million

This code works as per my requirement.
sed 's!\(.*\)|\(.*\)|\(.*\)|!s/\2//g!' file2 >temp_file
sed -f temp_file file1

Once again thank you very much for providing an update within no time.

Appreciate all users help.

We can close this thread.

Imas.

5

Hmmm... I guess that code gives unexpected results if your files are like the following:

Name:    000f9fbc6738.net.in|Addresses: 10.84.1.101, 10.84.2.222,212.241.66.170
Name:    001371e8ed3e.net.in|Addresses:  10.241.65.153, 10.84.1.101
Name:    00e06f5bd42a.net.in|Addresses:  10.72.19.218, 172.18.72.250

000f9fbc6738|10.84.2.222|27|
001371e8ed3e|10.84.1.101|27|
00e06f5bd42a|10.72.19.218|27|
6

Yes sir,

Good catch, however how to get rid of this issue.

Thanks
-imas

7

It starts to get too cumbersome to do with sed alone.
I would switch to awk.
Check if the code provided by ghostdog74 works as expected.