Search code examples
unixfilterfilteringsolaris

Filter non-repeating values from a file in Solaris

I have a file in Solaris. The following is a sample from the file:

CELLO;998;CMGW4;20070926;1030;00000000;0000;15;/cpm_data;172.20.19.5;22;ON;true;UTC+06:00;Etc/GMT-6 
CELLO;999;CMGW1;20070920;1730;00000000;0000;15;/cpm_data;172.20.11.4;21;OFF;true;UTC+06:00;Etc/GMT-6
CELLO;999;CMGW3;20070629;1845;00000000;0000;15;/cpm_data;172.20.19.4;22;ON;true;UTC+06:00;Etc/GMT-6 
CELLO;999;CMGW4;20070926;1130;00000000;0000;15;/cpm_data;172.20.19.5;22;ON;true;UTC+06:00;Etc/GMT-6 
CELLO;99;CMGW5;20070930;1630;00000000;0000;15;/cpm_data;172.20.11.6;22;ON;true;UTC+06:00;Etc/GMT-6  
CELLO;9;CMGW4;20120330;1215;00000000;0000;15;/cpm_data;172.20.19.5;22;ON;true;UTC+06:00;Etc/GMT-6

I am interested in the 3rd column (separated by ;) i.e the filed starting with CMGW. I want to get only the 3rd column unique values. Considering the sample above, the output that I want is:

CMGW1
CMGW3
CMGW4
CMGW5

Any help is appreciated.

Solution

  • You can do it like this:

    awk -F\;  '{print $3}' data | sort -u

    Where data is the name of your data file.