Extract one of multiple files with similar syntax

[kHz]

I run an rpm -qa and get output of all packages installed on a Linux server. Then on another server I have a list of all RPM packages available. I need to compare the rpm list to the 'all' list and get just that one match even though there are similar names.

Example (rpm -qa list):
glibc-2.6.3-1.44.3

Example (all RPM packages available):
glibc | 2.6.3 | 1.44.3
glibc-html | 2.3.1 | 1.44.2

What I need to match is just the glibc on the first line of the example but I get all the glibc packages because they contain glibc when doing a grep. I tried using something with the shell by just getting the length of the filename (glibc which is 5) in the rpm -qa list and comparing that with each filename in the "all" list and if the file is a length of 5 and then matches the name then I use it. It still isn't giving me the correct output though.

Is there an easier way to only match glibc and not all other glibc* packages?

Thanks!!

 

[kHz]

There are many packages that match this scenario. Probably 6,000+ lines in the 'all' list, but only 250 or so in the rpm -qa generated list.

Thanks!!

 

[PHV]

Typed, untested:
awk -F'[ |]+' '{NR==FNR{a[$1];next}$1"-"$2"-"$3 in a' rpmlist alllist > matchlist

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886