[sllug-members]: Script Question

[Jason Clint]

I agree with roger the problem with diff is that the files may or may
not be in the same order.  I worked trying to use diff for a bit and it
requires a lot of trimming once you get the results.

Roger with this script you gave me it seems you define the files
"$file1=somefile", "$file2=someotherfile".  Then you go in and cat,
grep, sort and grep again all in one command.  First so I can understand
how this is done can you explain what this command does "[[:space:]]+2\
[[:space:]]" does that tell it to ignore the space?

Thanks.

Junior Systems Administrator
Sento IT Jason Clint

(801) 431-9297
(801) 661-4139


-----Original Message-----
From: Roger Smith [mailto:roger at itigger.com] 
Sent: Monday, May 05, 2008 10:51 AM
To: sllug-members at sllug.org
Cc: Jason Clint
Subject: Re: [sllug-members]: Script Question

On Monday 05 May 2008 10:44:37 am Jason Clint wrote:
> I am trying to do a Linux script to compare 2 files in some way and
only
> return a certain result.
>
> File 1 = /etc/passwd 
> File 2 = nonusers.txt
>
> File 2 has all the nonusers in it and I need to make sure my script
> doesn't accidently pull out any of them.  This is a for a manual ldap
> import and when I ran the script the first time I had a user named
"ns"
> so it pulled out "ns" and "nscd" which is a nonuser.  Anyway I was
> trying to use a script to compare the 2 and only return a result that
> didn't have "nologin" or even just return all results except the ones
> that were in the status list nonusers.txt.  I'm kind of new at shell
> scripting so any help would be greatly appreciated.

diff works, assuming the orders are the same. This is another way of
going 
about it.

# define files to compare
File1 = /etc/passwd 
File2 = nonusers.txt

# list all login entries : show unique entries and count how often they
# appear : show entries that appear once
cat $File1 $File2|grep -v nologin| sort |uniq -c|grep -E "[[:space:]]+1\
[[:space:]]"

# list all login entries : show unique entries and count how often they
# appear : only show entries that appear twice
cat $File1 $File2|grep -v nologin| sort |uniq -c|grep -E "[[:space:]]+2\
[[:space:]]"

-- Roger