unique values

[sandeepmur]

Hi,

I am trying to obtain unique values from a file like this:

$> uniq abc.xls > test.xls

but the above does not seem to be working.. I also tried using the -u option to no avail..

is there any other alternative ??

thnx

 

[bfitzmai]

The only criteria that I see with the command uniq is the repeated lines must be adjacent to each other. You also do not need the > between the file names... In your example, the command should look like:

uniq abc.xls test.xls

 

[sandeepmur]

I am doing something like this:

netstat | grep tcp | awk -f awkscript.awk > abc.xls
uniq abc.xls >> netstat.xls
cp netstat.xls temp.xls
rm netstat.xls
uniq temp.xls > netstat.xls
rm temp.xls
rm abc.xls


Even if I remove the ">" in the uniq command, it doesnt make any difference..

any further ideias ??
thnx

 

[vgersh99]

netstat | grep tcp | ( awk -f awkscript.awk; cat netstat.xls) | sort | uniq > netstat.xls

vlad
+----------------------------+
| #include<disclaimer.h> |
+----------------------------+

 

[vgersh99]

ooops,
nevermind the above - this is wrong
not tested:

( (netstat | grep tcp | awk -f awkscript.awk) ; cat netstat.xls ) | sort | uniq > netstat.xls

vlad
+----------------------------+
| #include<disclaimer.h> |
+----------------------------+

 

[PHV]

In the sort man page take a look at the -k and -u options.

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244

 

[sandeepmur]

My problem is that apart from the fact that I want the file netstat.xls to contain only unique values, I am putting the code above as a shell script which is executed every hour and appends to netstat.xls only unique values which are not already present in the file..

changing ">" to ">>" simply appends the file without caring for unique values..

thnx

 

[PHV]

netstat | grep tcp | awk -f awkscript.awk > abc.xls
sort -o netstat.xls -u abc.xls netstat.xls

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244

 

[sandeepmur]

Hi,

Thanks for the response.. Let me see if I got the above right..

netstat | grep tcp | awk -f awkscript.awk > abc.xls
wud create a file abc.xls with the output from the netstat command.

sort -o netstat.xls -u abc.xls netstat.xls

I am not sure why -o option is needed when -u abc.xls netstat.xls wud probably do the same thing..

could you please explain the above ? also, will this actually append the file netstat.xls when there is new data ?

Since the file generated is very difficult to validade manually owing to large number of connections, I need to be certain that it works fine..

thnx in adv,

 

[PHV]

man sort (as already suggested)

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244