File decimation

[ooids]

Hi,

I need to write a script that decimates a file by a user specified amount.
e.g. Keep every 5th, 10th, 30th line etc.

Is there a standard unix command that i can use to do this or should i be looking at sed or awk?

Any ideas would be great.

Cheers!

 

[feherke]

Hi

For example, to output each third line, so 3, 6, 9, ...

sed -n '3~3p' inputfile
[gray]# or[/gray]
awk '!(NR%3)' inputfile

Feherke.

http://rootshell.be/~feherke/

 

[ooids]

Thanks feherke,

The sed command fails - "Unrecognized command: 3~3p" - any ideas why?

I can get the awk to work but only using "nawk" so thanks for that.

Cheers!

 

[ooids]

Update to my last post - the sed command runs on my Linux (Redhat) box - just not on my Sun (Solaris 2.8).

Thanks for the help!

 

[feherke]

Hi

Yes, there are incompatibilities between [tt]sed[/tt]s. I have GNU [tt]sed[/tt] 4.1.2 and GNU [tt]awk[/tt] 3.1.1. But the [tt]awk[/tt] code works with the [tt]gawk[/tt] --traditional option, so theoretically that is backward compatible. Seems to be exceptions too.

Feherke.

http://rootshell.be/~feherke/

 

[vgersh99]

to print every 3-rd line:

sed -n 'n;n;p;' file

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

 

[ooids]

Thanks,

Can you expain "n" and "p" - i have rarely used sed...

 

[vgersh99]

doing 'man sed' yields:

-n Suppress the default output.

n Copy the pattern space to the
standard output if default
output is not suppressed
Replace the pattern space with
the next line of input.

p Print. Copy the pattern space
to the standard output.

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

 

[ooids]

I was more after an example of decimation using your notation as it differs from the previous example that worked...

Cheers!