Hello,
I have a file with some columns of numeric data e.g.

5 5
5 6
5 7

I need to total them, currently im using the following method


for i in cat <filename> | awk print column2 (for example)
do
total=3D$i
total=3D=E8xpr $total + $i`
done

However this method is really slow on large files, could someone
recommend a quicker way?

Thanks

[ Post a follow-up to this message ]

On 2006-01-23, cconnell_1@lycos.com wrote:
> Hello,
> I have a file with some columns of numeric data e.g.
>
> 5 5
> 5 6
> 5 7
>
> I need to total them, currently im using the following method
>
>
> for i in cat <filename> | awk print column2 (for example)
> do
> total=$i
> total=èxpr $total + $i`
> done
>
> However this method is really slow on large files, could someone
> recommend a quicker way?

To add the second column:

awk -v col=2 '{ total += $col } END { print total }' FILENAME

To add the first column, change col=2 to col=1

--
Chris F.A. Johnson, author   |    <http://cfaj.freeshell.org>
Shell Scripting Recipes:     |  My code in this post, if any,
A Problem-Solution Approach  |          is released under the
2005, Apress                 |     GNU General Public Licence

[ Post a follow-up to this message ]

Chris F.A. Johnson wrote:
> On 2006-01-23, cconnell_1@lycos.com wrote: 
>
>    To add the second column:
>
> awk -v col=3D2 '{ total +=3D $col } END { print total }' FILENAM
E
>
>    To add the first column, change col=3D2 to col=3D1
>
> --
>    Chris F.A. Johnson, author   |    <http://cfaj.freeshell.org>
>    Shell Scripting Recipes:     |  My code in this post, if any,
>    A Problem-Solution Approach  |          is released under the
>    2005, Apress                 |     GNU General Public Licence


Thanks, I think there is a typo somewhere as I get byte@osn01:/tmp-->
awk -v col=3D2 '{ total +=3D $col } END { print total }' test
awk: syntax error near line 1
awk: bailing out near line 1

Any ideas?

[ Post a follow-up to this message ]

On 2006-01-23, cconnell_1@lycos.com wrote:
>
> Chris F.A. Johnson wrote: 
>
> Thanks, I think there is a typo somewhere as I get byte@osn01:/tmp-->
> awk -v col=2 '{ total += $col } END { print total }' test
> awk: syntax error near line 1
> awk: bailing out near line 1
>
> Any ideas?

Are you using Solaris? If so, use nawk or xpg4/awk (I forget the
exact location).


--
Chris F.A. Johnson, author   |    <http://cfaj.freeshell.org>
Shell Scripting Recipes:     |  My code in this post, if any,
A Problem-Solution Approach  |          is released under the
2005, Apress                 |     GNU General Public Licence

[ Post a follow-up to this message ]

Yes its solaris.
nawk works fantasic.Thx

[ Post a follow-up to this message ]

>>>>> "cconnell" == cconnell 1 <cconnell_1@lycos.com> writes:

cconnell> for i in cat <filename> | awk print column2 (for example)

Why the unnecessary cat?

Why not just:

cat filename | cat | cat | cat | tr | cat | awk ...

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment PERL training
!

[ Post a follow-up to this message ]

On 2006-01-23, Randal L. Schwartz wrote: 
>
>cconnell> for i in cat <filename> | awk print column2 (for example)
>
> Why the unnecessary cat?
>
> Why not just:
>
>         cat filename | cat | cat | cat | tr | cat | awk ...

usage: tr [-csu] string1 string2
tr [-cu] -d string1
tr [-cu] -s string1
tr [-cu] -ds string1 string2

--
Chris F.A. Johnson, author   |    <http://cfaj.freeshell.org>
Shell Scripting Recipes:     |  My code in this post, if any,
A Problem-Solution Approach  |          is released under the
2005, Apress                 |     GNU General Public Licence

[ Post a follow-up to this message ]

While on the subject, if one wants to print selective columns in a file
using awk, is it possible to do it without cat to make it faster?
i.e I always pipe the output from a file into awk from cat, can you use
awk on its own?
Thanks

[ Post a follow-up to this message ]

cconnell_1@lycos.com wrote:
> While on the subject, if one wants to print selective columns in a file
> using awk, is it possible to do it without cat to make it faster?
> i.e I always pipe the output from a file into awk from cat, can you use
> awk on its own?

Yes.  In general, you can almost always eliminate redundant "cat"s
by changing this form:

cat foo | bar

into this form:

bar < foo

With awk, there is an even shorter way, because awk can take an input
filename as a command line argument directly (just like cat does), so
you can change this

cat /etc/passwd | awk -F: '{print $1}'

into this:

awk -F: '{print $1}' /etc/passwd

You could of course use redirection so that the shell opens the file
and makes it awk's standard input:

awk -F: '{print $1}' < /etc/passwd

but that more typing.

- Logan

[ Post a follow-up to this message ]

On 2006-01-24, cconnell_1@lycos.com wrote:
> While on the subject, if one wants to print selective columns in a file
> using awk, is it possible to do it without cat to make it faster?
> i.e I always pipe the output from a file into awk from cat, can you use
> awk on its own?

Did you look at the solutions you were given? Don't they answer
your question?

--
Chris F.A. Johnson, author   |    <http://cfaj.freeshell.org>
Shell Scripting Recipes:     |  My code in this post, if any,
A Problem-Solution Approach  |          is released under the
2005, Apress                 |     GNU General Public Licence

[ Post a follow-up to this message ]