UNIX for Dummies Questions & Answers

Hi Gurus,

I have a file which has some fields separated with multiple spaces or single space.

data
1 2 3 4 5 6
4 5 5 7 7 8 9
4 6 10

and so on.....


The problem I am facing is the output of the awk program automatically squeeze the multiple spaces to single space.

I used the command

Code:

awk '{$1=1;print $0}' data

the output I got
1 2 3 4 5 6
1 5 5 7 7 8 9
1 6 10

I need the multiple space format(the actual content)

And I think this happens pint $0 is behaving like $1,$2 and so on.(I also want to know if that then why it is not behaving in the same way when we run "awk '{print $0}' file_name)

Longlive,
Sanjay

If you only want to replace 1st field to 1,

Code:

sed '/\([0-9]*\)/1/' data

OR (if 1st field has only one length data)

Code:

sed '/\(.\)/1/' data

OR

Code:

sed '/\([^ ]*\)/1/' data

* can be removed if 1st field is of lenght 1.
I guess when we say

Code:

$1=1

in awk, it splits all fields and combines all them as per FS value (which is single space) to make $0. When we don't use $1, $2 etc in awk, field splitting doesn't occur, so spacing is retained. Experts may comment on this awk behavior.

I dont know how standard this is (works on gawk), but you can use "split" with a regular expression to split up the arguments, and this doesnt compress down multiple spaces:

Code:

split($0,a,"\\s");

you can then use a[1]...a[n] for your column values