Hai

I am using

bash-2.03$ bash --version
GNU bash, version 2.03.0(1)-release (sparc-sun-solaris)

I am not able to use gawk command its showing command not found , why ?

Eg:
awk 'NR==1' fix.txt | gawk 'BEGIN { FIELDWIDTHS = "3 2" } { printf($1"|"$2); }'
---------------------------------------
bash: gawk: command not found
Broken Pipe

I want to find num of columns with whitespace as field seperator, is it possible with awk

Thank u

Not sure whether gawk is available in Solaris!

this will not work...as one column may also contain null(space) value..
the null value will be considered as seperator.

Eg:Inputfile
3445 56 67
23 2 45
07 06


output of 3rd line 2nd column must be null(empty)

Yes of course, it would be considered as a seperator!

So, may I know what is your expected output from the input file considering the null space also as a separator ?

3445 56 67
23 2 45
07 space 06


My out put for 3rd line 2nd column must be null(space) and should not be 06.
While counting number of column it should show as 3 column

Sorry, but I think that your logic is wrong. How would you split this up:

3445space56space67
23space2space45
07spacespacespacespace

Does this mean that the columns for the last row should be 3? And if you are talking about NULL value - that would mean that the value simply does not exist and (in this case) the third row will have be:
07space06

Then how do you decide whether 06 is the second column or the third?

Because my data is Fixed width column...with that i can get 1st column and 2nd ..and so on