Shell Programming and Scripting

Hi again,


Can someone please help me on extracting only certain columns from any text (Tab Delimited).


Eg.

Read all the text files under zzz directory (Tab Delimited).

If the cell value in column 8 contains the following strings ( HP OR IBM OR Dell)

Then
Copy Colum 1, Colum 4, Colum 8.



I tried this code but it didn't work.


Find . -name '*PC*.txt' | gawk -F\\\t $1,$4,$8 '$8=="HP" or "IBM" or "Dell" {print}'


thanks in advance

You need to repeat the comparison operator, the == doesn't accept a list of alternatives with "or" between them on the right hand side.

Code:

find . -name '*PC*.txt' |
gawk -F '\t' '$8 == "HP" || $8 == "IBM" || $8 == "Dell" { print $1, $4, $8 }'

Also note that the fields to print are passed to the print function, not given as command-line parameters.

You could use a regular expression match like $8 ~ /^(HP|IBM|Dell)$/ if you want to make it more succint.

As per your problem description, you could replace the find with gawk -options 'script' zzz/*PC*.txt

Hi era,



Thanks alot ,

I am using this gawk -options 'script' zzz/*PC*.txt

and it works .

But I have a minor issue, that all the txt files are not located under the ZZZ folder.

There are other folders under zzz which contains txt files.


How can I reference the input directory to all files inside the zzzz and any sub folders?

For some reason this script didn't work.
find . -name '*PC*.txt' | gawk -F '\t' '$8 == "HP" || $8 == "IBM" || $8 == "Dell" { print $1, $4, $8 }'

Once again thanks for your help.