awk pipeline 2

[ovince]

hi,

I would like to use file names within awk so I tried something like this:


for file in box*.dat;
do awk 'BEGIN {"echo $file" | getline fl; close("echo $file"); print fl} ' $file
; done

any idea what I do wrong?

thanks

 

[PHV]

In fact, the real question is: what do you want to do ?

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[ovince]

I would like to take ALL box*.dat files, to calculate average of say coulumn 4, and to write out the results into different files that has almost the same name (say DHbox*.tex). What is under * should remain in order to distinguish the files

 

[feherke]

Hi

for file in box*.dat; do
  awk '{t+=$4;n++}END{print t/i}' $file > $file.txt
done

Feherke.

http://rootshell.be/~feherke/

 

[ovince]

thanks. Steel I would like to know why the preliminary idea does not work and how to introduce file names into awk to be able to work with within awk

 

[feherke]

Hi

Steel I would like to know why the preliminary idea does not work

Because the code string with the $file variable was delimited with single quotes ( ' ). The $file is a shell variable and the shell does not expand the variable names in strings quoted with single quotes.

how to introduce file names into awk to be able to work with within awk

awk '{print "the file is : [red]'[/red]$file[red]'[/red]"}' $file

[gray]# or[/gray]

awk [red]"[/red]{print [red]\[/red]"the file is : $file[red]\[/red]"}[red]"[/red] $file

[gray]# or[/gray]

awk [red]-vfile=$file[/red] '{print "the file is : "[red]file[/red]}' $file

[gray]# or[/gray]

awk '{print "the file is : "[red]FILENAME[/red]}' $file

Feherke.

http://rootshell.be/~feherke/