find awk - add line to files?

because_I_can

Hi,

I have numerous files that i want to add a line to.

I can do this using:

find . -name "runme.sh" | xargs awk 'BEGIN {print "setenv whatever"} {print $0};'

But this will only print to std_out. I want to print this directly back into the same file. How can i do this?
Thanks
John

As an added extra i would like to add the "setenv whatever" after the first line of every file. So instead of 'BEGIN i should use????

Thanks again

Tobias Greeshman

To add at the end of a file append the ">>" character to print to the end of a file.

As for the "setenv..." well IIR awk sees data from input files as records and fields and you can modify them as they come in. With this in mind it should be pretty straight forward to redirect the modified output back to file. Use ">" so it will overwrite the old file though.

The Manual for awk is here

daymobrew

I'm not too familar with xargs so I use 'for' loops. This code worked for me.

for f in `find . -name "runme.sh"`
do
  # NR means record/line number. Direct output to temp file.
  awk 'NR == 2 {print "setenv whatever"} {print}' $f > $f.$$
  # Rename the temp file, overwriting the original file.
  mv $f.$$ $f
  # Display something to show progress.
  echo "$f - done"
done

Input file:

Line 1
2nd line
Third line

became

Line 1
setenv whatever
2nd line
Third line

Syth

silas wrote:

Use ">" so it will overwrite the old file though.

Using > will not work. Files are read in in streams, so it'll start overwritting the original file before it's finished reading it, so you'll loose data. Your best bet is to use a temporary file. Look into 'mktemp' to generate a temporary file.

Dero

Would sed not be a better tool for this? Especially with gnu sed as it can change files in place with the -i flag. 2i will insert at the 2nd line of the input file.

sed -i '2i setenv whatever' runme.sh

If your sed doesn't have -i:

sed '2i setenv whatever' runme.sh > tmpfile; mv tmpfile runme.sh

Edit: Didn't notice the 2nd line requirement.