work on each line returned by GREP

[IMAUser]

Hi,

I need to do something like

for i `ls -1 *.sql `
do
for k in `grep "MY TEXT" $j `
do
..
...
..
done

done


And one of the sql files may contain lines like

MY TEXT ABC
fkjgnhdfg
MY TEXT PQR
dfkljgn
MY TEXT XYZ

So a grep would return three lines and I want to work on the three lines separately, but with the command syntax I have in my script each line comes out as three lines because of the space in the three words.

Is there a way to make grep return one line and not three in the k-shell script.

Many Thanks,

 

[MoreFeo]

How about something like

...
   grep "MY TEXT" $j | while read k;
   do
      #work with $k
   done
...

 

[columb]

If you mean that 'MY TEXT ABC' is seen as three separate variables try

IFS='
'

This sets the input record separator to newline (the default is space) so that each line is seen as a separate record.

Ceci n'est pas une signature
Columb Healy

 

[feherke]

Hi

Just to make the smart @$$. ;-)

This sets the input record separator to newline (the default is space)

Includes space, but not only space.

IFS
Internal field separators, normally space, tab, and new-line that are used to separate command words [gray](...)[/gray]

Feherke.

http://rootshell.be/~feherke/

 

[IMAUser]

I used the solution posted by MoreFeo and it worked the trick. Will try the others later on...

Thanks for the suggestions.