Shell Programming and Scripting

Hi All,

I need to write a script that will go through 600+ files and perform find and replace. I was going to use sed but there is a level of complexity that is doing my head in.

To explain: I have 600+ files that have a line in them that reads (for example)

FILE=DCLCLHST
REMARKS='CLAIM_HISTORY' $
SEGNAME=DCLCHST, SEGTYPE=SO
--Followed by anothe 50 lines or so---

In the line REMARKS='CLAIM_HISTORY' $ the $ is supposed to have a , in front so it reads REMARKS='CLAIM_HISTORY' ,$

I have played around with awk and came up with awk '/REMARKS/{if ($2=="$") $2=",$"; print > FILENAME}' *.mas

Of course the problem here is that the entire file gets over written and only the REMARKS line (correct mind you) exists within the file.

So I guess the question is how can I perform a find and replace on the $ only on the line that contains the word REMARKS without overwriting the rest of the file. Thanks in advance

thats because u missing the else part.. in else part simply print the $0

Code:

sed -i.bak '/^REMARKS/s/\$/,$/' *.mas

Now, your *.mas files are modified, and originals are backed up as *.mas.bak

Forgive me picking holes in this but I am trying to get to grips with sed.

In ksh I get an illegal option message for the -i. Taking that away the original files aren't modified, just echoed to stdout.

Thanks for showing me the '/^REMARKS pattern in front of the substitution, I wasn't aware you could do that.

Cheers

-i option is not POSIX® standard , see man sed

OP you have to use a temporary file.