Hi all, I'm trying to modify a file using ksh script. I need to parse the file and then look for a specific occurence of a pattern and then assign it a value. After that I need to save the file with the same filename. Is there any way to accomplish it without creating a temporary file ? For modification, which is a better bet (sed/awk) ? Thanks in advance.

Hi abby, it is unusal for unix commands to overwrite him self. So choose C, C++. No RISK no fun Frank

Frank, Seems like you wanna do "Search and Replace" operation on a text file right? For example if you want to find every single occurrence of "string_1" in "my_file" and replace it with "string_2" you might try following "perl -i.bak -p -e 's/string_1/string_2/g;' my_file" Remove the *.bak file if you don't want to keep it If you do not have perl on your system you can try "sed 's/string_1/string_2/g;' my_file" This will send changed file to std out, that you can save later Hope this helps - Faraz

Faraz, read the question from abby and give here an answer. Seems so that you only read my answer and comment it. I didn't see any value in your posting to the question "how to make changes in a file whitout creating a temp file". No RISK no fun Frank