Remove the footer from a flat file by searching a pattern

Shell Programming and Scripting
1

Hi,
I have more than 30,000 records in a flat file. I want to remove footer from the file by searching a string pattern for the footer.
Example.... let the flat file test.dat contains below records.

record1
record2
record3
..
..
..
record31000
Total records 31000

I want to remove the last line ("Total records 31000") from the file by searching for the string "Total records". I want to modify the input file by removing the line, i don't want to use a separate file as output file due to the large size of the input file. please suggest me.

Thanks in advance.

2

Try this...

sed -i '/Total records/ d' input_file

--ahamed

3

Thanks Ahamed.
But I think your code needs an output file for writing the output. I need to modify the input file as the size of the input file is large.

4

No, the "-i" option will modify the original file itself!

--ahamed

5

using awk ..

$ awk '!/Total records/' infile
6

Hi ahamed,
I think the option "-i" is shell specific. This command is working fine in bash, but not working in ksh. Please suggest me if you have any suggestions(code). thanks...

Hi jayan_jay,
This command is fine but needs an output file for writing the output. I need to modify the input file as the size of the input file is large. Please suggest me if you have any idea for this. Thanks...

7

which OS are you using? If Solaris, try using /usr/xpg4/bin/sed

--ahamed