Last Friday as I was working, I was asked to get some information from the application logs for a particular period of time. I needed to extract some data and count the results. The given period was last twelve months.

Luckily I am working with Linux, so I opened a shell session and typed the following after doing some research on the Internet:

[java]
grep -B4 "String1" *.log | grep "String2" | cut -d -f2 | sort -u | wc -l
[/java]

Basically what the command does is:

1. In all files with extension log , grab 4 lines before occurrence of “String1″ and pass the outcome result to the next command
2. Find occurrence of “String2″ and pass the outcome result to the next command
3. Split by SPACE delimiter and grab the second field and pass the outcome result to the next command
4. Sort in unique manner and pass the outcome result to the next command
5. Count the results and output

Simple and efficient. The whole process took only few seconds.

I do not think I could have done this so easily if I would be working on Win platform…

You gotta’ love Linux!