Running sed and counting number of lines processed

Shell Programming and Scripting
1 
/bin/sed -n ';4757335,$ p' | wc -l
/bin/sed -n ';4757335,$ p' | egrep "Failed" | egrep -c "PM late arrrival"

how can i combine the above two sed commands into one? i want to count the number of lines between the specified line number and the end of the file. AND and i want to count how many lines contain a specified set of strings within those lines.

i dont want to run these sed commands separately because they would add greatly to the time it takes for the script to complete.

2

I think using awk is the best option here:

awk 'NR>4757335&&/Failed/&&/PM late arrrival/{c++}NR>4757335{d++}END{print c,d;}' filename
1 Like
3

wonderful!!! i have yet to test it, but could you please explain your beautiful awk command to me? please????

4

SkySmart, please test it and let me know if it is giving the desired o/p, I will explain the code.

5

it works. the command returns two numbers:

35300 313306
6

NR>4757335&&/Failed/&&/PM late arrrival/ - Lines greater than line number: 4757335 and matches pattern: Failed and matches pattern PM late arrrival

{c++} - Increment variable: c for lines that matches all 3 conditions above.

NR>4757335 - Lines greater than line number: 4757335

{d++} - Increment variable: d for lines that matches condition above.

END{print c,d;} - In the end print value of variables: c and d

1 Like