Shell Programming and Scripting

Just wondering if someone could assist me with shell script I'm trying to write. I need to read the final column of a text file (shown below) and workout what the average number is. The text file will have a variable number of lines, I just want the script to pull out the values in the final field (which I've done using cut), add them all together then divide by the total number of rows. Really struggling with this! Help much appreciated

abcd e544f222e0jk55 1503
abcd e544f222e0jk56 1504
abcd e544f222e0jk57 1505
abcd e544f222e0jk58 1506
abcd e544f222e0jk59 1507
abcd e544f222e0jk60 1508
abcd e544f222e0jk61 1509
abcd e544f222e0jk62 1510
abcd e544f222e0jk63 1511
abcd e544f222e0jk64 1512
abcd e544f222e0jk65 1513
abcd e544f222e0jk66 1514
abcd e544f222e0jk67 1515
abcd e544f222e0jk68 1516
abcd e544f222e0jk69 1517
abcd e544f222e0jk70 1518
abcd e544f222e0jk71 1519
abcd e544f222e0jk72 1520
abcd e544f222e0jk73 1521
abcd e544f222e0jk74 1522
abcd e544f222e0jk75 1523
abcd e544f222e0jk76 1524
abcd e544f222e0jk77 1525
abcd e544f222e0jk78 1526
abcd e544f222e0jk79 1527
abcd e544f222e0jk80 1528
abcd e544f222e0jk81 1529
abcd e544f222e0jk82 1530
abcd e544f222e0jk83 1531
abcd e544f222e0jk84 1532

Hi.

One with awk...

Code:

$ cat file1
abcd e544f222e0jk55 1503
abcd e544f222e0jk56 1504
abcd e544f222e0jk57 1505
abcd e544f222e0jk58 1506
abcd e544f222e0jk59 1507
abcd e544f222e0jk60 1508
abcd e544f222e0jk61 1509
abcd e544f222e0jk62 1510
abcd e544f222e0jk63 1511
abcd e544f222e0jk64 1512
abcd e544f222e0jk65 1513
abcd e544f222e0jk66 1514
abcd e544f222e0jk67 1515
abcd e544f222e0jk68 1516
abcd e544f222e0jk69 1517
abcd e544f222e0jk70 1518
abcd e544f222e0jk71 1519
abcd e544f222e0jk72 1520
abcd e544f222e0jk73 1521
abcd e544f222e0jk74 1522
abcd e544f222e0jk75 1523
abcd e544f222e0jk76 1524
abcd e544f222e0jk77 1525
abcd e544f222e0jk78 1526
abcd e544f222e0jk79 1527
abcd e544f222e0jk80 1528
abcd e544f222e0jk81 1529
abcd e544f222e0jk82 1530
abcd e544f222e0jk83 1531
abcd e544f222e0jk84 1532

$ awk '{T+= $NF} END { print T/NR }' file1
1517.5

many thanks for the reply scottn, I get the following:

awk: line 1: syntax error at or near END

script is:

Code:

#!/bin/sh
ls -l logs/db
#echo "please enter the name of the file you wish to analyse:"
#read filetoint
cat stats.txt
awk '!(T+= $NF) END { print T/NR }' stats.txt

Hi.

I changed my awk slightly. Try that one.

It doesn't look like the error from Solaris, but if it is, use /usr/xpg4/bin/awk or nawk.

change

Code:

awk '!(T+= $NF) END { print T/NR }' stats.txt

to

Code:

awk '!{T+= $NF} END { print T/NR }' stats.txt

---------- Post updated at 22:06 ---------- Previous update was at 22:02 ----------

@scottn

I've found a 6 yr. old posting of Ygor here.
Now I'm almost able to "read" your code.
So "T" means Total and "$NF" means 3rd column, right?
I really can't understand how "$NF" stands for 3rd column....

many thanks for this, I'm actually scripting it on ubuntu 9.10? now returns the following error:

awk: line 1: syntax error at or near {

Hi pseudocoder.

Yes, I'm sure this question has been asked and answered many many times.

The first awk you quote is my original one, which I changed, but which rich@ardz quoted.

It works fine (at least for me )

Code:

$ awk '!(T+= $NF) END { print T/NR }' file1
1517.5

T is a running total, $NF is the value of the last field and NR is the number of rows.

---------- Post updated at 10:13 PM ---------- Previous update was at 10:10 PM ----------

Can you paste (i.e. not type) exactly the command you are running? This awk is not by any means complicated.

Thanks.