UNIX for Dummies Questions & Answers

Hi all,

Here is my problem:

I'm testing a login script for our mac Machines. This script output's in a csv file.

The login script looks like this:

Code:

echo "${computername}|${data1}|${data2}|${data3}" >> file.csv

The csv file looks like this:

Computername1|data1|data2|data3
Computername2|data1|data2|data3
Computername2|data1|data2|data3
Computername2|data1|data2|data3
Computername3|data1|data2|data3

Now I want to erase the data from computername2 because I only want to see the last output from every computernameX.

So before the login script output's the "Computername2|data1|data2|data3" line I want it to erase it's previous output.
When I use

Code:

grep -v Computername2 file.csv > file2.csv

the file2.csv is empty.
When I only use grep (without -v) the file2.csv is similar to file.csv.
When I manually edit the file.csv file en hit enter after all the lines and save the file then it works fine.
So somehow the grep command "sees" only one line. Now I searched the web for all day now but I can't figure out what to do. I think it has something to do with new lines or carriage return.

So what to do? Maybe use another command then grep? Or change the way of output in the login script?

Code:

perl -ne '@record=split/\|/;$record{$record[0]}=join "|", $_; _END_ {for (sort keys %record){print $record{$_};}' file.csv

Thanks for your reply.

The first error I got is:

Missing right curly or square bracket at -e line 1, at end of line
syntax error at -e line 1, at EOF
Execution of -e aborted due to compilation errors.

So I added the right curly bracket at the end like this:

Code:

perl -ne '@record=split/\|/;$record{$record[0]}=join "|", $_; _END_ {for (sort keys %record){print $record{$_}};}' file.csv

So then I get this error:

Can't call method "_END_" without a package or object reference at -e line 1, <> line 1.

I must say I have no experience with perl and I'm pretty new to unix as well. Still I wonder if there is a way to change the way of output of the initial data in the first place. Since the grep command actually works fine when I manually hit enter after all the lines.

if grep -v Computername2 is not working, can you try awk instead:-

Code:

awk -F\| '$1!="Computername2"' file.csv

Awk doesn't seem to work either. If I use:

Code:

awk -F\| '$1!="Computername2"' file.csv >> file2.csv

File2 is the same as file1. If I output to the terminal window it output's everything in one single line. So again that's where the problem seems to be.

---------- Post updated 29-11-12 at 12:00 PM ---------- Previous update was 28-11-12 at 04:02 PM ----------

SOLVED!

The problem was in the first output. Every line ended with a Carriage Return (CR). That's why the grep command "sees" only one line of data. For the grep command to work every line has to end with a Carriage Return AND a Line Feed (CR+LF).

So what I did was change the way of output of the initial data:

Code:

echo "${computername}|${data1}|${data2}|${data3}\r" >> file.csv

As you can see I added the \r in the end. By doing this you get the CR and LF at the end of every line.

After this the grep command worked fine.

Thanks everyone for your help.