Hello everybody.

I have a strange problem (aren't they all )

The challenge:
I have two files
#1 holds a simple list of 3000+ lines, each line beeing an order ID like for example 53267

This file is named ordreliste_2008.txt

#2 holds more than 33000 lines. Each line holds an information string like this
55072;;386;4;;175
the first 5 digits are the order ID.

This file is named ordrelinje.csv

I want to run a while loop iterating file #1 and grep the order ID in file #2.
The result will eventually be >> into a new file

I'm using this simple while loop

while read LINE
do
echo "this is LINE: $LINE" #simple check to see the value of the line
grep "$LINE" ordrelinje.csv
echo $? #simple check to see the result of last command
done < /home/noricom_bergen/ordreliste_2008.txt

My result is all negative - no output

I have tried to write '$LINE', $LINE - same result

If I do this:

while read LINE
do
echo "this is LINE: $LINE" #simple check to see the value of the line
grep 55072 ordrelinje.csv
echo $? #simple check to see the result of last command
done < /home/noricom_bergen/ordreliste_2008.txt

I get an output

The same happens if I put in
while read LINE
do
echo "this is LINE: $LINE" #simple check to see the value of the line
test=52072
grep $test ordrelinje.csv
echo $? #simple check to see the result of last command
done < /home/noricom_bergen/ordreliste_2008.txt

If I however put in test=$LINE the output is nothing


It is only when I use the variable $LINE together with grep that I get this peculiar error.
My conclution is that the grep itself is working ok - I've proved that, but there is a catch with the variable $LINE.

Anyone out there who can help me out?

Hi.

Just a guess, but your files might have DOS linebreaks. Try running dos2unix on ordreliste_2008.txt first.

Dave

From what I understand, $LINE will contain the full line, i.e., 55072;;386;4;;175. If you really only want the ID, you have to split the line, one simple way is to use command substitution:

${LINE%%;*} will remove everything after (and including) the first semicolon encountered in the string.


EDIT: Sorry, I have just seen that I misunderstood what you were trying to do.

I would suggest

to make sure there are no newlines or whitespace around $LINE and if there is, go from there.

I agree that there probably are invisible characters in $LINE.

Your script appears to work fine when file #1 contains only numbers.
You could try cleansing your file with this command:
(after you make a backup just in case you need it)

The command searches for anything that is not a number and replaces it with a single new-line.