Files in shell scripts

I have an input file InputFile.csv that contains 2 types of rows. One type is a file name and the second is information:

File1.txt,
Row12,Y
row23,N
File2.txt,
Row1,Y
Row19,Y

The first type is the name of a file that already exists and the second are rows that should be removed or not removed based on the Y or N. What this script needs to do is read the input file and create separate files and put the rows for a given file into these new files. These can then be passed to other utilities for other work.

Code:

while read LINE
do
IFS=","

set -A array ${LINE}

SEG_ONE=${array[0]}
SEG_TWO=${array[1]}

if [[ ${SEG_ONE} == File* ]]
then
RM_FILE=${SEG_ONE}.remove
else
if [[ ${SEG_TWO} == "Y" ]]
then
echo ${SEG_ONE} >> ${RM_FILE}
fi
fi
done < InputFile.csv

for WORK_FILE in `ls *.remove`
do
echo [${WORK_FILE}]
done

I would expect my output to be:
[File1.txt.remove]
[File2.txt.remove]

Instead I am getting:
[File1.txt.remove
File2.txt.remove]

So the ls sees it as one file name. Can anyone explain why and what I can change to make the expected output happen?

 

hi,

the problem comes from the IFS, you should restore it.

to avoid this, you could set IFS "locally" for `read'
(I've removed useless array):

Code:

while IFS=',' read SEG_ONE SEG_TWO
do
...
done < file2read

also, you should tell what shell you're using, it seems it's KSH, that I'm not used to.

and, `ls' is useless inside scripts, for example:

Code:

printf '[%s]\n' *.remove

simply does the job.