Shell Scripting Help .. Please Help

[Praveen Hari]

Hi All,
I am trying write a shell script to accomplish following:

This is what I want:
I have a file1 with following content:
xxx1 100 105
xxx2 200 210
xxx3 150 175

For each line of this file, I need to genearte a new file with contenats and file name
in the pattern:

file: xxx1.bif
http://yyy.com/test.cgi?cmd=item-100&group=xxx1
http://yyy.com/test.cgi?cmd=item-101&group=xxx1
..
..
http://yyy.com/test.cgi?cmd=item-105&group=xxx1



file: xxx2.bif
http://yyy.com/test.cgi?cmd=item-200&group=xxx2
http://yyy.com/test.cgi?cmd=item-201&group=xxx2
..
..
http://yyy.com/test.cgi?cmd=item-210&group=xxx2


I am trying following script and it doesn't work:

infile=test.txt
url="http://yyy.com/test.cgi?cmd=item-"
uri="group="
suffix=".bif"

cat $infile | while read name start end
do
filename=`expr $name$suffix`
if [$end -gt $start]
then
while [$end -ge $start]
do
echo $url$end$uri$name >> $filename
echo "<>" >> $filename
end=`expr $end-1`
done
fi

done



Any help is appreciated. The inner loop fails.

[Steven E. Protter]

Whenever you say a script fails, provide the error message. That way maybe someone can diagnose by site instead of having to copy the script and data and run the script.

Not knowing the error code, I'll suggest this:

change this:
echo $url$end$uri$name >> $filename

to

echo "${url} ${end} ${uri} ${name}" >> $filename

The spaces may not be necessary. If I can't sleep I'll actually try and run your script and debug it later.

SEP
Will work for points.

[Praveen Hari]

I tried changing it as you specified:
I get following message when I execute the script:

./test.csh: [105: not found
./test.csh: [210: not found
./test.csh: [175: not found


I believe it may be due to some syntax error. But I am not able to figure out what it is...
Thanks
Praveen

[Mark Grant · Accepted solution]

Problem seems to be lack of spaces in your expr.

Assuming you don't mind the "<>" after each entry in each file, the following works.

infile=test.txt
url="http://yyy.com/test.cgi?cmd=item-"
uri="&group="
suffix=".bif"

cat $infile | while read name start end
do
filename=`expr $name$suffix`
if [ $end -gt $start ]
then
while [ $end -ge $start ]
do
echo $url$end$uri$name >> $filename
echo "<>" >> $filename
end=`expr $end - 1`
done
fi

done

[Mark Grant]

Problem seems to be lack of spaces in your expr.

Assuming you don't mind the "<>" after each entry in each file, the following works.

infile=test.txt
url="http://yyy.com/test.cgi?cmd=item-"
uri="&group="
suffix=".bif"

cat $infile | while read name start end
do
filename=`expr $name$suffix`
if [ $end -gt $start ]
then
while [ $end -ge $start ]
do
echo $url$end$uri$name >> $filename
echo "<>" >> $filename
end=`expr $end - 1`
done
fi

done

[Mark Grant]

Problem seems to be lack of spaces in your expr.

Assuming you don't mind the "<>" after each entry in each file, the following works.

infile=test.txt
url="http://yyy.com/test.cgi?cmd=item-"
uri="&group="
suffix=".bif"

cat $infile | while read name start end
do
filename=`expr $name$suffix`
if [ $end -gt $start ]
then
while [ $end -ge $start ]
do
echo $url$end$uri$name >> $filename
echo "<>" >> $filename
end=`expr $end - 1`
done
fi

done

[ALPER ONEY]

hi,
try to give a space before the value being tested in every "test" command.
Ex:instead of if [$end -gt $start] or while [$end -ge $start], please use if [ $end -gt $start ], while [ $end -ge $start ].
Regards.
ALPER ONEY

[Massimo Bianchi]

I think the inner loop fails because you re-assign the variable of the for, and at exit there is a big mess.

try this:

cat $infile | while read name start end
do
START=0
FINAL=0
filename=$name$suffix
if [$end -gt $start]
then
START=$start
FINAL=$end
while (($FINAL >= $START))
do
echo $url$START$uri$name >> $filename
echo "<>" >> $filename
let START=START+1
done
fi

done


Massimo

[john korterman]

Hi,
if you combine the advice given by Alper and Massimo, I think you should be there.

regards,
John K.

[Rodney Hills]

Here is a short script that uses a perl 1-liner to do what you want.

#!/usr/bin/sh
for $file in file* ; do
perl -ane 'print "http://yyy.com/test.cgi?cmd=item-$_\&group=$F[0]\n" foreach ($F[1]..$F[2])' $file >${file}.bif
done

perl is handy for breaking up text and generating other data.

HTH

-- Rod Hills

[Praveen Hari]

Mark,Alpher,MAssimo,

Your all suggestions worked.
Rodney, I haven't tried Perl. I will try.

Thanks all for quick answers.
I like this forum.

Praveen