unix shell script
 

How do I cut fields from multiple files and redirect hem to a new file?

This is the script... the only problem I am having is getting the cut command to parse all of the files....


cd /no5odd/cidata
for i in imoddhs*.out
do
SM=`echo $i | sed -e "s/^imoddhs//" -e "s/.out$//"`
echo "$SM\\nRT_MISC\\nBR\\n/unixa/users/cyrtmisc$SM\\n" |odbe> /dev/null
wait
done
cut -f1,6,10 cyrtmisc$SM > cynewrtmisc <<--this line does not work
wait
lpr -s < /unixa/users/cynewrtmisc rop0
rm cynewrtmisc ; rm cyrtmisc*
exit 0

I don't think anyone can help without knowing:

1. The format of your input files
2. the output you expect
3. the output you are actually getting

input files
 

Here is an example of one of the multiple input files - but I am not having a problem with the format - the "cut" utility is just giving me a cannot open error...

# pg cyrtmisc5
4ce TR303INT 672 0 0 5007 CKTNULL ENV_AUTO
4 XENAU FDRS CAC NAV 14 DBNO DBNO

OUTPUT:

# cyrt_misc.sh
Collecting data...
Compiling data... please be patient...
cut: cannot open cyrtmisc5 <<----------------------------
Report completed... printing to switch ROP
cyrt_misc.sh: /unixa/users/cynewrtmisc: cannot open
rm: cyrtmisc* non-existent

manual entry
 

If I enter the command at the prompt - does exactly what I want it to do...

#cut -f1,6,10 cyrtmisc$SM > cynewrtmisc

creates the "cynewrtmisc" file exactly the way I want it to... just won't work out of the script...

What is the point of this? You are just throwing the value away.
In any event your file cyrtmisc5 doesn't exist...

I got it.... thanks anyway...

You can do this much more efficiently by using the shell's facilities, you know:

Since you asked for help, and got it (whether or not it did you any good doesn't really matter) maybe you could be solidary and share your sollution as well

(the only way to learn is by seeing as many sollutions as possible and I'm curious;))

Nearly.:) The only way to learn is by seeing and understanding as many solutions as possible. Curiousity is always good, though.

You're right of course, so let's have it...