Yet more shell scripting

[Wraith]

I have been visiting this forum for a couple of weeks now, originally searching for shell scripting advice. Some threads here have been absolutely invaluable.

I have finally got stuck in my Assignment for uni though, I am trying to write a script to calculate missed pracs for students.

I have three files containing prac marks over a semester, they look a little like:

#Prac Marks for class 1
#fomat is SID MARK
21353535 10
24535365 6
34213444 9

I need to compile a file of missed pracs, where no mark is recorded in any particular prac.

what I have so far is:

cut Students -f1 -d: | grep -v \# > Missing-temp; #extract SID from the first column of my Students file.

while read sid extra; do #read SID using extra to catch any extra stuff not needed

if cut Prac1 -f1 -d ' ' | grep -v \# | grep "$sid"; then #cut field 1 out of Prac1, this is the SID field, remove any lines starting with # using grep then find whether the relevant SID is in the Prac file.
echo "Found $sid" >>error; #just for debugging purposes
else
echo $sid: Prac1, >> Missing1; #this is my required format
fi;

if cut Prac2 -f1 -d ' ' | grep -v \# | grep "$sid"; then
echo "Found $sid">>error;
else
echo $sid: Prac2, >> Missing2;
fi;

if cut Prac3 -f1 -d ' ' | grep -v \# | grep "$sid"; then
echo "">>error;
else
echo $sid: Prac3, >> Missing3;
fi;

done < Missing-temp; #pipe Missing-temp into read, end loop

join -a 1 -a 2 Missing1 Missing2 | join -a 1 -a 2 - Missing3 >> Missing4; #join files together, using -a to ignore any missing bits

sed 's/,$//' <Missing4 | sort -g >Missing; # removes trailing commas and sorts numerically.

rm -f Missing1 Missing2 Missing3 Missing4 Missing-temp; #get rid of temp files


The script works fine on my test data but when I submit it, I get the following error:
File: Missing
Expected: 312349311: Prac1, Prac2
Got: 312349311: Prac1

I have tested for this kind of output in my test files and it works fine. Does anyone have any suggestions as to why the script is not working?

[ghostdog74]

you can use awk

awk -v file="Prac1" ' BEGIN {  # get all sids of Prac1 into array
		while( (getline line < file )>0 ) { 
			if (line ~ /^#/ ) continue #skip comments 
			mprac[line]
		}
      }
      /^#/ {next} #skip comments line in Students file
	{ 		
		if (!( $1 in mprac )) {
			print "Sid: " $1 " not in " FILENAME
		}
	
	}
	
' "Students"

[Wraith]

Thanks for the advice. I finally figured out that I needed to sort my Prac files before joining them *doh!*

Oh, well