A script for detecting in-your-face plagiarism in C source files

gotta_catch_em_all

#!/bin/bash


FILES=($*)
NUM_FILES=${#FILES[@]}

function num_diffs() {
    i=1
    for f in $*; do
        cat $f | uncrustify -q -c uncrustify.conf | grep -v "#" > tmp.c
        cpp tmp.c > F$i.c
        i=$((i+1))
    done
    diff -d -c1 -wiB F1.c F2.c | grep -- "^---" | wc -l
}

echo -n "  ID,  NAME"
for ((i=0; i < NUM_FILES; i++)); do
    printf ",%4d" $i
done
echo ""

for ((i=0; i < NUM_FILES; i++)); do
    printf "%4d," $i
    echo -n ${FILES[i]} | head -c 6
    for ((j=0; j < i + 1; j++)); do
        echo -n ",  NA"
    done

    for ((j=i+1; j < NUM_FILES; j++)); do
        OUT=`num_diffs ${FILES[$i]}  ${FILES[$j]}`
        printf ",%4d" $OUT
    done
    echo ""
done