Bash script to check for diff - linux

Question

I'm trying to get this script to work for checking the output of a program (test21) against known correct result text files. I've tried playing around with a couple ways of writing it, but I can't get it to work. Any ideas, thanks!
#!/bin/bash
arr=$(find resultfile*)
x=0
for i in $(find resultfile*);do
arr[$x]=$i
x=$(($x + 1))
done
for i in ${arr[#]}; do
echo $i
done
x=0
for i in $(find testfile*);
do
if diff ${arr[$x]} <(./test21< testfile1.txt 2>&1); then
echo Everything is correct
else
echo Everything is wrong
fi
x=$(($x+1))
done
It says for this,I get:
Segmentation fault: 11 ./test21 < "testfile1.txt" 2>&1
My program test21 takes in an input text file and parses it, which could result in the segmentation faults. But if I use ./test21 testfile1.txt my program runs perfectly fine. I hardcoded in testfile1.txt, but it should be
if diff ${arr[$x]} <(./test21< $i 2>&1); then
However, it's reading it as $i, even if I do "$i".

Answer 1

I get:
Segmentation fault: 11 ./test21 < "testfile1.txt" 2>&1
My program test21 takes in an input text file and parses it, which
could result in the segmentation faults. But if I use ./test21 testfile1.txt my program runs perfectly fine.
Then just use that; instead of
./test21< testfile1.txt 2>&1
write
./test21 testfile1.txt 2>&1

Answer 2

Lets say that you want to run a program myprogram which will use input.txt file as input.
You want to compare output of your program with some known correct result which is written in correct.txt file. We will save output of your program in actual.txt file.
#!/bin/bash
./myprogram < input.txt > actual.txt
cmp -s actual.txt correct.txt
if [ $? -eq 1 ]; then
echo "WRONG"
else
echo "CORRECT"
fi
Note that actual.txt will capture only output that came from stdout. If out want to redirect both stdout and stderr you then want to use this command:
./myprogram < input.txt &> actual.txt
If you want to ignore stderr and throw it in "black hole" you will use:
./myprogram < input.txt 2> /dev/null > actual.txt