Brainfuck interpreter in c printing trouble - c

Question

I'm trying to code a very simple brainfuck interpreter in C, and I run into problems while trying to outprint certain characters by what I understand.
This is all my code:
#include <stdio.h>
int bla(char tabukaz[30000], int ukaz, int num) {
int sum = 0;
int index = ukaz;
while (sum > -1) {
index -= num;
if (tabukaz[index] == ']')
sum += num;
else if (tabukaz[index] == '[')
sum -= num;
}
return index;
}
int main () {
int tab[30000];
int tabukaz[30000];
int c;
int i = 0; int ukaz = 0;
unsigned char ch;
for (int i = 0; i < 30000; i++) {
tab[i] = 0;
tabukaz[i] = 0;
}
while ((c=getchar()) != EOF) {
ch = (unsigned char)c;
if (ch == '>' || ch == '<' || ch == '+' || ch == '-' || ch == '.' || ch == '[' || ch == ']')
{
tabukaz[ukaz] = ch;
}
switch (ch) {
case '>': i++; break;
case '<': i--; break;
case '+': tab[i]++;break;
case '-': tab[i]--; break;
case '.': putchar(tab[i]); break;
case '[':
if (tab[i]==0) {
ukaz = bla(tabukaz, ukaz, -1);
}
break;
case ']':
if (tab[i]!=0) {
ukaz = bla(tabukaz, ukaz, 1);
}
break;
default:
break;
}
ukaz++;
}
return 0;
}
This is the input in question (I tried to avoid the other text in the actual input (keep in mind everything down here is part of the input, even the unnecessary text) We were provided with a make file which will write the output into a text file, and compare it to a predefined text, the problem is my text file comes out as a binary file and I cant figure out why. The problem may be hidden in how I handle [ and ] as I didn't have that problem in the earlier tests without them
+++++ +++++ initialize counter (cell #0) to 10
[ use loop to set 70/100/30/10
> +++++ ++ add 7 to cell #1
> +++++ +++++ add 10 to cell #2
> +++ add 3 to cell #3
> + add 1 to cell #4
<<<< - decrement counter (cell #0)
]
> ++ . print 'H'
> + . print 'e'
+++++ ++ . print 'l'
. print 'l'
+++ . print 'o'
> ++ . print ' '
<< +++++ +++++ +++++ . print 'W'
> . print 'o'
+++ . print 'r'
----- - . print 'l'
----- --- . print 'd'
> + . print '!'
> . print '\n'
As a suggestion made by somebody I did this:
while ((c=getchar())!=EOF) {
ch = (unsigned char)c;
if (ch == '>' || ch == '<' || ch == '+' || ch == '-' || ch == '.' || ch == '[' || ch == ']')
{
tabukaz[ukaz]=ch;
stukaz++;
}
}
while (stukaz>0) {
switch (tabukaz[ukaz]) {
case '>': i++; break;
case '<': i--; break;
case '+': if(tab[i]==255) tab[i] = 0;
else tab[i]++;
break;
case '-': if (tab[i]==0) tab[i] = 255;
else tab[i]--;
break;
case '.': printf ("%c", tab[i]); break;
case '[':
if (tab[i]==0) {
ukaz = bla(tabukaz, ukaz, -1);
}
break;
case ']':
if (tab[i]!=0) {
ukaz = bla(tabukaz, ukaz, 1);
}
break;
default: break;
}
stukaz--;
ukaz++;
}
However the problem now extends to the tests before that, as it even outputs those as binary files, I'm thinking there's something wrong with the [ and ] code and thus it doesn't increment the fields properly printing unwanted characters, how this extended to tests without them only when putting another loop around it I have no idea.
EDIT: the problem with the above loop is not the while loop not going trough, the problem is that it will never get into the switch, any solution to that?

Answer 1

The test is wrong while scanning for a matching bracket.
while (sum > -1) {
index -= num;
if (tabukaz[index] == ']')
sum += num;
else if (tabukaz[index] == '[')
sum -= num;
}
You set num to 1 for a backwards scan, but to -1 for a forward scan.
case '[':
if (tab[i]==0) {
ukaz = bla(tabukaz, ukaz, -1);
}
break;
case ']':
if (tab[i]!=0) {
ukaz = bla(tabukaz, ukaz, 1);
}
break;
So the test should be (sum != 0) so it stops when the brackets are balanced from either direction. Of course, sum is initialized to 0, so I recommend a do { ... } while(); loop so the test is at the end.
Also, remember that you're incrementing the instruction pointer at the end of the loop.
ukaz++;
So you may want to set the pointer to bla(...) - 1, so the increment puts the pointer in the correct place.
You might also like to look at my own brainfuck interpreter, which is very similar to yours. One of my reviewers there gives an excellent explanation of optimizing the loop execution with a jump table.