Forum
What's wrong with this code? - Printable Version

+- Forum (https://www.boriel.com/forum)
+-- Forum: Compilers and Computer Languages (https://www.boriel.com/forum/forumdisplay.php?fid=12)
+--- Forum: ZX Basic Compiler (https://www.boriel.com/forum/forumdisplay.php?fid=11)
+---- Forum: Help & Support (https://www.boriel.com/forum/forumdisplay.php?fid=16)
+---- Thread: What's wrong with this code? (/showthread.php?tid=337)



What's wrong with this code? - LTee - 03-24-2011

Okay, this is driving me mad. :-D

I wrote a little routine to 'shuffle' the numbers 1 to 10 into an array, randomly. It's not exactly the best way of doing this but it was supposed to be a quick test. Regardless, it doesn't work. Or rather, it does work - the first time you run it. The second loop of the same code never completes and I can't decide if I found a bug or if I've made a mistake.

Here's the code:
Code:
dim levelMap(10) as UBYTE

randomize
for x = 1 to 10
    cls
    generateLevelMap()
    for n = 1 to 10: print levelMap(n): next n
    pause 0
next x

SUB generateLevelMap()
    dim lm as UBYTE
    dim lmPos as UBYTE = 0
    
    'clear the map first
    for lm = 1 to 10
        levelMap(n) = 0
    next lm
    
    'debug code to print the map, to make sure it's empty
    print
    for lm = 1 to 10
        print levelMap(n);",";
    next lm
    print
    pause 0
    
    'now fill it randomly
    print
    for lm = 1 to 10
        do
            'get a random position in the array
            lmPos = int(rnd * 10) + 1
            
            'debug code to print the number we randomly picked
            print lmPos;",";
            
            'debug code to print the array element if it ISN'T empty
            if levelMap(lmPos) <> 0 then
                print ink 2;levelMap(lmPos);",";
            end if
            
        loop until levelMap(lmPos) = 0
        
        'write the current number into the empty random element
        levelMap(lmPos) = lm
        
        'debug code to show that we handled this number okay
        print inverse 1;lm
        
    next lm
    
    pause 0
    cls
END SUB

I've stuck a load of debug info in there so that you can see what it's doing.

Basically it does this:
1. Clear the array to all zeroes
2. For each number, 1 to 10....
3. Look in a random array position to see if it's zero.
3a. if it isn't, loop around to 3 and try a different one
3b. if it is, write the number into that array position and move on to the next number
4. The result should be the numbers 1 to 10 in a random order in the array, after a variable number of iterations.

First run, no problem.

Second run, although the debug code shows that the array was cleared, the do...loop until loop never completes - by the time it's checking the array within that loop it contains the OLD values again, despite me having zeroed them out just a few seconds earlier.

Am I missing an obvious goof? :-D


Re: What's wrong with this code? - LCD - 03-24-2011

Code:
for lm = 1 to 10
      levelMap(n) = 0
   next lm

Maybe if you change it to:
Quote:for lm = 1 to 10
levelMap(lm) = 0
next lm
Then it will really put a zero in each position of the array.

Look also here:
Code:
for lm = 1 to 10
      print levelMap(n);",";
   next lm
But this is only a suspection...


Re: What's wrong with this code? - LTee - 03-24-2011

Ha ha, you're right! What a clown I am. :-D

That'll teach me to copy/paste stuff from one file to another! Thanks, LCD!


Re: What's wrong with this code? - boriel - 03-24-2011

Another pending feature: a command-line flag to enforce the user to *declare* (DIM) any used variable. This way, the levelMap(n) should complain about undeclared "n" var (or 2 Variable not Found) ;-) Free Basic also has this feature. I will put it, but need to re-re-check everything in the 1.2.8 version which now seems to be relatively stable at the moment...


Re: What's wrong with this code? - LTee - 03-24-2011

I'd like that! That's not the first time I've tripped up because of a bad copy/paste, and occasionally I've forgotten to declare something entirely and ended up with a float when a ubyte would've done.

And I'd agree on 1.2.8, by the way - it seems to be working great for me!