Thread: functions using arrays

The ascend function returns 1 if it does ascend and -1 if it does not. the same is for the descend.

Not exactly - the function returns an undefined value for a size of 1, and a 1 or -1 if the LAST element of the array is "ascending" or "descending" in relation to the first element. But it says nothing at all meaningfull about the elements in between. So if you have 10 12 13 14 15 17 19 9, then it's not ascending according to your function, whilst 10 1 2 3 4 5 6 12 is - I don't think either is ascending properly - but maybe I don't understand what ascending and descending arrays are correctly - I never studied math at a higher level.

Code:

 if (seconval < 0){ 
    firsval = 1.0;  
    seconval= 1/firsval;  
  }

This is still quite wrong. I think you can see that if you actually read those lines.

Other than those few things, I think you're pretty close. Have you actually tested your code with some real data, and checked the results?

--
Mats

That looks much better.

Multiplying by -1 may not work quite as well as "x = -x" when it comes to speed in the compiler - not that it really makes any difference here, but if you can do things SIMPLE, why make it complicated?

And by the way, "count will never be > 1" is not true. If "firsval" is 0.25 for example, count would be 4. But the result would of course not at all be the right answer.

--
Mats

I still don't know what ascend actually does, but first of all, I'm a bit dubious that "next" should be an integer, when the array contain floats.

Is an array that contains 1.0, 1.1, 1.2, 1.3, 1.4 ascending or not? If it is, then your code doesn't work.

--
Mats

Well, maybe you should just compare the "current" with the "next" element in the array. If at any point it the elements don't follow the rule, you break the loop and return "not ascending". If you finish the loop without finding any "offending" items, then you return "is ascending".

When I say "next", I don't mean the "next" variable you have, but rather array[x] and array[x+1].

There should be no need for a variable like your "next".

--
Mats

Yes, but you still need to "stop" looping when you find that it's not ascending, otherwise you'll have "false positives" if the last two numbers are ascending, even if all the others where descending, for example.

--
Mats

also what does this mean? i don't know how to fix it?
errors:
cc1: warnings being treated as errors
calculator.c: In function 'fill_arr':
calculator.c:123: warning: format '%f' expects type 'float *', but
argument 2 has type 'float **'


code:

Code:

void fill_arr(float farry[], int *sizearr){
     int fills = 0;
     while (fills < *size && scanf(“%f”, &farry) !=EOF)
          fills ++;
      } 
      *sizearr = fills;
  }

You could do:

Code:

float ascend_arr(float aarry[], int sizearr){ 
  int ascend; 
  for(ascend = 1; ascend < sizearr;){   
      if (aarry[ascend] < aarry[ascend + 1]) 
         ascend++; 
      else{
         return -1;
      }
   } 
  return 1; 
}

By the way, why is the function returning a float - for C++ I would use bool, for C an int would be the correct return type.

Code:

void fill_arr(float farry[], int *sizearr){
     int fills = 0;
     while (fills < *size && scanf(“%f”, &farry) !=EOF)
          fills ++;
      } 
      *sizearr = fills;
  }

When you pass an array to another function, you are actually passing a pointer to the first element. So farry[] is the same as *farry. If you take the address of a *farry, it becomes a pointer to a pointer (**farry), right?

There are two problems here:
1. Your code is filling in to the address of the pointer to the array - which of course is not the same as the array.
2. You are not actually filling in your ARRAY itself.

You want to pass &farry[fills] to scanf.

Good to see that you use "-Wall" and "-Werror" - they are both good for stopping simple errors that can take hours of debugging to find.

--
Mats

--
Mats

Originally Posted by trprince

thank you... so now my fill array function works. I put in what you had for the ascend array and it works but not in that it returns a -1 for the ascending correctly not a 1

Ah, classic fence-post error (twice actually). [1] You are starting at element 1 and comparing with element 2. And so on, untill you come to sizearray-1, which you compare to sizearray - which is one BEYOND what you have filled in. You should start your loop at 0 and check until sizearray-1. So your "end condition" should be sizearray-1.

Also, put the "ascend++" in the for-loop and reverse the if-statement - that way you can do away with the else altogether, and it makes the whole thing a lot simpler.

[1] The term "fence-post error" comes from "how many posts do you need to X <unit of length> of fence if you put a post every Y <unit of length>", where the answer depends on whether you start with a post at the end, or in the middle of Y units, and also wether you need one at the end or not.

--
Mats