I don't understand this array problem (potential bug)

New comment
 

Here is some code I made (a test script):

double a[10];

void OnStart()
  {
//---
   for (int i=0;i<10;i++) {
      a[i] = i;
      printf("initial state: a[%d]=%d",i,a[i]);
   }
   
   double sum;
   
   for (int i=0;i<10;i++) {
      addElements(a,i,sum);
   }
   
   for (int i=0;i<10;i++) {
      printf("final state: a[%d]=%d",i,a[i]);
   }
   
  }
  
bool addElements(double& a[], int i, double& sum) {
   int len = ArraySize(a);
   
   sum = 5*a[i];
   return (true);   
}

 Why all values are 0 in initial state and final state? 

Even if it's something I don't get about arrays in MQL5, I really don't even know where to start to understand this situation. 

Half of my code is broken because of this stuff.

Help pls? 

 

Hmm... first, never assign an integer value to a double. You can make that a[i] = i*1.0 ; would be better for the compile.

Second, you should refrain from having formal parameters (double& a[]) with the same name as public variables (double a[10]), because the compiler would be confused between the public variable and the formal parameter. It would choose what is instructed to choose and it won't give a damn on what you actually want. 

Plus that in the second for cycle, you do not reassign the values of the array with something like a[i]=sum; after you call addElements.

In the end, I don't use printf.

I put a normal Print("a[",i,"]=",a[i]); and I got something else.

 
TheEconomist:

Hmm... first, never assign an integer value to a double. You can make that a[i] = i*1.0 ; would be better for the compile.

Second, you should refrain from having formal parameters (double& a[]) with the same name as public variables (double a[10]), because the compiler would be confused between the public variable and the formal parameter. It would choose what is instructed to choose and it won't give a damn on what you actually want. 

Plus that in the second for cycle, you do not reassign the values of the array with something like a[i]=sum; after you call addElements.

In the end, I don't use printf.

I put a normal Print("a[",i,"]=",a[i]); and I got something else.

Phew... Good pointers... I'm getting back to the code to see if names of the arrays are different. 

Very nice remark about printf 

I thought about assigning 1.0*i, but with no changed results (just an usual warning at compilation).

Thank you very very much for your comment. At least now I have a track.

I will come back with results. 

 

In my code, The array names were different altogether. When I created the test script, I made the mistake of declaring array a and formal parameter array with the same name.

I was confused about printf messages. Print function works. printf doesn't work in this case.

It seems the code works actually. But I will double check everything and post here back if something related to arrays eventually pops up. 

 

But now another little thing appeared, not related to my problem, but still a bug (from my point of view). 

This is the new code (a little changed).

 

double a[10];

void OnStart()
  {
//---
   for (int i=0;i<10;i++) {
      a[i] = i;
      //printf("initial state: a[%d]=%d",i,a[i]);
      Print("initial state: a[",i,"]=",a[i]);
   }
   
   double sum;
   
   for (int i=0;i<10;i++) {
      addElements(a,i,a[i]); //THIS DOESN'T WORK
      //a[i] = sum; // THIS WORKS
   }
   
   for (int i=0;i<10;i++) {
      //printf("final state: a[%d]=%d",i,a[i]);
      Print("final state: a[",i,"]=",a[i]);
   }
   
  }
  
bool addElements(double& inputArray[], int i, double& inputSum) {
   int len = ArraySize(inputArray);
   
   inputSum = 5*inputArray[i];
   return (true);   
}

 why addElements(a,i,a[i]) doesn't work? Weird values appear there. I guess it's about pointers.

Am I missing something or it's a bug ?

Documentation on MQL5: Common Functions / Print
  • www.mql5.com
Common Functions / Print - Documentation on MQL5
 

I get this, running this script:

2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[9]=9
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[8]=8
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[7]=7
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[6]=6
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[5]=5
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[4]=4
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[3]=3
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[2]=2.000000000000028
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[1]=4.056481920730334e+031
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[0]=0

 Yes, we have a bug confirm. Works perfectly with the other line:

2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[9]=45
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[8]=40
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[7]=35
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[6]=30
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[5]=25
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[4]=20
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[3]=15
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[2]=10
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[1]=5
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[0]=0

 

Your code has one a little but fatal mistype. Try this 

double a[10];

void OnStart()
  {
//---
   for (int i=0;i<10;i++) {
      a[i] = i;
      printf("initial state: a[%d]=%G",i,a[i]);
   }
   
   double sum;
   
   for (int i=0;i<10;i++) {
      addElements(a,i,sum);
   }
   
   for (int i=0;i<10;i++) {
      printf("final state: a[%d]=%G",i,a[i]);
   }
   
  }
  
bool addElements(double& a[], int i, double& sum) {
   int len = ArraySize(a);
   
   sum = 5*a[i];
   return (true);   
}

 
TheEconomist:

I get this, running this script:

2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[9]=9
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[8]=8
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[7]=7
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[6]=6
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[5]=5
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[4]=4
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[3]=3
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[2]=2.000000000000028
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[1]=4.056481920730334e+031
2010.09.27 17:36:21 ifmihai (USDCAD,D1) final state: a[0]=0

 Yes, we have a bug confirm. Works perfectly with the other line:

2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[9]=45
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[8]=40
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[7]=35
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[6]=30
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[5]=25
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[4]=20
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[3]=15
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[2]=10
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[1]=5
2010.09.27 17:38:58 ifmihai (USDCAD,D1) final state: a[0]=0

Same here. a[2] and a[1] have exactly the same values in my case too. 
 
Rosh:

Your code has one a little but fatal mistype. Try this


Very good remark. Really helped. 
 

Yes, but the mistype was in printf. This doesn't cancel the fact there is something fishy about passing an array item to a procedure expecting a double as reference. 

New comment