C / C++ / MFC

CodeProject 10 Million members celebration meetup - Switzerland

Marco Bertschi25-Jun-13 23:00

Marco Bertschi

25-Jun-13 23:00

I think I understand your problem;
You will definately have to log the variable value every time it changes - AFAIK there is no way of doing this 'automatically'.
The easiest way to log is probably to write the variable value into the stdout stream:

C++

int main(void){
   int myVariable = 0;
   cout<<"myVariable has the value "<<myVariable;  //Output: 'myVariable has the value 0'

   myVariable = 345;
   cout<<"myVariable has the value "<<myVariable;  //Output: 'myVariable has the value 345'

   return 0;
}

However, if you want to log the value outside of the IDE and are not developing a console app, you would need to write it to a file[^].

cheers,
Marco

Marco Bertschi

ErfanNoury25-Jun-13 23:07

ErfanNoury

25-Jun-13 23:07

Traditional "inline" logging always comes handy, but I was wondering maybe there could be a way to do it more precisely and automatically. Since there are many statements changing the value of the variable, this won't be a practical solution.
Thank you

CodeProject 10 Million members celebration meetup - Switzerland

Marco Bertschi25-Jun-13 23:19

Marco Bertschi

25-Jun-13 23:19

ErfanNoury wrote:
Since there are many statements changing the value of the variable, this won't be a practical solution.

I thought that you were thinking about something like that - The most important precondition is that you know when a variable value changes.
It is impossible to find this out when the variable of type char, int or any other native type.
The easiest way would probably be to encapsulate the variable with a setter where you can reckon that the value has changed:

int main(void){
   int myVariable;
   
   SetMyVariable(myVariable, 0);

   SetMyVariable(myVariable, 345);

   return 0;
}

void SetMyVariable(int variable, value){
   variable = value;
   cout<<"myVariable has the value "<< value;

   return;
}

Marco Bertschi

ErfanNoury25-Jun-13 23:33

ErfanNoury

25-Jun-13 23:33

Yeah it's a good idea! I think it is better to wrap the variable into a class and through operator overloading I can log the value of the variable whenever it changes.

CPallini26-Jun-13 9:12

When timeSetEvent() start?

26-Jun-13 9:12

Yes, that's fine. Probably you want to do that just in the debug build.

Veni, vidi, vici.

econy25-Jun-13 8:00

Re: When timeSetEvent() start?

25-Jun-13 8:00

According to MSDN, timeSetEvent() run in its own thread, then call timeSetEvent(), when the callback of the timeSetEvent be invoked is undetermined, not like system.threading.timer(), it has a due time to start. right?

David Crow25-Jun-13 9:56

David Crow

25-Jun-13 9:56

econy wrote:
...the callback of the timeSetEvent be invoked is undetermined...

The callback function is called once upon expiration of a single event, or periodically upon expiration of periodic events. That sounds deterministic to me.

"One man's wage rise is another man's price increase." - Harold Wilson

"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons

"Show me a community that obeys the Ten Commandments and I'll show you a less crowded prison system." - Anonymous

Re: When timeSetEvent() start?

econy25-Jun-13 10:29

Re: When timeSetEvent() start?

25-Jun-13 10:29

I mean, when is it start to count in a new thread?
OS decide which thread to start, right?

David Crow25-Jun-13 10:31

David Crow

25-Jun-13 10:31

econy wrote:
I mean, when is it start to count in a new thread?

When timeSetEvent() returns, I presume.

Convertin from c++ 6 to Visual studio C++ 10

mamoony25-Jun-13 7:41

25-Jun-13 7:41

Hi I am new in c++. Recently I am converting a C++ project to Visual Studio 10 c++ . I am getting an error.

error C2664: 'std::_Vector_iterator<_Myvec>::_Vector_iterator(const std::_Vector_iterator<_Myvec> &)' : cannot convert parameter 1 from 'int' to 'const std::_Vector_iterator<_Myvec> &'

The code is:

XNodes::iterator _tagXMLNode::GetChildIterator( LPXNode node )
{
    XNodes::iterator it = childs.begin();
    for( ; it != childs.end() ; ++(it) )
    {
        if( *it == node )
            return it;
    }
    return NULL;     //error is in here
}

How to fix it?

Thanks in advance,
Sai

CPallini25-Jun-13 8:43

25-Jun-13 8:43

Change from

Quote:
return NULL; //error is in here

C++

return it;

Veni, vidi, vici.

mamoony25-Jun-13 9:01

25-Jun-13 9:01

Hi,

Thanks.

But the logic of the code is if the node matches wuth it then return it else return nothing.

Thanks,
Saimanti

CPallini25-Jun-13 9:09

25-Jun-13 9:09

The 'return it' I suggested returns childs.end() that is the past-the-end iterator (a 'typed' NULL, roughly speaking).

Veni, vidi, vici.

mamoony25-Jun-13 9:41

25-Jun-13 9:41

Thank you. It worked. But now it is giving another error..

error C2451: conditional expression of type 'std::_Vector_iterator<_Myvec>' is illegal

The code below:

bool _tagXMLNode::RemoveChild( LPXNode node )
{
    XNodes::iterator it = GetChildIterator( node );
    if( it )  //error is here
    {
        delete *it;
        childs.erase( it );
        return true;
    }
    return false;
}

Thank in advance,
Saimanti

mamoony25-Jun-13 9:42

25-Jun-13 9:42

I did..

bool _tagXMLNode::RemoveChild( LPXNode node )
{
    XNodes::iterator it = GetChildIterator( node );
    if( it != childs.end())
    {
        delete *it;
        childs.erase( it );
        return true;
    }
    return false;
}

Is this right?

Thanks,
Saimanti

CPallini25-Jun-13 9:53

25-Jun-13 9:53

It should be correct.

Veni, vidi, vici.

mamoony25-Jun-13 9:54

25-Jun-13 9:54

Thank you very much...

Saimanti

JackDingler26-Jun-13 11:37

JackDingler

26-Jun-13 11:37

C++

bool _tagXMLNode::RemoveChild( LPXNode node )
{
    XNodes::iterator it = GetChildIterator( node );
    if( it != childs.end() )  // Fixed it
    {
        delete *it;
        childs.erase( it );
        return true;
    }
    return false;
}

timeSetEvent() question

econy25-Jun-13 6:45

Re: timeSetEvent() question

25-Jun-13 6:45

is this funtion create an event in a asynchronized way? i.e use a background thread?

if I use a timeSetEvent() in a call back function:

void CALLBACK TimerProc1( UINT uID, UINT uMsg, DWORD dwUser, DWORD dw1, DWORD dw2 ) {
 read bus data;
 if (data == true) then
  timeSetEvent( 5, 1, TimerProc3, 0, TIME_ONESHOT ) ;
}

No timeKillEvent(), then is that means the code will create many events in many threads, right? that is , no release resource.

econy25-Jun-13 6:58

Variable synchronization question

25-Jun-13 6:58

if I use same parameters in a timeSetEvent(), then call timeSetEvent() twice, is that mean I created 2 different events?

or Sencond call reset the timer?

econy25-Jun-13 5:19

Re: Variable synchronization question

25-Jun-13 5:19

I have an application, there are 2 loops, one is message loop, one is a timeSetEvent() loop.

There is a global variable in both loops.

Now I meet a problem about redraw the screen. The situation like:

wndProc (UINT message, WPARAM wParam, LPARAM lPara)
{
 if (GetTickCount() - GVarTck > 4000) then 
  redraw screen;
  
  switch (message) { handle message;}

}
///////////////////////

OnPaint() {
 if (g_timeEvent == 0)
   timeSetEvent(50, 5, RefreshScreen,0, TIME_ONESHOT)


}

void CALLBACK RefreshScreen()
{
   do Refresh thing;
   GVarTikc = GetTickCount();
   timeSetEvent(50, 5, RefreshScreen,0, TIME_ONESHOT)
   //With GetTickCount() meassured, the time of refreshScreen() is
   // about 20 ms.
}

Richard MacCutchan25-Jun-13 6:06

Richard MacCutchan

25-Jun-13 6:06

Why do you need to use a timer to decide when to redraw the screen? You should do it in response to a WM_PAINT message, and you can force a repaint by calling the InvalidateRect() function.

Use the best guess

Cycle of Message loop

econy25-Jun-13 4:59