|
Change the order of A and B in the C class and see the result:
class C
{
B b;
A a;
public:
C() : a()
{
wcout << L"D\n";
}
};
"We make a living by what we get, we make a life by what we give." --Winston Churchill
|
|
|
|
|
Thanks George,
Will the initialization sequence impact the way we program, I mean, because of the sequence of initialization, what points do we need to consider in your experiences?
regards,
George
|
|
|
|
|
Yes, it does impact the way you write a class.
In addition to that, the rules for guaranteed object initialization to take place (By the way, I consider POD types objects too!) are complicated. Thus, I always initialize my objects before I use them. In a class, this is done in the initialization list in the order the member variables appear in the class. Using the initialization list is the most efficient technique. Avoid assignment whenever possible! Please note, however, that, with built-in types (POD types), there is no extra cost in using assignment over initialization but having consistant code is always a plus.
Our example code demonstrated the classes A and B were initialize with or without an initialization list in the order of appearance in class C. Placing theses classes in class C's initialization list might seem like overkill to you. However, by placing them in the initialization list, you erase any doubt about whether you classes have been properly initialized.
Finally, order of initialization due to placement of member variables can get a little tricky if a member variable requires other member variables to be initialized before its initialization. Thus, you have to order your member variables accordingly. If you cannot order your member variables due to a really complicated initialization dependency, you may have to uses other methods such as assignment in the constructor body.
I hope I was clear,
George
References:
[Meyers2005] Scott Meyers,Effective C++: 55 Specific Ways to Improve your Programs and Designs, Third edition, Addison-Wesley.
"We make a living by what we get, we make a life by what we give." --Winston Churchill
|
|
|
|
|
Thanks George!
Cool! My question is answered.
regards,
George
|
|
|
|
|
please, George, don't give too much to George_George. he's been abusing this board for several months, and he doesn't even search the web before asking very simple question.
so, please, don't give him too easy answers...
|
|
|
|
|
Toxcct,
I understand your pain, and have been following George_George exploits. He bites the hand that feeds him! However, my goal is to help and I won't let posters such as George_George change my focus just because he is a SOAB.
Geo
"We make a living by what we get, we make a life by what we give." --Winston Churchill
|
|
|
|
|
George L. Jackson wrote: just because he is a SOAB
lol, i feel lighter now that someone else have the same feeling than me.
BTW, I do have the same goal too, and I generally answer his technical questions, but there a time when he already has enough to find his answer. it's not our job to pound out a full working code. we are here to advice and to help when others are in real need. posting just to post is not my religion.
modified on Tuesday, December 18, 2007 10:33:26 AM
|
|
|
|
|
Sometimes, I get over helpful. However, I just modified his code a bit and asked him to play with the order of the member variables to lead him to his own Aha! moment. Of course, his post proved:
He was what I thought he was, and I let him off the hook.
Geo
"We make a living by what we get, we make a life by what we give." --Winston Churchill
|
|
|
|
|
I agree a bit with you, but disagree too. Maybe is not "fair" to give him the easy answer, but for other users that maybe don't have had the "problem" or "question" before (i.e. me) an easy answer is full worthy to learn. Actually I have to say that altough GG is sometimes annoying, some of his questions had been interesting and the answers ave been very usefull for me to get some points that I hadn't even thought about.
Greetings.
--------
M.D.V. 
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
“The First Rule of Program Optimization: Don't do it. The Second Rule of Program Optimization (for experts only!): Don't do it yet.” - Michael A. Jackson
|
|
|
|
|
The compiler will give the class initialize implicitly.In the book "Inside the C++ Object Model", it detailed specifically.
Later buggers harm more.
|
|
|
|
|
Hi
I need to make the header font bold in my ListControl. How do I do that?? Whihc message should I handle?? I dont see any variable in the NMLVCUSTOMDRAW structure for doing this. Please help
With Reagards
Vikas
|
|
|
|
|
Hello,
I have declared a namespace for 3 of my classes and changed the declarations accordingly using : : but I don’t know how to change the declarations of message maps.
In the following I get error saying
Dialog::OnCreate' : member from enclosing class is not a type name, static, or enumerator
error C2065: 'OnCreate' : undeclared identifier
error C2440: 'type cast' : cannot convert from 'int *' to 'int (__thiscall CWnd::*)(struct tagCREATESTRUCTA *)'
BEGIN_MESSAGE_MAP(f::Dialog, CDialog)
//{{AFX_MSG_MAP(CDynDialogEx)
ON_WM_CREATE()
//}}AFX_MSG_MAP
ON_MESSAGE(WM_HELP, OnHelpMsg)
ON_WM_MEASUREITEM()
ON_WM_DRAWITEM()
In the above OnCreate() is the message handler function
Prithaa
|
|
|
|
|
I don't understand well you question. Do you put namespace around your CDialog inherited class?
Can you post the relevant code?
If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.
|
|
|
|
|
prithaa wrote: I have declared a namespace for 3 of my classes and changed the declarations accordingly using : :
Why?
Adding
using namespace f; did not help?
Let's think the unthinkable, let's do the undoable, let's prepare to grapple with the ineffable itself, and see if we may not eff it after all.
Douglas Adams, "Dirk Gently's Holistic Detective Agency"
|
|
|
|
|
Make sure you understand what the various message map macros expand to.
The message map is a static member variable of the class. If you wrap
the dialog class declaration in a namespace then wrap the dialog
implementation in that namespace.
Move the message map table definitions into the same namespace the
dialog class is declared in.
namespace f {
class CAboutDlg : public CDialog
{
protected:
afx_msg void OnButton1();
DECLARE_MESSAGE_MAP()
};
}
namespace f {
CAboutDlg::CAboutDlg() : CDialog(CAboutDlg::IDD)
{
}
BEGIN_MESSAGE_MAP(CAboutDlg, CDialog)
ON_BN_CLICKED(IDC_BUTTON1, OnButton1)
END_MESSAGE_MAP()
void CAboutDlg::OnButton1()
{
}
}
|
|
|
|
|
Hi,
I need to create a unique id which consists only of numbers (0-9), and needs to be of a certain length (not more than 37 chars).
How can I do this?
I know that here is GUID, but this creates a hexidecimal number.
Thanks
|
|
|
|
|
Well, I think it is far easier to change the GUID representation than writing your own algo to create a unique ID. Unfortunately, a 128-bit wide number (as GUID is) requires more than 38 decimal digits.
If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.
|
|
|
|
|
yes - that's exactly my problem
|
|
|
|
|
yes - that's exactly my problem
If I use part of the GUID that is generated and convert that to decimal. will that still be globaly unique?
|
|
|
|
|
SWDevil wrote: If I use part of the GUID that is generated and convert that to decimal. will that still be globaly unique?
No.
BTW why have you such a requirement constraint (just curious)?
If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.
|
|
|
|
|
The requirement is a 64 char string.
But there is a first part that I have to include, and for the second part I have to generate a unique ID.
Both parts together should not be mopre than 64 chars.
|
|
|
|
|
BTW, do you really need a Global Unique Identifier?
GUID exist because ID creation must be shared (anyone may create his unique IDs via Guidgen) and not bound to a centralized mechanism
(as the one assigning car license numbers), do you have such a strict requirement? 
If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.
|
|
|
|
|
Could you get away with using a timestamp (probably including milliseconds?)
Thats nowhere _globally_ unique, but unique to every point in space. And maybe that is enough?
Let's think the unthinkable, let's do the undoable, let's prepare to grapple with the ineffable itself, and see if we may not eff it after all.
Douglas Adams, "Dirk Gently's Holistic Detective Agency"
|
|
|
|
|
I am thinking of using GetTickCount, but I understand that this might return a negative value? Is this true?
|
|
|
|
|
SWDevil wrote: ...GetTickCount...might return a negative value?
No. It returns a DWORD, that is unsigned.
It wraps around after 49,7 days, though.
Win95 used to crash after that uptime. Not that it was generally able to stay up that long without crash anyway.
time64 returns the seconds since UTC. It wraps around somtimes in the year 3000
If you really need, you could mix in the milliseconds since startup (GetTickCount)
time64 relies on the computers clock being set - tampering with the clock *could* make your UID non-unique. That may or may not be a problem.
Let's think the unthinkable, let's do the undoable, let's prepare to grapple with the ineffable itself, and see if we may not eff it after all.
Douglas Adams, "Dirk Gently's Holistic Detective Agency"
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
