|
structures reserve memory for all the variables inside it. but Union takes up memory for only the thing that we assign value. and overwrites the same if a value is entered to antoher variable inside it.
VuNic
|
|
|
|
|
VuNic wrote: Union takes up memory for only the thing that we assign value
takes up memory for variable who has maximum amount of byte.
Divyang Mithaiwala
System Engineer & Software Developer
|
|
|
|
|
|
Your link contains content that verifies what the person-you-are-responding-to said...
The compiler allocates a piece of storage that is large enough to hold the largest variable type in the union What is the confusion? 
Peace!
-=- James
If you think it costs a lot to do it right, just wait until you find out how much it costs to do it wrong!
Avoid driving a vehicle taller than you and remember that Professional Driver on Closed Course does not mean your Dumb Ass on a Public Road!
DeleteFXPFiles & CheckFavorites (Please rate this post!)
|
|
|
|
|
hi VuNic,
VuNic wrote: Union takes up memory for only the thing that we assign value
i'm sorry to say that you are wrong.
a union occupies as much memory as the wider field of the union.
lets say we have this union :
union myU {
long l;
char c;
};
then, even if you assign c which is a char, so 1 byte, the union will weight 4 bits because the longuest field is l (long is 4 bytes).
myU u;
u.c = 5;
ASSERT(sizeof(u) == 4);
|
|
|
|
|
|
i'm sorry, but the author of that article is really not aware on what unions are !!! 
all the fields of a union starts at the same address, so, they overlap, and the cosequence of this is that the union length (whatever data it's benn assigned) is the length of the larger field...
|
|
|
|
|
The SDK(telephony) which i'm using is based on 100's of Unions. in that its said that Not all of the items inside the union would be used at a time. moreover only one would be used at a time. so it saves memory when you use Unions. hmm.. i'll check it out again.
VuNic
|
|
|
|
|
The compiler has to allocate enough space for the largest item in the union. If you create a union that contains a char (1 byte), a 32-bit int (4 bytes), and a double (8 bytes), the size of the union will be 8 bytes - always (disregarding alignment and packing). If you create a union containing structures, the size of the union will be that of the largest-sized structure.
The reason the compiler has to use the largest possible size is that the union cannot change its storage size at runtime (just like a structure).
Also there is nothing inside of the union that indicates which data item inside of it is in use. Since this information is unavailable, the space allocated for the union has to use the largest possible size, just in case the largest member of the union happens to be in use, and the the union is passed around to other functions by value (for example). And, even if that information WAS available, there would be nothing stopping the called function from setting a different larger member of the structure, so all of its storage space would have to be there, anyway.
The documentation you read is correct - not all of the members of the union will be being used at the same time - but it says nothing about the storage required for them.
Peace!
-=- James
If you think it costs a lot to do it right, just wait until you find out how much it costs to do it wrong!
Avoid driving a vehicle taller than you and remember that Professional Driver on Closed Course does not mean your Dumb Ass on a Public Road!
DeleteFXPFiles & CheckFavorites (Please rate this post!)
|
|
|
|
|
union EU
{
int eu1;
int eu2;
}eu;
int main(int argc, char* argv[])
{
eu.eu1=10;
eu.eu2=13;
printf("%d",eu.eu1);
}
VuNic
|
|
|
|
|
easy 
13
|
|
|
|
|
Hello,
I want to apply the transcoding of the images using C++.
Background :
As the bigger images with the high resolution are taking a long time to load I want to convert the images into low resolution format and to a restricted size. I am working with c++ and using the GDI+ functions to load the images.
Problem : To apply transcoding of images before actual loading.
Thanks in advance ...
Regards,
Mukta.
|
|
|
|
|
Hi,
where you want to load the image? In memory or any control?
Bye,
Cool Ju
Dream Ur Destiny
|
|
|
|
|
Hi,
Thanks for the quick reply.
I want to transcode the image and keep it in the memory first.
As and when the user clicks the thumbnail of that image I want to load the transcoded image in the control.
Regards,
Mukta.
|
|
|
|
|
Hi,
AFAIK, without loading the image you cannot do any operation on it. Atleast you have to load it in memory,to do some operations(in your case transcoding).
Transcode the image by keeping it in memory and then display or load it in the control. This will reduce the time delay.
Bye,
Cool Ju
Dream Ur Destiny
|
|
|
|
|
we are using dynamic link library for printing purpose.
how to change orientation of default printer through program?
|
|
|
|
|
LPDEVMODE lpdevmode;
...
...
...
lpdevmode->dmOrientation=
|
|
|
|
|
thanx
we have tried this option already, the change is not reflecting in the actual device.
|
|
|
|
|
|
Hi All,
I have a dialog based app, containing two classes:
1)CMyDlg
2)CMyClass
Now, i can grab all the messages of CMyDlg in PreTranslateMessage(), but i dont want that. I would like to grab all the CMyDlg messages in CMyClass.
Anybody having any ideas? Check the link below to findout more abt it.
http://www.codeguru.com/forum/showthread.php?p=1346397[^]
Best regards,
JT!
|
|
|
|
|
You can handle CMyDlg's messages in CMyClass by subclassing the CMyDlg instance after it's been created.
Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact"
|
|
|
|
|
Thank you Ryan for you help.
Ryan Binns wrote: >subclassing the CMyDlg instance after it's been created.
subclassing the CMyDlg instance after it's been created. i have taken your post as follows:
1) The CMyDlg's instance is created in CMyApp, and there the dialog is DoModal()ed. So there in CMyApp im supposed to subclass it?
2)How to subclass an instance? can you be more specific with reference to my code at the link i posted in my previous post?
Best regards,
JT!
|
|
|
|
|
jusstujoo wrote: 1) The CMyDlg's instance is created in CMyApp, and there the dialog is DoModal()ed. So there in CMyApp im supposed to subclass it?
You'll have to do it after it's been created. In your WM_CREATE handler in your CMyDlg class would be a good place.
jusstujoo wrote: 2)How to subclass an instance? can you be more specific with reference to my code at the link i posted in my previous post?
You call SubclassWindow() on your dialog window. Have a look at the MSDN documentation to see exactly what you have to do. It's not too difficult.
Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact"
|
|
|
|
|
Hello Ryan
Do you really think that i should do this SubclassWindow()?
Isint there means by which i can get msgs in CMyClass without writing anything in CMyDlg??
Well... let me try. Isint there a way that i can do CMyDlg-subclassing in CMyClass?
Thank you.
Best regards,
JT!
-- modified at 4:32 Wednesday 8th March, 2006
|
|
|
|
|
jusstujoo wrote: Isint there means by which i can get msgs in CMyClass without writing anything in CMyDlg??
Not easily. It is far, far harder without putting anything in CMyDlg. Yes, it's possible, but complex, difficult and, in my opinion, not worth the effort.
Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact"
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
