|
The core code will be the same, the samples are console apps to keep them simple.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
How to design for the receive dialog when I send a message to it? My send dialog can send a message but the receive side is a Dos application, how do I design a dialog that I can receive and see the message I send?
|
|
|
|
|
1983ttj wrote:
but the receive side is a Dos application, how do I design a dialog that I can receive and see the message I send?
I repeat - the code in your DOS app will work in an MFC app.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
((int)((int)(r+100000.5f)-100000))
((int)((int)(r+0.5f))
Is the former one better? Why?
|
|
|
|
|
Rnd, Floor or Ceil from math.h is the way to go.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
There is no rnd in math.h.
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
|
(int) (r+0.5f) is the way to go if you are rounding to the nearest integer with ?.5 always rounding up.
The second version avoids overflow problems. My question would be why would someone think adding 100000 would improve the rounding function. When dealing with floating point numbers, you really have to be careful when adding and subtracting values.
(NOTE: this only works for positive numbers, negative numbers must use -0.5f and not 0.5f)
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
-- modified at 10:26 Friday 9th September, 2005
|
|
|
|
|
Oh dear god, I just realized what they are doing.
(int) (f) does a truncation of the floating point number. In otherwords, it rounds to zero.
2.5 -> 2
0.3 -> 0
-4.5 -> -4
The "(int) (r+0.5f)" trick only works for positive values.
2.3 -> 2
2.7 -> 3
-0.2 -> 0
-0.7 -> 0 (WRONG - should be -1)
By doing 100000.5f, they AVOID (not remove) the negative problem.
Example, but we use 10.5f instead of 100000.5f
-0.2 + 10.5f = 10.3f -> 10 -> 0
-0.7 + 10.5f = 9.8f -> 9 -> -1
But dear god, WHAT A HACK.
The proper way of doing it is:
int i = (int) (r + (r > 0.0f ? 0.5f : -0.5f));
This is why Christian wants people to use math.h because it is easy to create things that sort of work, but often they don't always work.
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
|
please, somebody could publish a simple example on SNMP in C++?
Thanks!
Sorry for my english
|
|
|
|
|
Search CP articles for SNMP. There are several examples.
"One must learn from the bite of the fire to leave it alone." - Native American Proverb
|
|
|
|
|
sebastianos wrote:
somebody could publish a simple example on SNMP in C++?
here is list :-
SNMP ARTICLE[^]
"Opinions are neither right nor wrong. I cannot change your opinion. I can, however, change what influences your opinion." - David Crow
cheers,
Alok Gupta
VC Forum Q&A :- I/ IV
|
|
|
|
|
Friends,
I want to develop a ActiveX control container,and save data in flat(uncompound) document.
My document CxxDoc is derived from COleDocument,and the constructor as following:
CxxDoc::CxxDoc()
{
// EnableCompoundFile();
// ^^^^^^^^^^^^^^^^^^^^^^^^^^I need a flat (uncompound) file.so i delete it.
...
}
Firstly, Is it OK????Can i save data in flat document?
Secondly,my client item is CxxClientItem,
class CxxClientItem : public COleClientItem
The WriteItem and ReadItem function (copied from the sample program of ActiceX control test container in Visual C++) as follows. They can work correctly in compound document. Please tell me how to change them from compound document to flat(uncompound) document. I'm a new and I really need your help.
void CxxClientItem::WriteItem( CArchive& ar )
{
USES_CONVERSION;
IStoragePtr pStorage;
IStreamPtr pStream;
IPersistStreamInitPtr pPSI;
TCHAR szItemName[64];
HRESULT hResult;
IPersistStoragePtr pPersistStorage;
CxxDoc* pDoc;
CLSID clsid;
(void)ar;
pDoc = GetDocument();
ASSERT_VALID( pDoc );
ASSERT( pDoc->m_lpRootStg != NULL );
// ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^It is wrong in uncompound document
ASSERT( pDoc->m_bCompoundFile );
// ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^It is wrong in uncompound document
ASSERT( m_lpStorage != NULL );
ASSERT( !ar.m_bForceFlat );
GetItemName( szItemName );
if( !pDoc->m_bSameAsLoad )
{
hResult = pDoc->m_lpRootStg->CreateStorage( T2COLE( szItemName ),STGM_CREATE|STGM_READWRITE|STGM_TRANSACTED|STGM_SHARE_EXCLUSIVE, 0, 0, &pStorage );
// ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^It is wrong in uncompound document
if( FAILED( hResult ) )
{
AfxThrowOleException( hResult );
}
m_lpNewStorage = pStorage;
m_lpNewStorage->AddRef();
m_bNeedCommit = TRUE;
}
else
{
pStorage = m_lpStorage;
}
......
}
void CxxClientItem::ReadItem( CArchive& ar )
{
USES_CONVERSION;
BOOL tUsedStream;
IStoragePtr pStorage;
IPersistStreamInitPtr pPSI;
IStreamPtr pStream;
TCHAR szItemName[64];
HRESULT hResult;
BOOL tQuickActivated;
IPersistStoragePtr pPersistStorage;
CLSID clsid;
(void)ar;
CxxDoc* pDoc = GetDocument();
ASSERT_VALID(pDoc);
ASSERT(pDoc->m_lpRootStg != NULL);
// ^^^^^^^^^^^^^^^^^^^^^^^^^^It is wrong in uncompound document
ASSERT(pDoc->m_bCompoundFile);
// ^^^^^^^^^^^^^^^^^^^^^^^^^^It is wrong in uncompound document
ASSERT(m_lpStorage == NULL);
ASSERT(m_lpLockBytes == NULL);
ASSERT( !ar.m_bForceFlat );
// get item name
GetItemName( szItemName );
// open storage for this item
hResult = pDoc->m_lpRootStg->OpenStorage( T2COLE( szItemName ), NULL,
STGM_READWRITE|STGM_TRANSACTED|STGM_SHARE_EXCLUSIVE, 0, 0, &pStorage );
// ^^^^^^^^^^^^^^^^^^^^^^^^^^It is wrong in uncompound document
if( hResult != S_OK )
{
TRACE1( "Warning: unable to open child storage %s.\n", szItemName );
// upon failure throw file exception (item will be cleaned up)
AfxThrowOleException( hResult );
}
ASSERT(pStorage != NULL);
// remember the storage
m_lpStorage = pStorage;
ASSERT(m_lpStorage != NULL);
m_lpStorage->AddRef();
......
}
|
|
|
|
|
Hi folks,
how can I get the ascii code (0-255) from the wParam of the WM_KEYDOWN message (virtual-key code) ?
Thanks in advance,
Achim Klein
We can do no great things, only small things with great love. - Mother Theresa
|
|
|
|
|
According to MSN[^], that's all that's in there. So I expect you just need to cast the value to throw off the extra bits you don't need.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
Hi Christian,
thanks for your instant feedback.
I've already tried to copy the bits according to this article, but unfortunately it doesn't work.
For example:
- the the Num-Pad keys returns lowercase characters
- the '+' and '-' keys haven't the normal ascii code
- all alphabetical keys from the keyboard are uppercase characters
It's a bit confusing...
We can do no great things, only small things with great love. - Mother Theresa
|
|
|
|
|
Have you tried WM_CHAR ?
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
It's a dialog based application.
I've set a breakpoint in the WM_CHAR handler but it seems that the handler never gets called.
We can do no great things, only small things with great love. - Mother Theresa
|
|
|
|
|
I created a VC6 dialog project and put this in PreTranslateMessage:
if (pMsg->message == WM_KEYDOWN)
{
CString s;
s.Format("%d %c", pMsg->wParam, pMsg->wParam);
AfxMessageBox(s);
}
I got the same result as you. I replaced WM_KEYDOWN with WM_CHAR and got upper and lowercase and correct values for the keypad.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
Yeah it works !
Thanks a bomb !
It's 3:23 a.m here in germany, and now that it works I can finally go to bed.
Thanks a lot !
We can do no great things, only small things with great love. - Mother Theresa
|
|
|
|
|
No worries - glad to help 
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
|
I check the WM_KEYDOWN message in the PreTranslateMessage() handler...
We can do no great things, only small things with great love. - Mother Theresa
|
|
|
|
|
|
WM_KEYDOWN does NOT return ASCII characters. It returns virtual-key codes for the keys which happen to be ASCII characters for some of the keys.
The TranslateMessage function looks for WM_KEYDOWN messages and if it can convert the data in that message to a character-code (ASCII codefor ASCII keyboards???) then it generates a WM_CHAR message.
http://blog.ngedit.com/2005/06/13/whats-broken-in-the-wm_keydownwm_char-input-model/[^]
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
