|
The InitInstance() of your DLLs application is a task of the Loader,
that is a part of your Windows OS... 
Usually -
the people use exported classes and functions of a DLL only
That is why
the wizard does not make the class of the DLLs application exported Check your definition of Irrationality[ ^]
1 - Avicenna
5 - Hubbard
3 - Own definition
|
|
|
|
|
I have no idea what "OpenHR" is, but I'll try to help based on what you've posted.
Vaclav_Sal wrote: Linking...
OpenHR.obj : error LNK2001: unresolved external symbol "public: __thiscall COpenHR_DLL_TestApp::COpenHR_DLL_TestApp(void)" (??0COpenHR_DLL_TestApp@@QAE@XZ)
Debug/OpenHR.exe : fatal error LNK1120: 1 unresolved externals
Error executing link.exe.
The first thing to note about a linker error is that everything compiled Ok: all the declarations of the required classes and methods are present. The linker error indicated that it can't find "COpenHR_DLL_TestApp::COpenHR_DLL_TestApp(void)" in "OpenHR.obj". This suggests that the file OpenHR.cpp attempts to construct a COpenHR_DLL_TestApp but the definition for the constructor isn't present (only its declaration). The first step is to have a look and see if it exists.Steve
|
|
|
|
|
the *cpp in which the class COpenHR_DLL_TestApp is implemented got to be included in the project. Or the lib.
Somehow is the ctor declared but not implemented.Press F1 for help or google it.
Greetings from Germany
|
|
|
|
|
Many thanks to all who replied, I really appreciate it.
But ... I am still confused.
I think the terminology used is part of the confusion.
Let me attempt to clarify what I have done so far.
So I have a application - MFC DLL with *.cpp and *.h files.
I have done this
1. Created "child" direcotory to mine main program - COpenHR_DLL_Test
2. Used wizard to build the MFC DLL "application"
3. Build the COpenHR_DLL_Test DLL.
4. Added this COpenHR_DLL_TestApp to my workspace.
5. Made COpenHR_DLL_TestApp dependent on main application ( OpenHR)
6. Put #include "OpenHR_DLL_Test\OpenHR_DLL_Test.h" into stdafx.h
6.Added the *.lib (in Project - Settings -Link ) OpenHR_DLL_Test/Debug/OpenHR_DLL_Test.lib
7.copied OpenHR_DLL_Test.dll into windows/system32 – I understand it is not necessary
8.Sucessfuly compiled and linked both project in workspace.
9.Added in
BOOL COpenHRApp::InitInstance()
{
COpenHR_DLL_TestApp m_COpenHR_DLL_TestApp;
10.Now it will compile with described linker error.
What you have told me I am missing the implementation of the constructor.
I just do not see where.
I have the header file of the DLL included and the libraly linked.
What else is there??
Here is quote from the *.h as generated by wizard.
// COpenHR_DLL_TestApp
// See OpenHR_DLL_Test.cpp for the implementation of this class
//
In the meantime I'll try using the MFC DLL extension option in the MFC DLL wizard.
Vaclav
|
|
|
|
|
Beeing a geenhorn in MFC DLL I missed AFX_EXT_CLASS !!
Or did the wizard missed it?
I blame it on wizard! It should be smarter - it is building MFC extension DLL in first place!
Or at least make some smart remarks about it!
Anyway, thanks for all you help.
class AFX_EXT_CLASS OpenHR_DLL_Test : public CWnd
|
|
|
|
|
Hai,
Can Anyone give me a link to connect to MySqlServer using CDataBase Class. I searched a lot but i am not able to find out a proper link.Can anyone help me on this??
Thanks,
|
|
|
|
|
|
|
Hi all,
I have developed one win32 dll in VC++ for windows mobile. I need to do some work after each 1 hour, So I want to use Timer in this dll.
Can any one please let me know that how can i use Timer in dll?
Regards,
Rahul Vaishnav
|
|
|
|
|
|
See my answer to (your ?) same post here[^].
ARWhen the wise (person) points at the moon the fool looks at the finger (Chinese proverb)
|
|
|
|
|
I have a parent process that creates a child process that in turn exchanges data with an external network addressed computer. I would like the parent to be able to monitor the amount of data the child is sending and receiving. The application runs on W2K through Win7 and is written in C/C++. I already have the handles and PIDs of both the parent and child and know the internal IP address, external IP address and the IP address and port on the external network computer. I found some sniffer articles on here that may apply but they require a driver installed which I can not allow. Any suggestions or help is appreciated.
|
|
|
|
|
|
I had a thought ... write a high speed tcp/ip proxy server to sit 'inbetween' your end-points - it simply needs to add up bytes/packets in/out and present that into in a shared-memory area or such
but thats an extra component ! in reality its doing the same as WinPCAP in stats mode, just from a higher level
'g'
|
|
|
|
|
I was hoping someone knew of an API call that I wasn't able to find perhaps something like:
GetProcessIoCounters or GetIpStatistics or GetTcpTable but I'm not sure if any of these are appropriate.
|
|
|
|
|
well, I have seen a C# way of doing it but cant remember all the steps to capture packets .. my first thought was also that it could cause performance issues on the local machine at least ...
I think they opened a socket for receive, bound something like IP-Address-Any to it and then received all the packets from the network adaptor, mapping the packet receive to an event that they could monitor/wait for ... when the event was tripped that had a packet obkect they decoded the basic fields from - I just cant remember who did it and what the pitfalls were
Actually David Crow has an article here about GetIPStatistics, the best thing to do is 'suck it and see' - if as a standalone it gives you what you're looking for you could look at including his functionality in your program (and crediting him of course) - see More on using IP Helper API’s[^]
'g'
|
|
|
|
|
That article is really nice. I wish I found that in my search when I was trying to do several of those things. Unfortunately the links in the reference section that I wanted to follow were all bad.
|
|
|
|
|
| Is the child process yours? If so forget all this fancy stuff and just have the child process report its progress to the parent (or otherwise publish them where the parent can access). Steve
|
|
|
|
|
No, only the parent process is my code. But I do have access to the child via the PID and its handle as well as knowing all the IP addresses involved in the transfers. I don't really care if the answer is totally accurate, but just a general knowledge that the link is alive and continually sending data back and forth.
|
|
|
|
|
Ok. Can you tell us any more about the "child" application? For example, it may be able to split out progress reports to stdout in which case pipes and output redirection could be used. This is just an example, but you get the idea: it may already provide a notification mechanism you can use.Steve
|
|
|
|
|
I have to threat the child process as a black box producer / consumer as one end of a network device. I have full control over launching the process but it must run as a normal independant app.
|
|
|
|
|
Hi
I'm trying to re-enable a disabled button by right clicking on the button. I know that the message is not sent to the parent window for disabled buttons. Is there a workaround solution on receiving messages for disabled controls?
Thanks
|
|
|
|
|
Edit: This is WRONG. I was obviously not thinking. Please ignore it.
I have not tried this. Why don't I tell you my idea and you can tell me if it is right?
( I am presuming that you are using MFC. For win32, you would use window subclassing. )
Derive a class from CButton.
Override OnRButtonDown(). Have it enable the (button) window.
Add a data member of this class to the class of the parent window.
In the parent window, when the button is created, associate it with this data member. If the parent window is a dialog, you can use the DDX mechanism for this - see DoDataExchange and DDX_Control. If you are manually creating the button, do it with this variable's Create() member function.
That's it.Please do not read this signature.
modified on Monday, March 8, 2010 9:57 AM
|
|
|
|
|
Avi Berger wrote: Derive a class from CButton.
Override OnRButtonDown(). Have it enable the (button) window.
I think you're missing the point. If the window (button) is disabled, it won't receive the RButtonDown message.
No, I think the trouble here is the OP is trying to make the button do something the UI design doesn't support; e.g., trying to have a disabled window respond to mouse input.
Two possible solutions.
1) Redesign your UI and provide a checkbox used to enable/disable the button in question. The checkbox would always be enabled.
2) Derive a custom CButton descendant. Add a BOOL data member that would hold status info re: custom disabled/enabled state. Provide an owner-drawn capability that would draw the button in a disabled state if the custom status is disabled, and would draw the button normally otherwise. This way the button could still respond to mouse input even if it was 'disabled'.
Solution 2 is NOT recommended. It'd be a pain to implement, and the button behavior would violate UI standards.
UI standards exist for a reason. Users will be confused and frustrated by non-standard behavior that invalidates their expectations.L u n a t i c F r i n g e
|
|
|
|
|
Thank you. I had evidently taken a vacation from thinking.
I agree with everything you had to say, and a 5 vote for fixing my blunder.
I had a fuzzy image that this could be for some simulator game type of thing where a control might have a lockout to be disabled, or a cover to be opened before being activated. Your solution 1, or hiding the button and displaying a different graphic button in its place, would be a better solution for this.
If I am not mistaken again, the RButtonDown message will go the parent window. The parent window could do hit testing on it (or ask the button object to do so). This approach has the drawback of needing code outside the button object to meddle with its operation. More importantly, it has the UI problems that you pointed out and, as you said, is NOT recommended.Please do not read this signature.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
