I am working on a plugin system in C++ whereby a C++ executable loads a dll and runs plugin_start(someclass&) via GetProcAddress.
I fully understand how to pass function pointers to the dll, and visa versa, and how the dll may use anything defined in a header file, but I would like the dll to be able to use someclass where someclass is declared in someclass.h BUT DEFINED in someclass.cpp.
The catch is, someclass is compiled into the calling executable which means when the dll tries to call a function it gets a linker error. I even understand why this is, what I don't understand is how to achieve what I want.
I imagine I can pass a pointer to the object, and a pointer to the function ie someclass* somefunction* and then call it as someclass->*somefunction() but this means I would have to pass a pointer to every function in every class.
Is there an easier way to do this, or should I stick to C-style functions and function pointers alone and forget trying to pass entire classes between the two?
Thanks,
Ben
#ifndef EVENTREGISTRAR_H
#define EVENTREGISTRAR_H
#include <vector>
typedef void (__stdcall *error_callback_t)(const char *error);
class EventRegistrar
{
public:
void OnError(error_callback_t fn);
void FireError(const char *error);
private:
std::vector<error_callback_t> errors;
};
#endif
-- Cpp
#include "PluginLoader.h"
void EventRegistrar::OnError(error_callback_t fn)
{
this->errors.push_back(fn);
}
void EventRegistrar::FireError(const char *error)
{
for (std::vector<error_callback_t>::iterator it = this->errors.begin();
it != this->errors.end(); ++it)
{
(*it)(error);
}
}
-- DLL
#include "../plugin.h"
#include <stdio.h>
void __stdcall error(const char *error) { printf("Error: %s\n",error); }
extern "C" int __stdcall plugin_start(plugin_start_data& data)
{
error_callback_t fn = error;
data.events.OnError(fn);
return LOAD_SUCCESS;
}
--Error
Error 1 error LNK2001: unresolved external symbol "public: void __thiscall EventRegistrar::OnError(void (__stdcall*)(char const *))" (?OnError#EventRegistrar##QAEXP6GXPBD#Z#Z) D:\Files\C++ Workspace\BLib\BLib\Example Plugin\main.obj Example Plugin
I did something like this a long time ago. I simply used a straight C interface to keep things simple.
There may be a better way but I think passing a pointer to the object is your best and most straight-forward approach.
Related
I am new to C++ but not to programming. I'm developing a plugin and learning the language at the same time. The plugin is for old software but still being used, so I'm using VS2015 and an SDK to match. I'm having a problem that I just don't know enough to solve but I know that it's the result of something that I'm doing wrong or don't understand. Please also consider that I'm using a third party SDK, with only .H/.HPP files and an occasional .CPP, but that's it. Everything else is wrapped in their libraries. Therefore, I don't have the liberty to change any behavior.
My code snippets are parts of their headers (can't change) and the .cpp is my modified sample code that comes along with their SDK and which I'm using as my base. It is also the area of code that causes the link error. Their samples all work, I can compile them and run them no problem. My code also works and is doing what I want. Things only break when I use my modified code. The reason I'm doing this is because I need access to the message passed into the plugin and can't find any other way to get it other than to try and override "PluginMain". The original sample code actually does call into PluginSetup.cpp because it runs other code within it as setup prior to continuing on. I've only posted the part of my code which is my attempt to override the function as I mentioned and I just included the variable declaration that causes the error. If I comment my variable declaration and other code related to it, program compiles and works again. If I move the variable declaration to another .cpp file in my codebase, code compiles no problem. It just don't like being in PluginSetup.cpp but part from maybe the main.cpp file (which I can't do anything with), PluginSetup.cpp is the first that gets called. So this is where I chose to put my override.
Am I using the friend function correctly? As you can see from the codebase, they've made the ctor as well as the friend function private.
This may also go hand in hand with a question I asked before on how to instantiate a class from this implementation using private friend function and ctors?
Hopefully, what I've posted is enough to give someone all that's needed to figure out what the problem might be.
ns1ns2Main.h
namespace ns1
{
namespace ns2
{
class Plugin;
...
}
}
extern "C" __declspec(dllexport) __MainError PluginMain(const char* caller_, const char* selector_, void* message_);
ns1ns2Plugin.h
#include "ns1ns2Main.h"
namespace ns1
{
namespace ns2
{
class Plugin
{
Public:
static Plugin* const instance();
private:
friend __MainError (::PluginMain) (const char*, const char*, void*);
Plugin();
Plugin(const Plugin&);
virtual ~Plugin();
};
}
}
PluginSetup.cpp
#include "ns1ns2Main.h"
#include "ns1ns2Plugin.h"
//-> My Modification Begins
__MainError (::PluginMain) (const char* caller, const char* selector, void* message)
{
ns1::ns2::Plugin plugin;
if (!plugin.instance())
{
plugin = ns1::ns2::Plugin();
}
if (strcmp(caller, kSPInterfaceCaller) == 0)
{
if (strcmp(selector, kSPInterfaceStartupSelector) == 0)
{
bool bStatus = ns1::ns2::pluginSetup(&plugin);
if (bStatus)
{
plugin_ = clpcsx::Plugin::instance();
plugin_->startup();
}
}
else if (strcmp(selector, kSPInterfaceShutdownSelector) == 0)
{
plugin_ = clpcsx::Plugin::instance();
plugin_->shutdown();
}
}
return error;
}
//<- My Modification Ends
namespace ns1
{
namespace ns2
{
void pluginLoaded()
{
// no-op
}
bool pluginSetup(Plugin* const plugin)
{
clpcsx::Plugin::create(plugin);
plugin->setStartupCallback(NS1_NS2_CALLBACK(clpcsx::Plugin, CLPCSX_PLUG_INST, startup));
plugin->setPostStartupCallback(NS1_NS2_CALLBACK(clpcsx::Plugin, CLPCSX_PLUG_INST, postStartup));
plugin->setPreShutdownCallback(NS1_NS2_CALLBACK(clpcsx::Plugin, CLPCSX_PLUG_INST, preShutdown));
plugin->setShutdownCallback(NS1_NS2_CALLBACK(clpcsx::Plugin, CLPCSX_PLUG_INST, shutdown));
return true;
}
void pluginDestroy(Plugin* const plugin)
{
clpcsx::Plugin::destroy();
}
}
}
Link Error
1>PluginSetup.obj : error LNK2019: unresolved external symbol "private: __cdecl ns1::ns2::Plugin::Plugin(void)" (??0Plugin#ns2#ns1##AEAA#XZ) referenced in function PluginMain
You have to tell the linker to include the libraries. Since this is VS you can add to the main .cpp file
#pragma comment(lib, "xxxx.lib")
where 'xxxx.lib' is the name of the library that has those ns functions. You need to make sure they are in the VS linker path too
I am working in Visual Studio C++.
I made an class with a non-static function and packaged it as a dll. Here is the code to generate my dll:
// head file
#ifndef FUNCTIONS_H_
#define FUNCTIONS_H_
#include <string>
#include <memory>
#ifdef MATHFUNCSDLL_EXPORTS
#define MATHFUNCSDLL_API __declspec(dllexport)
#else
#define MATHFUNCSDLL_API __declspec(dllimport)
#endif
MATHFUNCSDLL_API class Functions
{
public:
MATHFUNCSDLL_API void func(int, std::string);
};
extern "C" MATHFUNCSDLL_API Functions * __cdecl create_class();
#endif
// cpp file
#include "stdafx.h"
#include "Functions.h"
#include <iostream>
void Functions::func(int id, std::string name)
{
std::cout << "Your ID: " << id << ". Your name: " << name << std::endl;
}
Functions * create_class()
{
std::cout << __FUNCTION__ << std::endl;
return new Functions();
}
Now I have a C++ project that loads this dll dynamically. Here is the code:
#include <iostream>
#include <Windows.h>
#include "../../testDmcDLL/testDmcDLL/Functions.h"
typedef Functions *(__stdcall *f_funci)();
int main(int argc, char ** argv)
{
HINSTANCE hGetProcIDDLL = LoadLibrary("C:\\Documents\\Visual Studio 2013\\Projects\\testDmcDLL\\Debug\\testDmcDLL.dll");
f_funci func_create_class = (f_funci)GetProcAddress(hGetProcIDDLL, "create_class");
Functions * pf = func_create_class();
////LNK error////pf->func(1, "toto");
system("pause");
return 0;
}
I can make sure that hGetProcIDDLL and func_create_class have been initialized successfully (I've tested them with if, but here I removed the if).
When I run this project, I can see that create_class is shown on the console because there is std::cout << __FUNCTION__ << std::endl; in that function. So everything looks fine.
However, when I compile it with the code pf->func(1, "toto") uncommented, I get a linker (LNK2019) error:
Error 1 error LNK2019: unresolved external symbol
"__declspec(dllimport) public: void __thiscall
Functions::func(int,class std::basic_string,class std::allocator >)"
(__imp_?func#Functions##QAEXHV?$basic_string#DU?$char_traits#D#std##V?$allocator#D#2##std###Z)
referenced in function _main c:\documents\visual studio
2013\Projects\testLoadDmcDLL\testLoadDmcDLL\main.obj testLoadDmcDLL
The class definition is not quite right with the exports, it should be of the form;
class MATHFUNCSDLL_API Functions // MATHFUNCSDLL_API moved
{
public:
void func(int, std::string); // MATHFUNCSDLL_API removed
};
Once a class is exported, all its members are exported.
You don't mention how MATHFUNCSDLL_EXPORTS is defined during compilation (from the command line or possibly in the stdafx.h), but make sure it is defined when building the dll, but not when building the exe. Be sure to link against the .lib produced with the .dll.
Notes on the LoadLibrary and GetProcAddress usage; if you require the dll to be loaded dynamically, you need to get the C++ class member function bound to the exported function. I've not seen a successful implementation of this or if it is even reasonable possible. If the use of the LoadLibrary and GetProcAddress is required, consider using an abstract class and create the object in a factory of some sort.
You don't detail the motivation for the dynamic loading of the dll, but consideration could also be given the delay loading the dll.
If the motivation is to delay the loading of the dll, but the same dll is always used, then the delay-load linking may help. If the motivation is to load an unknown dll (by name/location) based on some runtime parameter (or configuration), then the virtual base class and a single C-style function as a factory for the object is probably the preferred solution.
There is a good code project article on this describing various solutions for this. In particular using the abstract base class is very portable.
If you don't rely on the import library but call GetProcAddress, you need to do that for every function you're importing. You never called GetProcAddress for __imp_?func#Functions##QAEXHV?$basic_string#DU?$char_traits#D#std##V?$allocator#D#2##std###Z (which is how your Functions::func is mangled in the DLL).
Also, be aware that you get a function pointer from GetProcAddress. While that points to the code implementing pf->func, function pointers aren't called with member function call syntax.
The root problem is that GetProcAddress really is designed for C, not C++.
I would like to understand the DLL mechanism and what the compiler does when I loads the DLL at run-time (i.e. I will not use the generated .lib).
Consider the following C++ code:
DLL interface header file
#ifdef MYDLL_EXPORTS
#define MYDLL_API __declspec(dllexport)
#else
#define MYDLL_API __declspec(dllimport)
#endif
class MYDLL_API Base
{
public:
Base();
virtual ~Base();
virtual int get_number() const;
virtual const char* what() const = 0;
private:
int i_;
};
class MYDLL_API Child : public Base
{
public:
Child();
virtual ~Child();
virtual int get_number() const override;
virtual const char* what() const override;
private:
int j_;
};
extern "C" {
MYDLL_API Base* __cdecl initializeObject();
}
DLL implementation source file
#include "MyDLL.hh"
Base::Base()
: i_(42)
{}
Base::~Base()
{}
int Base::get_number() const
{
return i_;
}
Child::Child()
: Base()
, j_(24)
{}
Child::~Child()
{}
int Child::get_number() const
{
return j_;
}
const char* Child::what() const
{
return "Hello!";
}
Base* initializeObject()
{
return new Child();
}
The goal of this DLL is to have a common interface defined by the Base class, but it allows specifics implementations compiled in different DLLs that are loaded at runtime (here the Child class is exposed for the purpose of the example).
At this stage, if I naively include the DLL's header:
#include "MyDLL.hh"
int main()
{
Base* b = new Child();
std::cout << b->get_number() << std::endl;
std::cout << b->what() << std::endl;
delete b;
getchar();
return 0;
}
The linker complains LNK2019 and LNK2001 errors: it can not resolves symbols. So, it behaves as expected (I did not use the .lib).
Consider now, the following code that I use to load the DLL at runtime:
#include "MyDLL.hh"
typedef Base* (*initFuncType)();
int main()
{
HINSTANCE handle = LoadLibrary(L"MyDLL.dll");
initFuncType init = nullptr;
init = (initFuncType)(GetProcAddress(handle, "initializeObject"));
if (init)
{
Base* b = init(); //< Use init() !
std::cout << b->get_number() << std::endl;
std::cout << b->what() << std::endl;
delete b;
}
getchar();
FreeLibrary(handle);
return 0;
}
This time it works, the linkage is done.
1st question: What happened? What changed for the compiler and the linker? The use of the function pointer on initializeObject() solves the problem.
The other issue I do not understand well is when I remove virtual and override of get_number():
int get_number() const;
I have a LNK2019 error because of the unresolved Base::get_number(void) const symbol in the _main function. I understand that the virtual keyword will resolve the member function dynamically (at run-time). In our case, the DLL is not loaded yet, the get_number symbol is not available.
2nd question: Does this means that methods must always be virtual using DLL run-time linking?
3rd question: How can I have the C++ function exportation with the Windows API? So that I could remove the extern "C" { ... } stuff.
Thanks for your reading! I hope I will read interesting answers! :)
There are two ways to link dll files.
The 2nd way (the way it works) is the C Binding approach, where you query the dll for a specific function name and it returns a functor to you.
Using the 2nd way you won't be able to extend the base classes, since they are not defined (you don't have any code to be copy pasted so to speak at linkage time).
In order to have a dll who's classes can be extended, you will need to use dynamic binding. You need to compile your .dll and also provide a Symbols Library (or an export library). You have this option in VS studio in project properties.
The mechanism is as following :
Compile Dll project -> output : myLib.dll , myLib.lib
Use exported symbols from myLib.lib inside your main project (main project takes myLib.lib as dependency)
at runtime,due to the binding, your program will know it requires myLib.dll to work so it will load it (if found, else you get runtime error)
Another advantage of using Export Library is that you can export C++ functions (which are mangled on export).
It's very hard to have C Binding on mangled functions.
C Binding on the otherhand, compared to the dynamic binding, won't make your program scream if myLib.dll isn't found , you will just get a null pointer to function.
I am trying to link to an external library in my QT application. The external library has a header file with the following relevant code I'm trying to call:
extern VGRABDEVICE_API bool V_AssignFrameSizeCallback(IGrabChannel* pChannel, void* pFunc);
In the demo C++ program provided, which has no problems compiling, the following relevant code is:
// in main.cpp
void _stdcall MyFrameSizeCallback(T x) {
do_stuff;
}
int main(int argc, char* argv[]) {
IGrabChannel* pChannel0 = something;
V_AssignFrameSizeCallback(pChannel0, MyFrameSizeCallback);
}
I am trying to incorporate this code into my QT application, but getting problems. In my mainwindow.cpp file:
void _stdcall MainWindow::MyFrameSizeCallback(T x) {
do_stuff;
}
void MainWindow::someFunction() {
IGrabChannel* pChannel0 = something;
V_AssignFrameSizeCallback(pChannel0, &MainWindow::MyFrameSizeCallback);
}
The error I'm getting is:
error: C2664: 'bool V_AssignFrameSizeCallback(IGrabChannel *,void *)' :
cannot convert argument 2 from 'void (__cdecl MainWindow::* )(T)' to 'void *'
There is no context in which this conversion is possible
What do I need to do? Thanks.
You have two problems. First, void* is a data pointer, not a function pointer. According to the C++ standard, casting between the two is not expected to work. Some platforms provide a stronger guarantee... for example Windows GetProcAddress and *nix dlsym mix the two.
Next, your &MainWindow::MyFrameSizeCallback is not a function pointer, it is a pointer-to-member-function. Calling it requires a MainWindow object, which the external library doesn't know anything about.
You need to provide an ordinary function, not a member function, to the library. If you have some way to get ahold of the MainWindow* object pointer, you can then call its member function to do the real work. Sometimes the library provides a "context" parameter which is passed to your callback; that's a great place to put the object pointer. Otherwise, you'll need to store your MainWindow* in a global variable. Easy if you have just one, while if you have more than one you might go with std::map<IGrabChannel*, MainWindow*>.
Code:
MainWindow* MainWindow::the_window;
void MainWindow::MyFrameSizeCallback(T x)
{
do_stuff;
}
void _stdcall MyFrameSizeCallbackShim(T x)
{
MainWindow::the_window->MyFrameSizeCallback(x);
}
void MainWindow::someFunction()
{
IGrabChannel* pChannel0 = something;
the_window = this;
V_AssignFrameSizeCallback(pChannel0, &MyFrameSizeCallbackShim);
}
If the parameter x isn't an IGrabChannel, change the map datatype and insertion logic accordingly. If the parameter x isn't some sort of unique predictable identifier, you may be limited to only doing callbacks to one MainWindow instance.
I'm trying to port a Windows app to Linux. This appplication marks some functions with the __stdcall attribute. However, I was told by a friend that stdcall is used only on Windows and has no meaning in Linux (but DOES exist in Windows GCC).
Searching Google - some results state that there IS stdcall in Linux.
Is there a stdcall in Linux?
Additionally, GCC indicates that:
__attribute__((__stdcall__)) and __attribute__((stdcall)) (without the underscores near stdcall).
Which one is preferred (if applied to Linux at all)?
The simplest solution is to just define __stdcall to nothing conditionally on Linux.
Here's a link to __stdcall description on MSDN:
http://msdn.microsoft.com/en-us/library/zxk0tw93(VS.80).aspx
It's only used to call WinAPI functions. To port such a Windows application to Linux, you need much more than just defining __stdcall to nothing:
#ifndef WIN32 // or something like that...
#define __stdcall
#endif
You would also need to call the Linux-specific API functions instead of Win32 API ones. Depending on the particular part of Win32 API and the size of the application (amount of code), it can be anywhere between moderately difficult and daunting.
Which specific functions are marked by the app as __stdcall?
Indeed, Windows port of GCC has to have __stdcall, because it's supposed to be able to generate conforming code for the Win32 platform. But since under Linux there is only one standard calling convention and it coincides with the default compiler output, this statement is not needed.
The reason your application is not compiling under Linux is almost certainly due to the fact, that it references Win32 API functions that are not defined under Linux -- you need to find appropriate Linux counterparts. Win32 API and Linux GLibc API-s are very much different and cannot be substituted easily.
Probably the easiest way to port your app to Linux would be to use Wine, i.e. modifying the Windows code in such a way, that it runs smoothly under Wine in Linux. This is the way even the most complex applications, like modern computer games, have been made to run under Linux.
Of course, if you really want it to be running natively under Linux, then porting is the only way to go.
stdcall is NOT just a calling convention; in addition to being a calling convention, it allows an isomorphism between C and C++ objects. Here's an example:
#define _CRT_SECURE_NO_WARNINGS // disable marking use of strcpy as error.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
class ICdeclGreeter {
public:
virtual ~ICdeclGreeter(){}
virtual void setGreeting(const char *greeting) = 0;
virtual void greet() = 0;
};
class IStdcallGreeter {
public:
virtual __stdcall ~IStdcallGreeter(){}
virtual void __stdcall setGreeting(const char *greeting) = 0;
virtual void __stdcall greet() = 0;
};
class CdeclGreeter : public ICdeclGreeter {
public:
char *greeting;
~CdeclGreeter() {
if (greeting != nullptr) {
free(greeting);
puts("[CdeclGreeter] destroyed");
}
}
void setGreeting(const char *greeting) {
this->greeting = (char *)malloc(strlen(greeting) + 1);
strcpy(this->greeting, greeting);
}
void greet() {
puts(greeting);
}
};
class StdcallGreeter : public IStdcallGreeter {
public:
char *greeting;
__stdcall ~StdcallGreeter() {
if (greeting != nullptr) {
free(greeting);
puts("[StdcallGreeter] destroyed");
}
}
void __stdcall setGreeting(const char *greeting) {
this->greeting = (char *)malloc(strlen(greeting) + 1);
strcpy(this->greeting, greeting);
}
void __stdcall greet() {
puts(greeting);
}
};
typedef struct pureC_StdcallGreeter pureC_StdcallGreeter;
typedef struct pureC_StdcallGreeterVtbl {
void (__stdcall *dtor)(pureC_StdcallGreeter *This);
void (__stdcall *setGreeting)(pureC_StdcallGreeter *This, const char *greeting);
void (__stdcall *greet)(pureC_StdcallGreeter *This);
} pureC_IStdcallGreeterVtbl;
struct pureC_StdcallGreeter {
pureC_IStdcallGreeterVtbl *lpVtbl;
char *greeting;
int length;
};
/* naive attempt at porting a c++ class to C;
on x86, thiscall passes This via ecx register rather than
first argument; this register cannot be accessed in C without
inline assembly or calling a reinterpretation of byte array
as a function. there is no "This" argument in any of below. */
typedef struct pureC_CdeclGreeter pureC_CdeclGreeter;
typedef struct pureC_CdeclGreeterVtbl {
void (*dtor)(pureC_CdeclGreeter *This);
void (*setGreeting)(pureC_CdeclGreeter *This, const char *greeting);
void (*greet)(pureC_CdeclGreeter *This);
} pureC_CdeclGreeterVtbl;
struct pureC_CdeclGreeter {
pureC_CdeclGreeterVtbl *lpVtbl;
char *greeting;
int length;
};
void test() {
ICdeclGreeter *g = new CdeclGreeter;
g->setGreeting("hi");
g->greet();
IStdcallGreeter *g2 = new StdcallGreeter;
g2->setGreeting("hi");
g2->greet();
// we can pass pointers to our object to pure C using this interface,
// and it can still use it without doing anything to it.
pureC_StdcallGreeter *g3 = (pureC_StdcallGreeter *)g2;
g3->lpVtbl->setGreeting(g3, "hello, world!");
g3->lpVtbl->greet(g3);
g3->lpVtbl->dtor(g3);
free(g2);
/*
// cdecl passes this via ecx in x86, and not as the first argument;
// this means that this argument cannot be accessed in C without
// inline assembly or equivelent. Trying to run code below will cause a runtime error.
pureC_CdeclGreeter *g4 = (pureC_CdeclGreeter *)g;
g4->lpVtbl->setGreeting(g4, "hello, world!");
g4->lpVtbl->greet(g4);
g4->lpVtbl->dtor(g4);
free(g);
*/
delete g;
}
int main(int argc, char **argv)
{
test();
system("pause");
return 0;
}
TLDR; it's not the same as cdecl makes C++ classes not usable from C on platforms using this convention because in order to send "This" to a method, you must set ecx register to address of "This" rather than just pushing it, and likewise if you want to implement a class in C that C++ can recognize, the method will need to get This pointer from ecx register which is not accessible to C without inline assemby or equivelent.
stdcall has this nice property that classes that use stdcall can easily be simultaneously usable from C or C++ without doing anything to them.
So you can only #define __stdcall as long as you don't deal with __thiscall; although there might be some other subtle distinctions.