• 0

[C++][MSVC] Cannot resolve this Linker error LNK2019, DirectX related


Question

Okay guys and gals, I am absolutely 100% stumped on this one. I genuinely cannot seem to resolve this one. I've been trying for 5 days, experimenting, googling, all to no avail. I usually like to solve my own coding problems and try for as long as I can to resolve it, but this time I need some outside advice!

Basically, I'm using DirectX for my input. I have a CInput class using dinput.h to grab keyboard inputs. My compiler is MSVC++ 2010.

CInput.h


// input.h: interface for the Cinput class.
//
//////////////////////////////////////////////////////////////////////
#if !defined(AFX_INPUT_H__9F947E83_2A3B_4B63_8675_83B41506943E__INCLUDED_)
#define AFX_INPUT_H__9F947E83_2A3B_4B63_8675_83B41506943E__INCLUDED_
#if _MSC_VER > 1000
#pragma once
#endif // _MSC_VER > 1000
#include <windows.h>
#define DIRECTINPUT_VERSION 0x0700
#include <dinput.h>

class CInput
{
private:
static CInput * _instance;
protected:
CInput(void);
public:
static CInput * Instance();
virtual ~CInput();
void GetInput();
bool GetIfKeyDown(int whichkey); //true as long as key still down
bool GetIfKeyDownEvent(int whichkey); //occurs only when key first pressed
};
#endif // !defined(AFX_INPUT_H__9F947E83_2A3B_4B63_8675_83B41506943E__INCLUDED_)
[/CODE]

CInput.cpp

[CODE]
// input.cpp: implementation of the Cinput class.
//
//////////////////////////////////////////////////////////////////////

#include "stdafx.h"
#include "Input.h"
#include <windows.h>
#include <OBJBASE.H>

extern HWND g_hWnd;
extern HINSTANCE hInst;
LPDIRECTINPUT lpdi; //pointer to direct input object
LPDIRECTINPUTDEVICE lpdikey; //the keyboard device
//storage for keyboard state
UCHAR keystate[256];
UCHAR oldkeystate[256];
bool keydown[256];
//////////////////////////////////////////////////////////////////////
// Construction/Destruction
//////////////////////////////////////////////////////////////////////
CInput::CInput()
{

if (DirectInputCreateA(hInst, DIRECTINPUT_VERSION, &lpdi, NULL)!=DI_OK)
{
// DirectInput not available; take appropriate action
}

//create keyboard device
if(lpdi->CreateDevice(GUID_SysKeyboard, &lpdikey, NULL)!=DI_OK)
{
//error
}
//set cooperation level
if(lpdikey->SetCooperativeLevel(g_hWnd, DISCL_BACKGROUND | DISCL_NONEXCLUSIVE)!=DI_OK)
{
//error
}
//set data format
if(lpdikey->SetDataFormat(&c_dfDIKeyboard)!=DI_OK)
{
//error
}
//acquire the keyboard device
if(lpdikey->Acquire()!=DI_OK)
{
//error
}

}
CInput::~CInput()
{
//unaquire keyboard
lpdikey->Unacquire();
lpdi->Release();
}
//this method refreshes the CInput class's data on which keys are currently down
//this data is stored in an array called keystate[ ]
void
CInput::GetInput()
{
//copy old keystate array
int i;
for (i=0;i<256;i++)
{
oldkeystate[i]=keystate[i];
}

if (lpdikey->GetDeviceState(256,keystate)!=DI_OK)
{
//error
}
//now check which keys are now down that weren't down last time
for (i=0;i<256;i++)
{
if ((keystate[i]>0)&&(oldkeystate[i]==0))
{
//key is now down and it wasn't before
keydown[i]=1;
}
else
{
keydown[i]=0;
}
}
}

//use this method with DirectX character constants
//which are defined in dinput.h such as DIK_SPACE
//relies on the GetInput( ) method being called first to
//update the data on which keys are currently down
//if this method doesn't seem to work then make sure you
//are calling GetInput( ) first!!!
bool
CInput::GetIfKeyDown(int whichkey)
{
if ((whichkey>256)|(whichkey<0)) return 0;
return (bool)(keystate[whichkey] & 0x80);
}
//method returns 1 or 0 depending on whether
//specified key has just been pressed
bool
CInput::GetIfKeyDownEvent(int whichkey)
{
if ((whichkey>256)|(whichkey<0)) return 0;
return (bool)(keydown[whichkey]);
}
CInput * CInput::Instance()
{
if (0==_instance)
{_instance=new CInput();
}
return _instance;
}
CInput * CInput::_instance=0;

[/CODE]

When I go to make a build, however, it compiles the code but throws up the linker error which I told you about. Here's my output from building:

[CODE]
1> Generating Code...
1>Input.obj : error LNK2019: unresolved external symbol _DirectInputCreateA@16 referenced in function "protected: __thiscall CInput::CInput(void)" (??0CInput@@IAE@XZ)
1>D:\Developer\ttSideScrollerWin32 - 2011 version\Debug\ttSideScrollerWin32.exe : fatal error LNK1120: 1 unresolved externals
========== Rebuild All: 0 succeeded, 1 failed, 0 skipped ==========
[/CODE]

From what I've read, LNK2019 usually gets thrown when something isn't being included in your project or hasn't been defined. However, the CInput Constructor has definitely been defined in my header, with both CInput.h and CInput.cpp included in my project. DirectX SDK is definitely installed, my Linker Input in my Project Properties includes dxguid.lib, dinput.lib and dinput8.lib. My VC++ Directories for Input and Libraries explicitly include DirectX SDK's folders for Input and Libraries. I've gone through everything I can think of, but everything seems to be in place. I presume it's some sort of compiler issue of some kind, and one that's probably beyond my expertise. However, I'm open to any suggestions. I'm returning to programming after a fair few months of inactivity and so I'm a little rusty, which probably hasn't helped me solve this issue.

So, please Neowin, help me!!!

2 answers to this question

Recommended Posts

  • 0

So.... yeah. I'm the world's biggest school-boy-erroring jackass.

Turns out, one small little thing I did right at the very beginning, which was re-naming dinput8.lib to dinput.lib in the very ignorant assumption would fix things, I forgot I did after I explored other avenues. Turns out, my code is pretty old now and was relying on an old version of DirectX SDK which used dinput.lib - which any DX programmer knows is pretty different to dinput8.lib (and both of which are pretty deprecated now anyway). A silly error made because I was rusty at coding. Installed a legacy version of DX which had dinput.lib, linked to that and it all compiles - thankfully. Of course, what I need to do now is re-write my Input class using a newer input library - anybody got any suggestions for input libraries in games? Just KB/M suppport is necessary but game controller support would be nice too!

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Hello, Also known for https://www.theguardian.com/technology/2009/jan/29/adware-internet.   Regards, Aryeh Goretsky    
    • Hello, I have used a few TEAM Group SSDs, USB flash drives, and Micro SDXC cards in the past. They all seemed to work fine. Regards, Aryeh Goretsky
    • "just $100 per TB"? Just? Are we trying to make this seem like the new normal? Kinda weird to make it sound like that is not a ridiculously expensive asking price.
    • The reviews you refer to mean nothing. Where there is no journalism there is no reason to call the gaming media's opinion pieces "reviews". For GP games there is indeed a metric for success - increasing subscriptions. Which turns in revenue. The only circumstance in which subs do not rise when great is being released is a Game Pass system where the company is close to fully saturated with customers in a subscription. However, in that case as the theory goes you spend aplenty in all kind of games - from shady live service cash cows and customer offending agitprop crap in purple colours to robust and entertaining single player games. And keep a solid level of profitability. Ignoring the simply innocuous but mid games MGS has released primarily of the second kind.
    • Report: Microsoft to use AWS to help GitHub deal with a major surge in demand by Pradeep Viswanathan Thanks to the surge of coding AI agents, GitHub's usage has skyrocketed over the past 12 months. To meet this demand, GitHub started with a plan in October 2025 to increase capacity by 10x. However, by early this year, the company realized that it needed 30x scale. This rapid growth has caused severe strain on the platform's reliability, resulting in several small outages over the past few months. In April, GitHub published a long blog post explaining the steps it is taking to resolve these reliability issues. In the post, the company also confirmed that it is working toward a multi-cloud architecture for better resilience. Today, Business Insider reported that GitHub is turning to Amazon Web Services to help deal with a major surge in AI-driven coding activity. It is important to note that GitHub is still in the process of moving completely to the Azure cloud. The current plan is to move the platform fully to Azure by 2027 so that it can scale better as per developer demand. Therefore, the current decision to utilize AWS might be part of a short-term plan to meet immediate demand. A Microsoft spokesperson confirmed that GitHub is using multiple cloud providers with the following statement: For Microsoft, the decision highlights the operational pressure behind the AI boom. GitHub has to stay reliable for developers at a time when rivals such as Codex, Cursor, Claude Code, and other AI coding tools are gaining attention. And the decision to use AWS for computing capacity seems practical given the circumstances.
  • Recent Achievements

    • Collaborator
      vjlex earned a badge
      Collaborator
    • Reacting Well
      Dys Topia earned a badge
      Reacting Well
    • Conversation Starter
      NovaEdgeX earned a badge
      Conversation Starter
    • One Year In
      Console General earned a badge
      One Year In
    • Week One Done
      Twozo Technologies earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      517
    2. 2
      +Edouard
      182
    3. 3
      PsYcHoKiLLa
      106
    4. 4
      Steven P.
      88
    5. 5
      ATLien_0
      68
  • Tell a friend

    Love Neowin? Tell a friend!