Support

If you have a problem or need to report a bug please email : support@dsprobotics.com

There are 3 sections to this support area:

DOWNLOADS: access to product manuals, support files and drivers

HELP & INFORMATION: tutorials and example files for learning or finding pre-made modules for your projects

USER FORUMS: meet with other users and exchange ideas, you can also get help and assistance here

NEW REGISTRATIONS - please contact us if you wish to register on the forum

Users are reminded of the forum rules they sign up to which prohibits any activity that violates any laws including posting material covered by copyright

MinGW / NetBeans IDE DLL for Bitmap

For general discussion related FlowStone

MinGW / NetBeans IDE DLL for Bitmap

Postby aronb » Mon Jul 24, 2017 3:48 am

Hi,

Has anyone created a DLL for Bitmap manipulation using the MinGW / gcc / gc++ / NetBeans IDE as outlined on the Flowstone Guru site?

The MinGW / gcc / gc++ / NetBeans IDE are the preferred method of making DLL's for Flowstone currently according to Flowstone Guru.

I have used (MinGW / gcc / gc++ / NetBeans IDE) to create DLL's doing other functions (audio delay, format manipulation etc.) but have not used it for Bitmap manipulation.

The Flowstone Example DLL has an example (and I have tested and it works) but they used Visual Studio 2008 to compile the DLL - unfortunately that version of Visual Studio is long gone... VS 2015 now...

Also, I noticed something in the "Helper Macros"
Code: Select all
#define GETFRAMESIZE(p) p ? *((int*)p) : 0
#define GETBITMAPWIDTH(p) p ? *((int*)p) : 0
#define GETBITMAPHEIGHT(p) p ? *((int*)p+1) : 0
#define GETBITMAPCHANNELS(p) p ? *((int*)p+2) : 0
#define GETBITMAPDATA(p) p ? ((BYTE*)p+12) : 0
#define GETBITMAPBYTES(p) p ? *((int*)p) * *((int*)p+1) * *((int*)p+2) : 0
#define NEWINTARRAY(p,n) if(n>0) { *((int**)&p)=new int[n+1]; ((int*)p)[0]=n; }

Notice the BYTE* in the GETBITMAPDATA (#define)... does C or C++ allow the use of BYTE* ? should that be int* ?

Any help or examples using the MinGW / gcc / gc++ / NetBeans IDE for Bitmap manipulation would be greatly appreciated.

Aron
User avatar
aronb
 
Posts: 154
Joined: Sun Apr 17, 2011 3:08 am
Location: Florida, USA

Re: MinGW / NetBeans IDE DLL for Bitmap

Postby TheOm » Tue Jul 25, 2017 6:25 pm

aronb wrote:The MinGW / gcc / gc++ / NetBeans IDE are the preferred method of making DLL's for Flowstone currently according to Flowstone Guru.

That's only the preferred method of the author of that post, nothing official.
I personally would recommend you use Visual C++.

aronb wrote:The Flowstone Example DLL has an example (and I have tested and it works) but they used Visual Studio 2008 to compile the DLL - unfortunately that version of Visual Studio is long gone... VS 2015 now...

The version of Visual Studio and/or the visual c++ compiler that you use doesn't matter.
You can use VS2015 or VS2017 without a problem.

aronb wrote:Notice the BYTE* in the GETBITMAPDATA (#define)... does C or C++ allow the use of BYTE* ? should that be int* ?

If you are building a dll you would usually be including the windows.h header, which defines BYTE as unsigned char.


Here's a simple example for flipping a bitmap, but I only tested with VS2017.
Code: Select all
#include <windows.h>
#include <new>
#include <cstring>

extern "C"
BOOL WINAPI DllMain(HINSTANCE, DWORD reason, LPVOID) {
   switch(reason) {
   case DLL_PROCESS_ATTACH:
   case DLL_THREAD_ATTACH:
   case DLL_THREAD_DETACH:
   case DLL_PROCESS_DETACH:
      return TRUE;
   default:
      return FALSE;
   }
}

class FSBitmapView {
   unsigned char* bytes_ptr;
   bool m_allocated;

   FSBitmapView(unsigned char* ptr, bool allocd) : bytes_ptr(ptr), m_allocated(allocd) {}

public:
   static FSBitmapView from_input(int input) {
      FSBitmapView view(reinterpret_cast<unsigned char*>(input), false);
      return view;
   }

   FSBitmapView() : FSBitmapView(nullptr, false) {}

   bool allocate_for_output(int& output, int width, int height, int channels) {
      deallocate();

      bytes_ptr = new(std::nothrow) unsigned char[12 + width * height * channels];

      if(bytes_ptr != nullptr) {
         this->width() = width;
         this->height() = height;
         this->channels() = channels;
         m_allocated = true;
         output = reinterpret_cast<int>(bytes_ptr);
      }

      return bytes_ptr != nullptr;
   }

   int& width() {
      return *reinterpret_cast<int*>(bytes_ptr);
   }

   int& height() {
      return *reinterpret_cast<int*>(bytes_ptr + 4);
   }

   int& channels() {
      return *reinterpret_cast<int*>(bytes_ptr + 8);
   }

   unsigned char* pixels() {
      return bytes_ptr + 12;
   }

   unsigned char* pixel_at(int x, int y) {
      return pixels() + y * width() * channels() + x * channels();
   }

   bool ok() const {
      return bytes_ptr != nullptr;
   }

   void deallocate() {
      if(m_allocated) {
         delete[] bytes_ptr;
         bytes_ptr = nullptr;
      }
   }

   ~FSBitmapView() {
      deallocate();
   }
};

static FSBitmapView bmp_out;

extern "C"
__declspec(dllexport)
void flip_bitmap(int num_params, int* inputs, int* outputs) {
   if(num_params < 1 || inputs == nullptr || outputs == nullptr)
      return;

   auto bmp_in = FSBitmapView::from_input(inputs[0]);
   const int w = bmp_in.width();
   const int h = bmp_in.height();
   const int channels = bmp_in.channels();

   if(!bmp_out.allocate_for_output(outputs[0], w, h, channels)) {
      return;
   }

   for(int y = 0; y < h; ++y) {
      for(int x = 0; x < w; ++x) {
         for(int c = 0; c < channels; ++c) {
            bmp_out.pixel_at(x, h - (y + 1))[c] = bmp_in.pixel_at(x, y)[c];
         }
      }
   }
}
TheOm
 
Posts: 103
Joined: Tue Jan 28, 2014 7:35 pm
Location: Germany


Return to General

Who is online

Users browsing this forum: No registered users and 31 guests