bmp Loader class

Philipp · 2015-11-04T01:07:26

Hello, I was building a class to load a bmp file 24 bit depth. To make it simple at first I have created a 2x2 image in paint with 4 pixels red, green, blue and white. when I open the image in a hex editor I count the 54 bit header offset and try and make sense of the values. The first 3 bytes I get is red but all the next BGR values are not what I expect them to be. also every pixel is 3 bytes but there is C++ no type with 3 bytes could I use uint or can that corrupt the colors when changing from BGR to RGB? Code so far: int bmpTextureLoader::loadBMP(const char* location, GLuint& texture) { UINT16 *datBuff[2] = { nullptr, nullptr }; UINT16 *pixels = nullptr; BITMAPFILEHEADER * bmpHeader = nullptr; BITMAPINFOHEADER * bmpInfo = nullptr; // The file... We open it with it's constructor std::ifstream file(location, std::ios::binary); if (!file) { return -1; } // Allocate byte memory that will hold the two headers datBuff[0] = new UINT16[sizeof(BITMAPFILEHEADER)]; datBuff[1] = new UINT16[sizeof(BITMAPINFOHEADER)]; file.read((char*)datBuff[0], sizeof(BITMAPFILEHEADER)); file.read((char*)datBuff[1], sizeof(BITMAPINFOHEADER)); // Construct the values from the buffers bmpHeader = (BITMAPFILEHEADER*)datBuff[0]; bmpInfo = (BITMAPINFOHEADER*)datBuff[1]; if (bmpHeader->bfType != 0x4D42) { return 2; } // First allocate pixel memory const int imageSize = bmpInfo->biSizeImage; pixels = new UINT16[imageSize]; int size = sizeof(pixels); size = size; file.seekg(bmpHeader->bfOffBits); file.read((char*)pixels, bmpInfo->biSizeImage); UINT16 tmpRGB = 0; // Swap buffer int cmpress = bmpInfo->biCompression; cmpress = cmpress; for (unsigned long i = 0; i < bmpInfo->biSizeImage; i += 3) { tmpRGB = pixels[i]; pixels[i] = pixels[i + 2]; pixels[i + 2] = tmpRGB; } // Set width and height to the values loaded from the file GLuint w = bmpInfo->biWidth; GLuint h = bmpInfo->biHeight; glGenTextures(1, &texture); // Generate a texture glBindTexture(GL_TEXTURE_2D, texture); // Bind that texture temporarily GLint mode = GL_RGB; // Set the mode glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST); glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST); //// Create the texture. We get the offsets from the image, then we use it with the image's //// pixel data to create it. //must be multiple of 2 1024 x 1024, 512 x 512 , 256 x 256 glTexImage2D(GL_TEXTURE_2D, 0, mode, w, h, 0, mode, GL_UNSIGNED_BYTE, pixels); //// Delete the two buffers. delete[] datBuff[0]; delete[] datBuff[1]; delete[] pixels; return 0; // Return success code }

Graphics and GPU Programming Programming

Started by Phil15 October 31, 2015 04:28 PM

24 comments, last by MarkS_ 8 years, 5 months ago

Phil15

184

Author

November 03, 2015 11:13 PM

Yes bmp sucks but I feel so close. This code works perfectly with bmp only one flaw when I try to free the pixels I get heap corruption. !!!
The only reason I use this format because I dont want to overload my head with compression and decompression, as I understand and correct me if am wrong everything else including targa uses compression.

Compression is optional in TGA files. I've never used it.

Ok I did some more research and now I am convinced. I will definitely look into rewriting my class to Tga at a later stage then.

MarkS_

3,509

November 03, 2015 11:22 PM

Ok I did some more research and now I am convinced. I will definitely look into rewriting my class to Tga at a later stage then.

The issue with all of the graphics formats, and why I and others are recommending you make your own, is that they are all general purpose. They are designed to cover nearly every eventuality and can be quite cumbersome and confusing. The compression part of the TGA format is an example of this. There is little reason to use it and most people don't, however, it is added for completeness and it can make things difficult.

Still, TGA is the way to go. Most people writing their own formats use TGA as an intermediate format. It is one of the most sane of the image formats and isn't restricted through patents.

Alundra

2,325

November 04, 2015 12:39 AM

The short answer is no.

But the good answer is yes if you send an alpha map, which is a grayscale texture, which is the alpha value.

It's also a common solution when you do a masked blending, you clip based on this mask texture if you don't want to use the alpha of the texture directly.

Still, TGA is the way to go. Most people writing their own formats use TGA as an intermediate format.

We prefered to use png files in all projects I worked on.

A common way is to use libPNG : http://www.libpng.org/pub/png/libpng.html

MarkS_

3,509

November 04, 2015 12:44 AM

We prefered to use png files in all projects I worked on.

Has the patent expired yet? I know that was a big deal a few years ago.

Alundra

2,325

November 04, 2015 12:51 AM

Where did you heard png is patented ?

PNG provides a patent-free replacement for GIF and can also replace many common uses of TIFF.

Source : http://ec.europa.eu/ipg/standards/image/png/index_en.htm (Last update:26/05/2015)

MarkS_

3,509

November 04, 2015 01:07 AM

Where did you heard png is patented ?
PNG provides a patent-free replacement for GIF and can also replace many common uses of TIFF.
Source : http://ec.europa.eu/ipg/standards/image/png/index_en.htm (Last update:26/05/2015)

Huh. Looks like I got that mixed up with GIF.

bmp Loader class

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

bmp Loader class

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

Reticulating splines