2d rasterizer

This is one of my first successful C projects. A C library that construts an image, lets you access pixels individually or use functions to draw shapes, then save it as a .bmp file.

Story behind 2d-rasterizer

Initially, a long time ago, I made module that would take an array of pixels and save it as a .bmp file.

Then, I restructured it to be a library. Both a static library (.a) and a dynamic library (.so).

Then, I made the thing more useful, better, faster. (See features)

I completed bmp_draw_rect, a function that can actually draw to the image.

After that, I automated the build process to build .dll library in OS Windows using MSVC compiler.

Then I could write a python wrapper for the library that works on both Windows and Linux, so I did that.

Features

A .bmp picture file has the rows padded to 4 bytes. I made it so that the picture is row-padded from the beginning, that is, the padding is present in the pixel array in memory when the picture is still being generated.

How is the pixel's position in the array calculated?

• pixel(x, y, img) macro, that returns an l-value of a pixel struct at x, y in the img's pixel array (img being an image struct)

Whats the benefit of this?

• The image is written to the .bmp file using only three fwrite calls. (File header, DIB header, the data)
• This makes it fast because there is very little I/O overhead

Compilation/build process

On GNU/Linux systems:

I developed the library using gcc 13.1.0, but really any version should work. Also I used GNU Make 4.3.

• make to compile everything:
  • libbmap.a - static version of the library
  • libbmap.so - dynamic version of the library
  • example.bin (it will be linked against the .so dynamic library)
• make libbmap.a to compile the static library only
• make libbmap.so to compile the dynamic library only
• make example_static.bin to compile all-static version of example.bin
  • this puts the necessary parts of C standard library into the executable (not recommended)
• make clean to remove all binaries (so you can re-start the build process froms scratch)

Know that make run is for development only - I use a very specific environment that might not work for you

On Windows

While you could use any compiler to build this library (perhaps MinGW), I chose MSVC. Therefore, to use the provided build script, you need to have Microsoft Visual Studio Build Tools (when installing, select Desktop development with C++ workload).

Here are the versions I have, although, sice this build process is very simple, it should not matter if you have older/newer.

Microsoft (R) C/C++ Optimizing Compiler Version 19.35.32215 for x64
Microsoft (R) Incremental Linker Version 14.35.32215.0
Microsoft (R) Library Manager Version 14.35.32215.0

Then, in the VS Developer command prompt, you can run these commands:

• build to build the dynamic library (.dll) and example.exe linked against it
• build static to build the static library (.lib) and example.exe linked against it
• build clean to remove all binaries (sou you can re-build from scratch)
• build run to build the dynamic library + build and run example.exe

Note: A .lib file is also created alongside the .dll dynamic library. However, this file is not a static library, rather a so-called import library. This is what you then link your program against (you still need the .dll, the code is in it).

Note 2: When executing the build command, the command prompt actually executes build.bat.

Note 3: To build 64-bit binaries, you need to open x64 Native Tools Comand Prompt for VS 2022. Like stated here, the MSVC tools build 32-bit applications by default:

Visual Studio includes C++ compilers, linkers, and other tools that you can use to create platform-specific versions of your apps that can run on 32-bit, 64-bit, or ARM-based Windows operating systems. Other optional Visual Studio workloads let you use C++ tools to target other platforms, such as iOS, Android, and Linux. The default build architecture uses 32-bit, x86-hosted tools to build 32-bit, x86-native Windows code. However, you probably have a 64-bit computer. When Visual Studio is installed on a 64-bit Windows operating system, additional developer command prompt shortcuts for the 64-bit, x64-hosted native and cross compilers are available. You can take advantage of the processor and memory space available to 64-bit code by using the 64-bit, x64-hosted toolset when you build code for x86, x64, or ARM processors.

Example image

The example.c file is compiled to example.bin (perhaps example.exe). When example.bin is run, it creates example.bmp, an example image.

Future goals

In the near future, I want to:

• Implement more drawing functions
• ✅ Work on the python wrapper (add missing null checks, make it more pythonic, etc.)
• ✅ Automate + document the build process on OS Windows

Implementing new functions to the library

To add a new function, one must:

• declare it in bmp.h
• implement it in a spearate .c file
• write a recipe for it in Makefile
  • that will most probably go like this:

  # compile my_module
  my_module.o: my_module.c
  	$(CC) $(CFLAGS) -fPIC -c $<
  

• in Makefile, add it to the BMP_LIB_MODULES variable
• in build.bat, add it to the modules variable

That way, upon finishing implementation, the function is exported to the library.

The Python wrapper

The library can be used in python. Here's how to do it (Note: it's the same example as when you run libbmap.py directly):

from python import libbmap as bmp

img = bmp.Image(100, 200)
img.draw_rectangle(bmp.Point(50, 50), bmp.Point(70, 80), bmp.Color(255, 0, 0))
img.save_bmp("imported_lib_example.bmp")

• the import statement imports libbmap.py as a module from the python folder
  • to do it like this, you should place your .py file into the root directory
  • this is possible because the folder containt __init__.py
• all you need is in the libbmap.py file, however you import it
• by default, the wrapper expects you to have the shared library file (.so/.dll) in the same folder whence you run the .py file
• you can alter this behavior by changing the SO_FILE variable at the start of libbmap.py

See also

• Jan Marjanovic's bmp inspector python script - helped me find problems in my generated files