Bin2Data

Bin2Data is a command-line utility, written in QB64-PE, that converts binary files into source code, making it easy to embed resources directly into your executables. It can also optionally compress the data.

Features

• Converts binary files into embeddable source code formats.
• Optionally compresses data using QB64-PE's _DEFLATE$ function.
• Supports multiple output formats:
  • QB64-PE DATA statements (.bi file).
  • QB64-PE CONST string (.bi file).
  • C-style header with a uint8_t array (.h file).
• Can also output a raw zlib/deflate compressed file (.deflate).
• Base64 encoding for QB64-PE DATA and CONST formats.
• Command-line interface with support for wildcards for batch processing.

Building and Using Bin2Data

1. Build from source

1. Clone this repository: git clone --recurse-submodules https://github.com/a740g/Bin2Data.git
2. Change into the directory: cd Bin2Data
3. Open Bin2Data.bas in the QB64-PE IDE and press F5 to compile it.

2. Command-Line Usage

You can run Bin2Data from your terminal. The help screen shows the available options.

Bin2Data: Converts binary files to QB64-PE data
Copyright (c) 2025 Samuel Gomes
https://github.com/a740g

Usage: Bin2Data [-w characters_per_data_line] [-i compression_level] [-d] [-c] [-p] [-r] [-s] [-o] [filespec]
   -w: A number specifying the number of characters per data line. 8-4096 (default 112)
   -i: A number specifying the compression level. 1-10
   -d: Generate DATA (.bas; default)
   -c: Generate a CONST (.bas; suitable for small files)
   -p: Generate a C array (.h)
   -r: Dump the raw compressed file (.deflate)
   -s: Disable compression and store the file instead
   -o: Overwrite output file if it exists

Note:
 * Will create filespec.bi/.h/.deflate (based on the switches)
 * Can bulk convert files using wildcards
 * filespec can be a URL
 * If filespec.bi/.h/.deflate exists, then it will not be overwritten (unless -o is specified)
 * Character per line may be changed in CONST mode due to QB64's 500 line continuation limit
 * C output is barebones. Use sizeof to get the array size

Using the Generated Files

QB64-PE Projects

For QB64-PE, you need Base64.bas from the Toolbox64 library.

Example: Using DATA statements (-d switch)

This is the default mode. It creates a .bi file containing a RESTORE label and DATA statements.

1. Generate the file: Bin2Data my_asset.png This will create my_asset.png.bi.

2. In your main QB64-PE program:

' Your program logic here
' ...

' Load the resource.
' The label name is generated based on the filename and size.
' You can find the exact label name in the generated .bi file.
RESTORE data_my_asset_png_bi_12345 
DIM buffer AS STRING
buffer = Base64_LoadResourceData

' Now 'buffer' contains the binary data of my_asset.png
' ...

' At the end of your main code, before SUBs and FUNCTIONs:

' Include the generated data file
'$INCLUDE:'my_asset.png.bi'

' At the bottom of you source code, after SUBs and FUNCTIONs:

' Include the library implementation (assuming you have cloned Toolbox64 in the include directory under your project directory)
'$INCLUDE:'include/Base64.bas'

Example: Using a CONST string (-c switch)

This mode is suitable for smaller files. It creates a .bi file containing CONST definitions.

1. Generate the file: Bin2Data -c my_icon.ico This will create my_icon.ico.bi.

2. In your main QB64-PE program:

' Include the generated constants file
'$INCLUDE:'my_icon.ico.bi'

' Your program logic here
' ...

' Load the resource using the generated CONSTs.
' The CONST names are based on the filename and size.
' Check the generated .bi file for the exact names.
DIM buffer AS STRING
buffer = Base64_LoadResourceString(DATA_MY_ICON_ICO_BI_123, SIZE_MY_ICON_ICO_BI_123, COMP_MY_ICON_ICO_BI_123)

' Now 'buffer' contains the binary data of my_icon.ico
' ...

' At the bottom of you source code, after SUBs and FUNCTIONs:

' Include the library implementation (assuming you have cloned Toolbox64 in the include directory under your project directory)
'$INCLUDE:'include/Base64.bas'

C/C++ Projects

When using the -p switch, Bin2Data generates a C header file (.h).

1. Generate the file: Bin2Data -p my_data.bin This will create my_data.bin.h.

2. In your C/C++ code:

#include "my_data.bin.h"
#include <stdio.h>
#include <string.h>

int main() {
    // The header defines symbols based on the filename and size, for example:
    // - const uint8_t my_data_bin_h_567[] = {...}; (the data array)
    // - #define SIZE_MY_DATA_BIN_H_567() 567       (original size)
    // - #define COMP_MY_DATA_BIN_H_567() 1         (is compressed)
    // - #define DATA_MY_DATA_BIN_H_567() ...       (pointer to data)
    
    // Note: If the data is compressed (COMP_...() is 1),
    // you will need a zlib/deflate library to decompress it.
    
    size_t dataSize = sizeof(my_data_bin_h_567);
    
    printf("Original size: %d\n", SIZE_MY_DATA_BIN_H_567());
    printf("Embedded size: %zu\n", dataSize);
    printf("Is compressed: %d\n", COMP_MY_DATA_BIN_H_567());

    // You can now use the my_data_bin_h_567 array.
    
    return 0;
}

Notes

• This tool requires a recent version of QB64-PE to build.
• When you clone a repository that contains submodules, you need to run git submodule update --init --recursive to fetch the submodule content.

Assets

• Icon by Umut Pulat