A Python library for reading and writing structured binary data.
A cross-platform Python 3 library for reading and writing structured binary data in an object-oriented (ish) style.
Why use binobj?
You may have used Python’s built-in struct library to load and dump binary data. It’s unwieldy for larger or more complex data structures, and the format strings are easy to get wrong. binobj is different in that it takes a class-based approach to declaring binary structures.
Take a look at this example using struct:
data = (b'BM', 1024, 0, 12, 40, 32, 32, 1, 1, 0, 0, 72, 72, 2, 2) header_bytes = struct.pack('<2sIIIIiiHHIIiiII', *data) loaded = struct.unpack('<2sIIIIiiHHIIiiII', header_bytes) n_pixels = loaded * loaded
The same example rewritten using binobj:
class BMP(binobj.Struct): magic = binobj.Bytes(const=b'BM') file_size = binobj.UInt32() _reserved = binobj.Bytes(const=b'\0\0\0\0', discard=True) pixels_offset = binobj.UInt32() # Legacy DIB header header_size = binobj.UInt32(const=40) image_width = binobj.Int32() image_height = binobj.Int32() n_color_planes = binobj.UInt16() n_bits_per_pixel = binobj.UInt16() compression_method = binobj.UInt32(default=0) bitmap_size = binobj.UInt32() v_resolution = binobj.Int32() h_resolution = binobj.Int32() n_palette_colors = binobj.UInt32() n_important_colors = binobj.UInt32() bmp = BMP(file_size=1024, pixels_offset=12, image_width=32, image_height=32, ...) header_bytes = bytes(bmp) loaded = BMP.from_bytes(header_bytes) n_pixels = loaded.image_width * loaded.image_height
binobj also has other advantages in that it supports strings in any encoding Python supports, toggling endianness on a per-field basis (necessary for ISO 9660 images), a variety of integer encodings, computed fields, validation, and more.
- This package will not work on a mixed-endian system. Those are pretty rare nowadays so chances are you won’t have a problem.
- This has been tested on Python 3.4-3.7, and PyPy3.5.
Sorry, I have no intention of supporting Python 2. Feel free to fork this and do a backport if you like! I’d be interested to see it and might even contribute.
You can install this with pip like so:
pip3 install binobj
- Be sure to use pip3 and not pip, because pip defaults to Python 2.
- If you get a “Permission Denied” error, try:
pip3 install --user binobj
Side note: Don’t use sudo (even sudo -EH) to force a package to install, as that’s a security risk. See this answer on Stack Overflow to find out why.
Testing and Development
This package uses Tox to run tests on multiple versions of Python.
To set up your development environment, you’ll need to install a few things.
- For Python version management, I use pyenv-virtualenv. Follow the installation instructions there.
- You’ll also need make. Depending on your platform you can install it in
one of several ways:
- macOS: brew install make
- Debian systems (e.g. Ubuntu): sudo apt-get install make
- Windows: Use Cygwin and install it during setup.
Once you have those installed, in the root directory of this repo run:
Running the Tests
To run the unit tests for all supported versions of Python, run make test. The environments will automatically be rebuilt if needed.
Issues and Feature Requests
To report an issue, request a feature, or propose a change, please file a report on the project’s GitHub page here.
Release history Release notifications
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size & hash SHA256 hash help||File type||Python version||Upload date|
|binobj-0.5.0-py3-none-any.whl (29.1 kB) Copy SHA256 hash SHA256||Wheel||py3|
|binobj-0.5.0.tar.gz (26.1 kB) Copy SHA256 hash SHA256||Source||None|