nrbf 0.1.2

Pure-Python parser for .NET BinaryFormatter (MS-NRBF) streams

python-nrbf

A pure-Python parser for .NET BinaryFormatter streams (MS-NRBF spec). No dependencies beyond the standard library.

Installation

pip install nrbf

# YAML output support for nrbfdump
pip install "nrbf[yaml]"

Library usage

import nrbf

# From a file
with open("data.dat", "rb") as f:
    data = nrbf.load(f)

# From bytes
data = nrbf.loads(raw_bytes)

Both functions return plain Python objects — dict, list, str, int, float, bool, or None. The .NET type name is preserved in a __class__ key on dicts so you can identify types if needed.

System.Collections.Generic.List<T> and Dictionary<K,V> are automatically unwrapped to Python list and dict. .NET enum values are unwrapped to plain ints.

Command-line usage

nrbfdump is installed automatically with the package. It reads a BinaryFormatter file (or stdin) and prints it. It automatically decompresses gzip input.

usage: nrbfdump [FILE] [--format FORMAT]

positional arguments:
  FILE                  input file (default: stdin)

options:
  -h, --help            show this help message and exit
  --format, -f          json (default), pprint, yaml

Examples:

# Decode a file to JSON
nrbfdump data.dat

# Pipe a gzip-compressed file
cat data.dat.gz | nrbfdump

# Pretty-print for quick inspection
nrbfdump data.dat --format pprint

# YAML output (requires: pip install "nrbf[yaml]")
nrbfdump data.dat --format yaml

What's decoded

All MS-NRBF record types are supported:

• All 16 primitive types (bool, int, float, double, DateTime, TimeSpan, Char, etc.)
• Strings, object references, nulls
• Single-dimensional and multi-dimensional arrays (primitive, object, string)
• Classes with full member type info (ClassWithMembersAndTypes, SystemClassWithMembersAndTypes, ClassWithId)
• System.Collections.Generic.List<T> → list
• System.Collections.Generic.Dictionary<K,V> → dict
• .NET enum values → int

License

MIT — see LICENSE.