cabfile

Read Windows .cab archives with a small, practical API.

The modern API is centered on CabFile and two helpers:

• CabFile(source) for member listing, reading, and extraction
• is_cabinet(source) -> bool for safe type checks
• probe(source) -> CabSummary for fast cabinet-level metadata

Quick start

import cabfile

with cabfile.CabFile("archive.cab") as cab:
        print(cab.namelist())
        payload = cab.read("hello.txt")

print(cabfile.is_cabinet("archive.cab"))
summary = cabfile.probe("archive.cab")
print(summary.file_count)

API shape

CabFile exposes two complementary layers:

• Member-centric methods for metadata + payload workflows:
  • read_members(names=None)
  • extract_members(target_dir, names=None)
• ZipFile-compatible methods/properties for drop-in usage patterns:
  • read, extract, extractall, namelist, infolist, getinfo, printdir, filelist, NameToInfo

CLI

The package includes a ZipFile-style CLI via the cabfile script or python -m cabfile.

# List archive members
cabfile -l archive.cab

# Test archive readability
cabfile -t archive.cab

# Extract all members to target directory
cabfile -e archive.cab output_dir

Note: Exit codes are process-style:

• 0 success
• 1 test command found unreadable archive data
• 2 usage or cabinet read error

Compatibility

Note: This project uses ctypes with Windows cabinet.dll APIs. It is intended for Windows environments.

Note: CAB decryption/password handling is not supported. ZipFile-shaped pwd parameters are accepted for compatibility but raise NotImplementedError when provided.

Background

Based on the Microsoft Cabinet SDK API surface (FDI), with compatibility ideas borrowed from zipfile.py.