---
id: xpress8
version: "0.1.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# xpress8 — Python bindings for the Microsoft Xpress8 (ESE) compression library

License: permissive · Maintenance: active · Downloads: 103.7K/mo

## What it is and what it does

xpress8 is a Cython wrapper around Microsoft's Xpress8 compression algorithm, sourced from the Extensible Storage Engine (ESE) library. It exposes a simple Python API for single-buffer compression and decompression, as well as support for chunked streams with 4-byte headers—the format used by Power BI's .pbix files. The package vendors the underlying C implementation from ESE and compiles it as a native extension.

The library is designed for scenarios where you need to work with Xpress8-compressed data, particularly .pbix files or Windows-internal data stores. It is not compatible with Xpress9, a related but distinct compression format. Installation is straightforward via prebuilt wheels for common Python versions and platforms, though you must specify the uncompressed size when decompressing.

Use it for:

- Decompress Power BI .pbix files to extract or analyze their internal structure and data.
- Compress or decompress data in Windows-internal formats that use Xpress8 encoding.
- Handle chunked Xpress8-compressed streams with embedded size headers in a Python application.
- Integrate Xpress8 compression into data pipelines that must interoperate with ESE-based systems.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

xpress8 provides Python bindings to Microsoft's Xpress8 compression library, enabling compression and decompression of data using the algorithm used by Power BI's .pbix format and Windows internal data stores.

Yes, if you need to work with Xpress8-compressed data (especially .pbix files). The package is actively maintained, has no known vulnerabilities, uses a permissive MIT license, and offers prebuilt wheels that minimize install friction. The 0.1.0 version and recent release date suggest early maturity—test thoroughly in non-critical contexts before relying on it for production workloads.

## Install

pip install xpress8
uv add xpress8
poetry add xpress8

## Installing xpress8

Before you install: Medium install friction due to compiled extension wheels; however, prebuilt wheels are available for Python 3.9–3.13 across macOS, Linux (x86_64 and aarch64), and Windows, reducing friction significantly. Repository is active with recent commits.

License in practice: MIT license is permissive and places no restrictions on use, modification, or distribution in proprietary or open-source projects. Vendored Microsoft sources retain their original copyright headers but are also MIT-licensed.

Quickstart:

from xpress8 import Xpress8

x = Xpress8()
compressed = x.compress(b"hello world " * 64)
restored = x.decompress(compressed, len(b"hello world " * 64))

Requires Python 3.8 or later; compiled extension, so platform-specific wheel must match your Python version and architecture.

Verify before relying:

- Performance characteristics (compression ratio, speed) compared to other compression libraries or Xpress9.
- Maximum practical data size limits beyond the default max_original_size of 65536 bytes.
- Stability and production-readiness status given the 0.1.0 version number and recent first release.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 103.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

xpress8 compression python, pbix file compression, microsoft xpress8 bindings, ese compression library, data compression decompression, windows compression format, chunked stream decompression, compression, microsoft-ese, pbix

[View on SkillFed](https://skillfed.io/packages/xpress8) · [View on PyPI](https://pypi.org/project/xpress8/)