---
id: pyhwp
version: "0.1b15"
license: GNU Affero General Public License v3 or later (AGPLv3+)
license_treatment: agpl
maintenance: active
---

# pyhwp — hwp file format parser

License: agpl · Maintenance: active · Downloads: 329.8K/mo

## What it is and what it does

pyhwp is a parser and processor for HWP (Hancom Word Processor) Document Format v5 files, the native format used by Korean word processing software. It allows you to read HWP files, analyze their internal structure, and extract embedded streams and content. The package includes experimental features for converting HWP documents to OpenDocument (.odt) or plain text (.txt) formats, making it useful for document analysis, format conversion, and content extraction workflows.

The package depends on cryptography, lxml, and olefile for its core functionality. It is actively maintained and supports multiple Python implementations (CPython, Jython, PyPy) across Python 2.7 and 3.5–3.8. However, it is distributed as a source tarball and remains in beta status, requiring the --pre flag during installation.

Use it for:

- Extract text and metadata from HWP documents for indexing or archival purposes
- Convert HWP files to OpenDocument format for compatibility with other office suites
- Analyze the internal structure and streams of HWP files for forensic or research purposes
- Batch process Korean word processor documents in automated workflows
- Migrate legacy HWP documents to open formats for long-term preservation

## Worth the install?

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

Parses and extracts data from HWP Document Format v5 files, with experimental support for conversion to OpenDocument (.odt) or plain text (.txt) formats.

Yes, with conditions. Install if you need to parse or extract data from HWP files and can tolerate beta-status software and high install friction. The package is actively maintained, has no known vulnerabilities, and fills a niche need for HWP document handling. However, be aware that the last release was in 2020, the experimental conversion features may not be production-ready, and AGPLv3+ licensing imposes copyleft obligations on derivative works.

## Install

pip install pyhwp
uv add pyhwp
poetry add pyhwp

## Installing pyhwp

Before you install: High install friction due to source distribution and no runtime dependencies listed in metadata. Package is actively maintained with recent commits, but last release was in 2020; requires --pre flag for installation due to beta versioning.

License in practice: Licensed under GNU Affero General Public License v3 or later (AGPLv3+). This is a copyleft license requiring that any modifications or derivative works distributed must also be released under AGPLv3+, and may impose network use restrictions depending on deployment context.

Quickstart:

pip install --pre pyhwp

import pyhwp
# Parse HWP file and extract streams
with open('document.hwp', 'rb') as f:
    doc = pyhwp.parse(f)

Requires cryptography, lxml, and olefile as runtime dependencies; source distribution requires build tools; Python 2.7 or 3.5+ support declared but package is in beta.

Verify before relying:

- Whether the package actually works reliably with Python 3.8+ given the last release was 2020 and Python has evolved significantly
- Current state of the experimental ODT and plain text conversion features and their stability
- Whether cryptography, lxml, and olefile are correctly declared as install-time dependencies despite not appearing in the metadata

## Package facts

- License: GNU Affero General Public License v3 or later (AGPLv3+) (agpl)
- Python support: unspecified
- Install friction: high
- Maintenance: active
- Downloads: 329.8K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

hwp file parser, hwp document extraction, korean word processor format, hwp to odt conversion, hwp document analysis, hancom hwp reader, hwp stream extraction, document-parsing, korean-office-format, format-conversion

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