ipadic

What it is and what it does

ipadic is a Python package that bundles IPAdic, a Japanese morphological dictionary originally included with MeCab. It is designed to work as a companion to morphological analyzers like mecab-python3 or fugashi, providing the dictionary data needed to perform Japanese text tokenization and part-of-speech tagging. The package contains no code of its own—it is purely a data distribution mechanism.

The package is explicitly discouraged by its maintainer. The underlying dictionary has not been updated since at least 2007 and lacks modern Japanese terminology. The authors recommend using Unidic instead, which is actively maintained. ipadic is retained only for compatibility with old benchmarks or models that depend on the original dictionary.

Use it for:

• Running legacy Japanese NLP models or benchmarks that were trained with the original IPAdic dictionary.
• Reproducing historical results or papers that explicitly used IPAdic for tokenization and tagging.
• Maintaining compatibility with existing code that hard-codes ipadic.MECAB_ARGS for analyzer initialization.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides the IPAdic Japanese morphological dictionary for use with mecab-python3 or fugashi, enabling Japanese text tokenization and part-of-speech tagging.

No, unless you are specifically maintaining legacy code or reproducing old benchmarks. The package is abandoned (last update 2020-07-20), the underlying dictionary is outdated (not updated since at least 2007), and the maintainers explicitly recommend alternatives. High install friction and unclear license treatment add further friction. Install only if you have a documented reason to use the original IPAdic.

Install

ipadic on PyPI

pip install ipadic

uv add ipadic

poetry add ipadic

Installing ipadic

Before you install

High install friction: the package is a large dictionary file with no runtime dependencies. Maintenance is abandoned—last commit was 2021-10-31. The authors explicitly recommend against using this package in favor of actively maintained alternatives.

License in practice

License treatment is unclear. The dictionary itself is copyrighted by the Nara Institute of Science and Technology and ICOT under terms in the COPYING file; the wrapper code is MIT or WTFPL. Verify the full license terms before use in proprietary or restricted contexts.

Quickstart

pip install ipadic

import ipadic
# Use ipadic.MECAB_ARGS with a compatible morphological analyzer

Requires a separate morphological analyzer (mecab-python3 or fugashi) to be installed and configured; ipadic alone is only a dictionary file.

Verify before relying

• Whether the dictionary's lack of updates since at least 2007 and missing modern terms is acceptable for your use case.
• Full license compliance requirements for the Nara Institute / ICOT dictionary terms in your deployment context.
• Exact Python version support requirements (listed as unspecified in metadata).

Package facts

Evidence: ipadic-1.0.0.tar.gz

Tags

More Linguistic packages