Summary

chardet 7 is a Python character encoding detector offering high accuracy and speed. It targets developers and users needing to process unknown text data, providing a drop-in replacement for older versions with significant performance gains. Its 0BSD license ensures broad applicability.

How It Works

The library utilizes a 13-stage detection pipeline, incorporating Byte Order Mark (BOM) detection, magic number identification, structural probing, byte validity filtering, and bigram statistical models. Optional mypyc compilation further accelerates processing. This comprehensive approach yields superior accuracy and speed compared to predecessors and competitors.

Quick Start & Requirements

• Primary install: pip install chardet
• Prerequisites: Python 3.10+; zero runtime dependencies.
• Documentation: chardet.readthedocs.io

Highlighted Details

• Accuracy: 99.3% on 2,517 files, a +11.1 percentage point improvement over chardet 6.0.0.
• Speed: 47x faster than chardet 6.0.0, 1.5x faster than charset-normalizer 3.4.6.
• Language Detection: 95.7% accuracy across 49 languages.
• MIME Type Detection: Identifies 40+ binary formats and common text types.
• Streaming: Supports incremental processing for large data.
• Filtering: Features encoding era and specific encoding include/exclude filters.

Maintenance & Community

chardet 7.x is a 2026 ground-up rewrite by Dan Blanchard, distinct from earlier codebases. Historical commits from the original author are preserved in a separate branch. No specific community channels are listed.

Licensing & Compatibility

Licensed under the permissive 0BSD license, allowing unrestricted commercial and closed-source use.

Limitations & Caveats

chardet 7.x is a complete rewrite, not a derivative of pre-version 7 code. While API-compatible, this architectural divergence may be a factor for users requiring strict code lineage.