v1.7.0

What's Changed

✨ New Features

• Expose function to read serialized schema from parquet metadata by Oliver Borchert (@borchero) in #82
• Add proper serialization support for failure info by Oliver Borchert (@borchero) in #75
• Allow to efficiently read and write collections by Oliver Borchert (@borchero) in #77
• Allow to serialize collections by Oliver Borchert (@borchero) in #74
• Introduce functionality to read and write typed parquets by Oliver Borchert (@borchero) in #66
• Enable custom data frame pre-processing during schema sampling by Moritz Potthoff (@MoritzPotthoffQC) in #71

🐞 Bug Fixes

• Fix .name property in columns returned by Schema.columns() by Daniel Elsner (@delsner) in #69
• Ensure compatibility with polars 1.31 by Jonas Süskind (@sueskind) in #79

🏗️ Testing

• Add benchmarks by Oliver Borchert (@borchero) in #78

⚙️ Automation

• Add nightly CI job by Oliver Borchert (@borchero) in #76
• Bump the gh-actions group with 4 updates by @dependabot[bot] in #68

♻️ Chores

• Update pixi lockfile by @quant-ranger[bot] in #81

Full Changelog: v1.6.0...v1.7.0