v2.1.0

What's Changed

✨ New Features

• Implement inner validation rules for Array columns by @copilot-swe-agent[bot] in #222

🐞 Bug Fixes

• Properly set nested nullability when converting to pyarrow by @copilot-swe-agent[bot] in #217
• Properly type-check non-existent Schema attributes by @copilot-swe-agent[bot] in #224
• Update migration guide for primary key column rules by @jjurm in #209

📚 Documentation

• Clarify FailureInfo write methods include rule columns by @copilot-swe-agent[bot] in #219
• Add example for SchemaClass.filter() to docstring by @DeflateAwning in #188
• Update migration guide for primary key column rules by @jjurm in #209
• Fix docs link in README by @delsner in #206

⚙️ Automation

• Replace macos-13 with macos-15-intel by @borchero in #218
• Bump the gh-actions group with 4 updates by @dependabot[bot] in #212

♻️ Chores

• Improve copilot instructions for tests by @borchero in #225
• Add initial instructions for copilot by @borchero in #216
• Update pixi lockfile by @quant-ranger[bot] in #213
• Update test suite for polars 1.35 by @borchero in #210
• Clean up mentions of the mypy plugin by @borchero in #205

Full Changelog: v2.0.0...v2.1.0

v2.0.0

What's Changed

Check out the migration guide ♻️

⚠️ Breaking Changes

• Make @dy.rule decorator apply to classmethods by @borchero in #198
• Remove support for reading collection metadata from schema.json by @borchero in #201
• Rename sql_schema to sqlalchemy_columns and prefix all conversion functions with to_ by @borchero in #183
• Split allow_inf_nan rule on floats by @borchero in #182
• Rename filter_relationship_one_to_one to require_relationship_one_to_one and add drop_unique parameter by @delsner in #67
• Make columns non-nullable by default by @AndreasAlbertQC in #174
• Raise error on nullable primary key by @AndreasAlbertQC in #173
• Rename primary_keys to primary_key by @borchero in #85
• Discontinue most parts of mypy plugin by @delsner in #103

✨ New Features

• Introduce lazy validation by @borchero in #179
• Properly support S3 in IO functions by @borchero in #139
• Make columns non-nullable by default by @AndreasAlbertQC in #174
• Support Python 3.14 by @borchero in #176
• Raise error on nullable primary key by @AndreasAlbertQC in #173
• Include validation failure information in exception after sampling exceeded maximum iterations by @MoritzPotthoffQC in #167
• Validate overrides key consistency for schema sampling by @MoritzPotthoffQC in #165

🐞 Bug Fixes

• Replace \d with [0-9] in regexes to avoid sampling non-ascii digits by @delsner in #149

📚 Documentation

• Add remaining documentation for v2 by @borchero in #202
• Revamp documentation with pydata theme by @borchero in #200
• Fix build for readthedocs by @borchero in #197
• Allow to use single backticks for code snippets by @borchero in #196
• Add data sampling docs by @delsner in #192
• Clean up small issues in docs by @AndreasAlbertQC in #180
• Document SQL generation by @AndreasAlbertQC in #178
• Add label for Python 3.14 by @borchero in #177
• Add documentation for serialization feature by @AndreasAlbertQC in #172
• Add column metadata documentation by @delsner in #171
• Convert docs to markdown by @AndreasAlbertQC in #170

⚙️ Automation

• Make nightly CI work with new polars packaging by @AndreasAlbertQC in #164

🛠 Builds

• Build for CPython on osx-64 by @borchero in #204

📦 Refactorings

• Make @dy.rule decorator apply to classmethods by @borchero in #198
• Remove support for reading collection metadata from schema.json by @borchero in #201
• Rename sql_schema to sqlalchemy_columns and prefix all conversion functions with to_ by @borchero in #183
• Split allow_inf_nan rule on floats by @borchero in #182
• Rename filter_relationship_one_to_one to require_relationship_one_to_one and add drop_unique parameter by @delsner in #67
• Rename primary_keys to primary_key by @borchero in #85
• Discontinue most parts of mypy plugin by @delsner in #103

♻️ Chores

• Fix deprecation warnings in pixi.toml by @borchero in #199

Full Changelog: v1.14.0...v2.0.0

v1.14.0

What's Changed

✨ New Features

• Add type error for pathological non-Column members in Schema by @aaronweissberg-lab in #159

🐞 Bug Fixes

• Fix max_exclusive=0 being ignored by @AndreasAlbertQC in #158

📚 Documentation

• Add multiple unique keys FAQ recipe by @DeflateAwning in #163
• Proofread primary key feature docs by @DeflateAwning in #161
• Document how to use primary keys by @AndreasAlbertQC in #157

⚙️ Automation

• Bump the gh-actions group with 9 updates by @dependabot[bot] in #152

♻️ Chores

• Update pixi lockfile by @quant-ranger[bot] in #156

Full Changelog: v1.13.0...v1.14.0

v1.13.0

What's Changed

✨ New Features

• Support Dataframes as Fields in Pydantic Models by @ytausch in #148

Full Changelog: v1.12.1...v1.13.0

v1.12.1

What's Changed

🐞 Bug Fixes

• Store Enum categories as list instead of pl.Series by @AndreasAlbertQC in #150

⚙️ Automation

• Add tests for serialization and printing of dy.Enum columns by @MoritzPotthoffQC in #147

Full Changelog: v1.12.0...v1.12.1

v1.12.0

What's Changed

✨ New Features

• Add support for Binary column types by @frewsxcv in #146

🐞 Bug Fixes

• Use covariant Sequence and Mapping in check parameter by @ytausch in #144
• Correctly compare Enums in matches by @AndreasAlbertQC in #141

Full Changelog: v1.11.0...v1.12.0

v1.11.0

What's Changed

✨ New Features

• Extend allowed types of the categories arg of dy.Enum by @jjurm in #138

Full Changelog: v1.10.0...v1.11.0

v1.10.0

What's Changed

✨ New Features

• Add support for deltalake by @AndreasAlbertQC in #134

🐞 Bug Fixes

• Correctly handle min and max bounds less than 1 for Decimal by @jhall-bp in #131

⚙️ Automation

• Include optional deps in nightly CI environment by @AndreasAlbertQC in #137
• Test optional dependencies separately by @AndreasAlbertQC in #135

📦 Refactorings

• Move IO specifics out of Schema, Collection by @AndreasAlbertQC in #109

♻️ Chores

• Update pixi lockfile by @quant-ranger[bot] in #133

Full Changelog: v1.9.0...v1.10.0

v1.9.0

What's Changed

✨ New Features

• Add join for Collections by @MoritzPotthoffQC in #127
• Allow specifying custom base collection when creating a dynamic collection by @MoritzPotthoffQC in #120

🐞 Bug Fixes

• Correct schema deserialization on failure by @florian-klein in #128
• Return original dtypes for failures in filter(..., cast=True) by @borchero in #118

🏎️ Performance Improvements

• Simplify filter evaluation for collections by @borchero in #119

📦 Refactorings

• Ensure filter and validate maintain order by @borchero in #121

Full Changelog: v1.8.2...v1.9.0

v1.8.2

What's Changed

🐞 Bug Fixes

• Correctly deserialize collections from schema by @borchero in #116

Full Changelog: v1.8.1...v1.8.2