Merge branch 'expr-ir/over-and-over-and-over-again' into expr-ir/strict-selectors

Author: dangotbanned

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -10,7 +10,7 @@ body:
 - type: markdown 
   attributes:
     value: >
-       ### Before submitting a request, please check if it has already been discused in the  
+       ### Before submitting a request, please check if it has already been discussed in the  
        [list of issues](https://github.com/narwhals-dev/narwhals/issues?q=is%3Aissue+is%3Aclosed%2Bopen+).
 - type: textarea
   attributes:

.github/workflows/downstream_tests.yml

@@ -48,10 +48,12 @@ jobs:
         run: |
           cd altair
           pytest tests
-      - name: Run mypy
-        run: |
-          cd altair
-          mypy altair tests
+      # Temporarily commented-out to get CI green
+      # https://github.com/narwhals-dev/narwhals/issues/3013
+      # - name: Run mypy
+      #   run: |
+      #     cd altair
+      #     mypy altair tests
 
   marimo:
     strategy:
@@ -187,6 +189,8 @@ jobs:
         run: |
           cd py-shiny
           . .venv/bin/activate
+          # temporary pin to get CI green
+          uv pip install "flake8-bugbear<25.10.21"
           make narwhals-install-shiny
       - name: install-narwhals-dev
         run: |
@@ -598,3 +602,43 @@ jobs:
         run: |
           cd gt-extras
           uv run --no-sync pytest
+
+  fairlearn:
+    strategy:
+      matrix:
+        python-version: ["3.12"]
+        os: [ubuntu-latest]
+
+    runs-on: ${{ matrix.os }}
+    steps:
+      - uses: actions/checkout@v5
+      - uses: actions/setup-python@v6
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install uv
+        uses: astral-sh/setup-uv@v6
+        with:
+          enable-cache: "true"
+          cache-suffix: fairlearn-${{ matrix.python-version }}
+          cache-dependency-glob: "pyproject.toml"
+      - name: clone-fairlearn
+        run: |
+          git clone https://github.com/fairlearn/fairlearn.git --depth 1
+          cd fairlearn
+          git log
+      - name: install-deps
+        run: |
+          cd fairlearn
+          uv pip install -e . -r requirements.txt matplotlib polars pyarrow pytest typing-extensions --system
+      - name: install-narwhals-dev
+        run: |
+          cd fairlearn
+          uv pip uninstall narwhals --system
+          uv pip install -e ./.. --system
+          uv pip freeze
+      - name: run-pytest
+        run: |
+          cd fairlearn
+          # TODO(FBruzzesi): I hope this will be simplified once there is a decision on
+          # https://github.com/fairlearn/fairlearn/issues/1555
+          pytest test/unit/preprocessing test/unit/metrics

.github/workflows/pytest-pyspark.yml

@@ -74,7 +74,8 @@ jobs:
           java-version: 17
 
       - name: install-reqs
-        run: uv pip install -e . --group core-tests --group extra --system
+        # temporary pin to get CI green
+        run: uv pip install -e . --group core-tests --group extra "pyarrow<22.0" --system
       - name: install pyspark
         run: echo "setuptools<78" | uv pip install -e . "pyspark[connect]==${SPARK_VERSION}" --system
       - name: show-deps

.github/workflows/pytest.yml

@@ -33,6 +33,8 @@ jobs:
         run: uv pip freeze
       - name: Run pytest
         run: pytest tests --cov=narwhals --cov=tests --cov-fail-under=75 --constructors=pandas,pyarrow,polars[eager],polars[lazy]
+      - name: install-test-plugin
+        run: uv pip install -e tests/test_plugin --system  
 
   pytest-windows:
     strategy:
@@ -54,6 +56,8 @@ jobs:
       - name: install-reqs
         # we are not testing pyspark, modin, or dask on Windows here because nobody got time for that
         run: uv pip install -e ".[ibis]" --group core-tests --group extra --system
+      - name: install-test-plugin
+        run: uv pip install -e tests/test_plugin --system
       - name: show-deps
         run: uv pip freeze
       - name: Run pytest
@@ -84,14 +88,16 @@ jobs:
           cache-dependency-glob: "pyproject.toml"
       - name: install-reqs
         run: uv pip install -e ".[dask, modin, ibis]" --group core-tests --group extra --system
+      - name: install-test-plugin
+        run: uv pip install -e tests/test_plugin --system
       - name: show-deps
         run: uv pip freeze
       - name: Run pytest
         run: pytest tests --cov=narwhals --cov=tests --cov-fail-under=100 --runslow --constructors=pandas,pandas[nullable],pandas[pyarrow],pyarrow,modin[pyarrow],polars[eager],polars[lazy],dask,duckdb,sqlframe,ibis --durations=30
       - name: Run doctests
         # reprs differ between versions, so we only run doctests on the latest Python
         if: matrix.python-version == '3.13'
-        run: pytest narwhals/*.py --doctest-modules
+        run: pytest narwhals --doctest-modules
 
   # Test against smaller dependency set, used e.g. on Gentoo.
   pytest-narrower-dependencies:
@@ -112,17 +118,24 @@ jobs:
           cache-suffix: pytest-narrower-deps-${{ matrix.python-version }}
           cache-dependency-glob: "pyproject.toml"
       - name: install-reqs
-        run: uv pip install -e ".[pandas]" --group tests --system
-      - name: show-deps
-        run: uv pip freeze
-      - name: Run pytest
+        run: |
+          uv pip install -e ".[pandas]" --group tests --system
+          uv pip freeze
+      - name: Run pytest (pandas and pandas[nullable])
         run: pytest tests --runslow --constructors=pandas,pandas[nullable]
       - name: install-more-reqs
-        run: uv pip install -U pyarrow --system
-      - name: show-deps
-        run: uv pip freeze
-      - name: Run pytest
+        run: |
+          uv pip install -U pyarrow --system
+          uv pip freeze
+      - name: Run pytest (pandas[pyarrow] and pyarrow)
         run: pytest tests --runslow --constructors=pandas[pyarrow],pyarrow
+      - name: install-polars
+        run: |
+          uv pip uninstall pandas pyarrow --system
+          uv pip install polars --system
+          uv pip freeze
+      - name: Run pytest (polars)
+        run: pytest tests --runslow --constructors=polars[eager],polars[lazy]
 
   python-314:
     strategy:
@@ -141,13 +154,9 @@ jobs:
           enable-cache: "true"
           cache-suffix: python-314-${{ matrix.python-version }}
           cache-dependency-glob: "pyproject.toml"
-      - name: install pyarrow nightly
-        run: |
-          uv pip uninstall pyarrow --system
-          uv pip install -i https://pypi.anaconda.org/scientific-python-nightly-wheels/simple pyarrow -U --system
       - name: install-reqs
         # Use `--pre` as duckdb stable not compatible with 3.14
-        run: uv pip install -e . --group tests --pre pandas polars duckdb sqlframe  --system
+        run: uv pip install -e . --group tests --pre pandas polars pyarrow duckdb sqlframe  --system
       - name: show-deps
         run: uv pip freeze
       - name: Run pytest
@@ -172,12 +181,8 @@ jobs:
           enable-cache: "true"
           cache-suffix: python-314t-${{ matrix.python-version }}
           cache-dependency-glob: "pyproject.toml"
-      - name: install pyarrow nightly
-        run: |
-          uv pip uninstall pyarrow --system
-          uv pip install -i https://pypi.anaconda.org/scientific-python-nightly-wheels/simple pyarrow -U --system
       - name: install-reqs
-        run: uv pip install -e . --group tests --pre pandas --system
+        run: uv pip install -e . --group tests --pre pandas pyarrow --system
       - name: show-deps
         run: uv pip freeze
       - name: Run pytest

.github/workflows/random_ci_pytest.yml

@@ -28,7 +28,7 @@ jobs:
           cache-dependency-glob: "pyproject.toml"
       - name: generate-random-versions
         run: python utils/generate_random_versions.py
-      - name: install-random-verions
+      - name: install-random-versions
         run: uv pip install -r random-requirements.txt --system
       - name: install-narwhals
         run: uv pip install -e . --group tests --system

.pre-commit-config.yaml

@@ -3,7 +3,7 @@ ci:
 repos:
 - repo: https://github.com/astral-sh/ruff-pre-commit
   # Ruff version.
-  rev: 'v0.14.0'
+  rev: 'v0.14.2'
   hooks:
     # Run the formatter.
     - id: ruff-format
@@ -21,6 +21,10 @@ repos:
       files: \.(py|rst|md)$
       args: [--ignore-words-list=ser, --ignore-words-list=RightT]
       exclude: ^docs/api-completeness.md$
+- repo: https://github.com/crate-ci/typos
+  rev: 'v1.39.0'
+  hooks:
+    - id: typos
 -   repo: https://github.com/pycqa/flake8
     rev: '7.3.0'
     hooks:
@@ -121,7 +125,11 @@ repos:
   rev: v6.0.0
   hooks:
   -   id: name-tests-test
-      exclude: ^(tests/utils\.py|tests/plan/utils\.py)
+      exclude: |
+        (?x)
+          ^(tests/utils\.py)
+          |^(tests/test_plugin/)
+          |^(tests/plan/utils\.py)
   -   id: no-commit-to-branch
   -   id: end-of-file-fixer
       exclude: .svg$

CONTRIBUTING.md

@@ -185,9 +185,16 @@ If you add code that should be tested, please add tests.
   - Never assume that your data is ordered in any pre-defined way.
   - Never materialise your data (only exception: `collect`).
   - Avoid calling the schema / column names unnecessarily.
-  - For DuckDB, use the Python API as much as possible, only falling
+
+- DuckDB:
+
+  In addition to the above:
+
+  - Use the Python API as much as possible, only falling
     back to SQL as the last resort for operations not yet supported
     in their Python API (e.g. `over`).
+  - Use standard SQL constructs where possible instead of non-standard
+    ones such as `GROUP BY ALL` or `EXCLUDE`.
 
 ### Test Failure Patterns
 

README.md

@@ -171,6 +171,7 @@ Join the party!
 - [altair](https://github.com/vega/altair/)
 - [bokeh](https://github.com/bokeh/bokeh)
 - [darts](https://github.com/unit8co/darts)
+- [fairlearn](https://github.com/fairlearn/fairlearn)
 - [formulaic](https://github.com/matthewwardrop/formulaic)
 - [gt-extras](https://github.com/posit-dev/gt-extras)
 - [hierarchicalforecast](https://github.com/Nixtla/hierarchicalforecast)

_typos.toml

@@ -0,0 +1,10 @@
+[default.extend-words]
+ser = "ser"  # Used in Series classes
+ND = "ND"  # Used in NativeNDFrameT
+ba = "ba"  # Used as column name in docstring examples (way too much?)
+iy = "iy"  # Used as column name (once in a test)
+pn = "pn"  # Used in docs: pn = PandasLikeNamespace(...)
+TYP = "TYP"  # Used in flake8 rule
+
+[files]
+extend-exclude = ["tests/data/*"]

docs/ecosystem.md

@@ -8,6 +8,7 @@ for their dataframe interoperability needs:
 * [altair](https://github.com/vega/altair/)
 * [bokeh](https://github.com/bokeh/bokeh)
 * [darts](https://github.com/unit8co/darts)
+* [fairlearn](https://github.com/fairlearn/fairlearn)
 * [formulaic](https://github.com/matthewwardrop/formulaic)
 * [gt-extras](https://github.com/posit-dev/gt-extras)
 * [hierarchicalforecast](https://github.com/Nixtla/hierarchicalforecast)