Add pyupgrade

.github/workflows/test-python.yaml

@@ -31,7 +31,13 @@ jobs:
  pip install -r py-polars/build.requirements.txt
  - name: Run formatting checks
  run: |
- cd py-polars && black --check . && blackdoc --check . && isort --check . && rustup override set nightly-2022-07-24 && cargo fmt --all -- --check && cd ..
+ cd py-polars
+ black --check .
+ blackdoc --check .
+ isort --check .
+ pyupgrade --py37-plus `find . -name "*.py" -type f`
+ rustup override set nightly-2022-07-24 && cargo fmt --all -- --check
+ cd ..
  - name: Run linting
  run: |
  cd py-polars && flake8 && cd ..
@@ -76,7 +82,13 @@ jobs:
  pip install -r py-polars/build.requirements.txt
  - name: Run formatting checks
  run: |
- cd py-polars && black --check . && blackdoc --check . && isort --check . && rustup override set nightly-2022-07-24 && cargo fmt --all -- --check && cd ..
+ cd py-polars
+ black --check .
+ blackdoc --check .
+ isort --check .
+ pyupgrade --py37-plus `find . -name "*.py" -type f`
+ rustup override set nightly-2022-07-24 && cargo fmt --all -- --check
+ cd ..
  - name: Run linting
  run: |
  cd py-polars && flake8 && cd ..

py-polars/Makefile

@@ -18,6 +18,7 @@ pre-commit: venv
  $(PYTHON_BIN)/isort .
  $(PYTHON_BIN)/black .
  $(PYTHON_BIN)/blackdoc .
+ $(PYTHON_BIN)/pyupgrade --py37-plus `find polars/ tests/ -name "*.py" -type f`
  $(PYTHON_BIN)/mypy
  $(PYTHON_BIN)/flake8
  make -C .. fmt_toml

py-polars/build.requirements.txt

@@ -16,14 +16,15 @@ pytest-cov==3.0.0
 hypothesis==6.49.1
 black==22.6.0
 blackdoc==0.3.5
-isort~=5.10.1
+isort==5.10.1
 mypy==0.971
 flake8==5.0.2
 flake8-bugbear==22.7.1
 flake8-comprehensions==3.10.0
 flake8-docstrings==1.6.0
 flake8-simplify==0.19.3
 flake8-tidy-imports==4.8.0
+pyupgrade==2.37.3
 ghp-import==2.1.0
 sphinx==4.3.2
 pydata-sphinx-theme==0.9.0

py-polars/polars/datatypes.py

@@ -23,7 +23,7 @@
  _DOCUMENTING = True
 
 
-def get_idx_type() -> Type[DataType]:
+def get_idx_type() -> type[DataType]:
  """
  Get the datatype used for polars Indexing
 

py-polars/polars/internals/expr.py

@@ -4,7 +4,7 @@
 import math
 import random
 from datetime import date, datetime, timedelta
-from typing import TYPE_CHECKING, Any, Callable, List, Sequence
+from typing import TYPE_CHECKING, Any, Callable, Sequence
 
 from polars import internals as pli
 from polars.datatypes import (
@@ -385,7 +385,7 @@ def exclude(
  self,
  columns: (
  str
- | List[str]
+ | list[str]
  | DataType
  | type[DataType]
  | DataType
@@ -1696,8 +1696,8 @@ def arg_min(self) -> Expr:
 
  def sort_by(
  self,
- by: Expr | str | List[Expr | str],
- reverse: bool | List[bool] = False,
+ by: Expr | str | list[Expr | str],
+ reverse: bool | list[bool] = False,
  ) -> Expr:
  """
  Sort this column by the ordering of another column, or multiple other columns.
@@ -1756,7 +1756,7 @@ def sort_by(
 
  return wrap_expr(self._pyexpr.sort_by(by, reverse))
 
- def take(self, index: List[int] | Expr | pli.Series | np.ndarray[Any, Any]) -> Expr:
+ def take(self, index: list[int] | Expr | pli.Series | np.ndarray[Any, Any]) -> Expr:
  """
  Take values by index.
 
@@ -2243,7 +2243,7 @@ def product(self) -> Expr:
  """
  return wrap_expr(self._pyexpr.product())
 
- def n_unique(self) -> "Expr":
+ def n_unique(self) -> Expr:
  """
  Count unique values.
 
@@ -2263,7 +2263,7 @@ def n_unique(self) -> "Expr":
  """
  return wrap_expr(self._pyexpr.n_unique())
 
- def null_count(self) -> "Expr":
+ def null_count(self) -> Expr:
  """
  Count null values.
 
@@ -2398,32 +2398,7 @@ def last(self) -> Expr:
  """
  return wrap_expr(self._pyexpr.last())
 
- def list(self) -> Expr:
- """
- Aggregate to list.
-
- Examples
- --------
- >>> df = pl.DataFrame(
- ... {
- ... "a": [1, 2, 3],
- ... "b": [4, 5, 6],
- ... }
- ... )
- >>> df.select(pl.all().list())
- shape: (1, 2)
- ┌───────────┬───────────┐
- │ a ┆ b │
- │ --- ┆ --- │
- │ list[i64] ┆ list[i64] │
- ╞═══════════╪═══════════╡
- │ [1, 2, 3] ┆ [4, 5, 6] │
- └───────────┴───────────┘
-
- """
- return wrap_expr(self._pyexpr.list())
-
- def over(self, expr: str | Expr | List[Expr | str]) -> Expr:
+ def over(self, expr: str | Expr | list[Expr | str]) -> Expr:
  """
  Apply window function over a subgroup.
 
@@ -3250,7 +3225,7 @@ def interpolate(self) -> Expr:
  def rolling_min(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3350,7 +3325,7 @@ def rolling_min(
  def rolling_max(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3449,7 +3424,7 @@ def rolling_max(
  def rolling_mean(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3546,7 +3521,7 @@ def rolling_mean(
  def rolling_sum(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3645,7 +3620,7 @@ def rolling_sum(
  def rolling_std(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3715,7 +3690,7 @@ def rolling_std(
  def rolling_var(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3785,7 +3760,7 @@ def rolling_var(
  def rolling_median(
  self,
  window_size: int | str,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3853,7 +3828,7 @@ def rolling_quantile(
  quantile: float,
  interpolation: InterpolationMethod = "nearest",
  window_size: int | str = 2,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  by: str | None = None,
@@ -3931,7 +3906,7 @@ def rolling_apply(
  self,
  function: Callable[[pli.Series], Any],
  window_size: int,
- weights: List[float] | None = None,
+ weights: list[float] | None = None,
  min_periods: int | None = None,
  center: bool = False,
  ) -> Expr:
@@ -5052,19 +5027,44 @@ def set_sorted(self, reverse: bool = False) -> Expr:
  """
  return self.map(lambda s: s.set_sorted(reverse))
 
- # Below are the namespaces defined. Keep these at the end of the definition of Expr,
- # as to not confuse mypy with the type annotation `str` with the namespace "str"
+ # Keep the `list` and `str` methods below at the end of the definition of Expr,
+ # as to not confuse mypy with the type annotation `str` and `list`
 
- @property
- def dt(self) -> ExprDateTimeNameSpace:
- """Create an object namespace of all datetime related methods."""
- return ExprDateTimeNameSpace(self)
+ def list(self) -> Expr:
+ """
+ Aggregate to list.
+
+ Examples
+ --------
+ >>> df = pl.DataFrame(
+ ... {
+ ... "a": [1, 2, 3],
+ ... "b": [4, 5, 6],
+ ... }
+ ... )
+ >>> df.select(pl.all().list())
+ shape: (1, 2)
+ ┌───────────┬───────────┐
+ │ a ┆ b │
+ │ --- ┆ --- │
+ │ list[i64] ┆ list[i64] │
+ ╞═══════════╪═══════════╡
+ │ [1, 2, 3] ┆ [4, 5, 6] │
+ └───────────┴───────────┘
+
+ """
+ return wrap_expr(self._pyexpr.list())
 
  @property
  def str(self) -> ExprStringNameSpace:
  """Create an object namespace of all string related methods."""
  return ExprStringNameSpace(self)
 
+ @property
+ def dt(self) -> ExprDateTimeNameSpace:
+ """Create an object namespace of all datetime related methods."""
+ return ExprDateTimeNameSpace(self)
+
  @property
  def arr(self) -> ExprListNameSpace:
  """Create an object namespace of all list related methods."""
@@ -5125,7 +5125,7 @@ def field(self, name: str) -> Expr:
  """
  return wrap_expr(self._pyexpr.struct_field_by_name(name))
 
- def rename_fields(self, names: List[str]) -> Expr:
+ def rename_fields(self, names: list[str]) -> Expr:
  """
  Rename the fields of the struct
 
@@ -5273,7 +5273,7 @@ def unique(self) -> Expr:
  return wrap_expr(self._pyexpr.lst_unique())
 
  def concat(
- self, other: List[Expr | str] | Expr | str | pli.Series | List[Any]
+ self, other: list[Expr | str] | Expr | str | pli.Series | list[Any]
  ) -> Expr:
  """
  Concat the arrays in a Series dtype List in linear time.
@@ -5309,7 +5309,7 @@ def concat(
  ):
  return self.concat(pli.Series([other]))
 
- other_list: List[Expr | str | pli.Series]
+ other_list: list[Expr | str | pli.Series]
  if not isinstance(other, list):
  other_list = [other]
  else:

py-polars/polars/internals/functions.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from datetime import date, datetime, timedelta
-from typing import TYPE_CHECKING, Optional, Sequence, overload
+from typing import TYPE_CHECKING, Sequence, overload
 
 from polars import internals as pli
 from polars.datatypes import Categorical, Date, Float64
@@ -273,7 +273,7 @@ def date_range(
 def cut(
  s: pli.Series,
  bins: list[float],
- labels: Optional[list[str]] = None,
+ labels: list[str] | None = None,
  break_point_label: str = "break_point",
  category_label: str = "category",
 ) -> pli.DataFrame:

py-polars/polars/internals/series.py

@@ -2557,7 +2557,7 @@ def set_at_idx(
  self._s = f(idx_array, value)
  return self
 
- def cleared(self) -> "Series":
+ def cleared(self) -> Series:
  """
  Create an empty copy of the current Series, with identical name/dtype but no
  data.
@@ -2578,7 +2578,7 @@ def cleared(self) -> "Series":
  """
  return self.limit(0) if len(self) > 0 else self.clone()
 
- def clone(self) -> "Series":
+ def clone(self) -> Series:
  """
  Very cheap deepcopy/clone.
 

py-polars/tests/test_df.py

@@ -2,7 +2,6 @@
 
 import sys
 import typing
-from builtins import range
 from datetime import datetime, timedelta
 from io import BytesIO
 from typing import Any, Iterator
@@ -594,7 +593,7 @@ def test_read_missing_file() -> None:
  pl.read_csv("fake_csv_file")
 
  with pytest.raises(FileNotFoundError, match="fake_csv_file"):
- with open("fake_csv_file", "r") as f:
+ with open("fake_csv_file") as f:
  pl.read_csv(f)