Skip to content

ENH: Add support for dtype string aliases to Series#astype #556

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Mar 1, 2023
Merged

ENH: Add support for dtype string aliases to Series#astype #556

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
4a421e8
Add support for nullable integer data types to Series#astype
skatsuta Mar 1, 2023
7470d27
Add support for nullable float data types to Series#astype
skatsuta Mar 1, 2023
34ed72b
Add support for nullable boolean data type to Series#astype
skatsuta Mar 1, 2023
25f99ee
Add support for nullable string data type to Series#astype
skatsuta Mar 1, 2023
ad85bfa
Refactor dtype arg type aliases and add missing dtype aliases
skatsuta Mar 1, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
54 changes: 43 additions & 11 deletions pandas-stubs/_typing.pyi
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,41 +78,73 @@ NpDtype: TypeAlias = str | np.dtype[np.generic] | type[str | complex | bool | ob
Dtype: TypeAlias = ExtensionDtype | NpDtype
DtypeArg: TypeAlias = Dtype | dict[Any, Dtype]
BooleanDtypeArg: TypeAlias = (
type[bool] | type[np.bool_] | pd.BooleanDtype | Literal["bool"]
# Builtin bool type and its string alias
type[bool] # noqa: Y030
| Literal["bool"]
# Pandas nullable boolean type and its string alias
| pd.BooleanDtype
| Literal["boolean"]
# Numpy bool type
| type[np.bool_]
)
IntDtypeArg: TypeAlias = (
Literal["int", "int32"]
| type[int]
# Builtin integer type and its string alias
type[int] # noqa: Y030
| Literal["int"]
# Pandas nullable integer types and their string aliases
| pd.Int8Dtype
| pd.Int16Dtype
| pd.Int32Dtype
| pd.Int64Dtype
| Literal["Int8", "Int16", "Int32", "Int64"]
# Numpy signed integer types and their string aliases
| type[np.byte]
| type[np.int8]
| type[np.int16]
| type[np.int32]
| type[np.int64]
| type[np.intp]
| Literal["byte", "int8", "int16", "int32", "int64", "intp"]
# Numpy unsigned integer types and their string aliases
| type[np.ubyte]
| type[np.uint8]
| type[np.uint16]
| type[np.uint32]
| type[np.uint64]
| type[np.intp]
| type[np.uintp]
| type[np.byte]
| type[np.ubyte]
| Literal["ubyte", "uint8", "uint16", "uint32", "uint64", "uintp"]
)
StrDtypeArg: TypeAlias = (
# Builtin str type and its string alias
type[str] # noqa: Y030
| Literal["str"]
# Pandas nullable string type and its string alias
| pd.StringDtype
| Literal["string"]
)
StrDtypeArg: TypeAlias = type[str] | pd.StringDtype | Literal["str"]
BytesDtypeArg: TypeAlias = type[bytes]
FloatDtypeArg: TypeAlias = (
pd.Float32Dtype
# Builtin float type and its string alias
type[float] # noqa: Y030
| Literal["float"]
# Pandas nullable float types and their string aliases
| pd.Float32Dtype
| pd.Float64Dtype
| Literal["Float32", "Float64"]
# Numpy float types and their string aliases
| type[np.float16]
| type[np.float32]
| type[np.float64]
| type[float]
| Literal["float"]
| Literal["float16", "float32", "float64"]
)
ComplexDtypeArg: TypeAlias = (
type[np.complex64] | type[np.complex128] | type[complex] | Literal["complex"]
# Builtin complex type and its string alias
type[complex] # noqa: Y030
| Literal["complex"]
# Numpy complex types and their aliases
| type[np.complex64]
| type[np.complex128]
| Literal["complex64", "complex128"]
)
TimedeltaDtypeArg: TypeAlias = Literal["timedelta64[ns]"]
TimestampDtypeArg: TypeAlias = Literal["datetime64[ns]"]
Expand Down
88 changes: 73 additions & 15 deletions tests/test_series.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1449,32 +1449,81 @@ def test_updated_astype() -> None:
s = pd.Series([3, 4, 5])
s1 = pd.Series(True)

# Boolean types

# Builtin bool types
check(assert_type(s.astype(bool), "pd.Series[bool]"), pd.Series, np.bool_)
check(assert_type(s.astype("bool"), "pd.Series[bool]"), pd.Series, np.bool_)
# Pandas nullable boolean types
check(
assert_type(s1.astype(pd.BooleanDtype()), "pd.Series[bool]"),
pd.Series,
np.bool_,
)
check(assert_type(s1.astype("boolean"), "pd.Series[bool]"), pd.Series, np.bool_)
# Numpy bool type
check(assert_type(s.astype(np.bool_), "pd.Series[bool]"), pd.Series, np.bool_)

# Integer types

# Builtin integer types
check(assert_type(s.astype(int), "pd.Series[int]"), pd.Series, np.integer)
check(assert_type(s.astype("int"), "pd.Series[int]"), pd.Series, np.integer)
check(assert_type(s.astype("int32"), "pd.Series[int]"), pd.Series, np.int32)
# Pandas nullable integer types
check(assert_type(s.astype(pd.Int8Dtype()), "pd.Series[int]"), pd.Series, np.int8)
check(assert_type(s.astype(pd.Int16Dtype()), "pd.Series[int]"), pd.Series, np.int16)
check(assert_type(s.astype(pd.Int32Dtype()), "pd.Series[int]"), pd.Series, np.int32)
check(assert_type(s.astype(pd.Int64Dtype()), "pd.Series[int]"), pd.Series, np.int64)
check(assert_type(s.astype("Int8"), "pd.Series[int]"), pd.Series, np.int8)
check(assert_type(s.astype("Int16"), "pd.Series[int]"), pd.Series, np.int16)
check(assert_type(s.astype("Int32"), "pd.Series[int]"), pd.Series, np.int32)
check(assert_type(s.astype("Int64"), "pd.Series[int]"), pd.Series, np.int64)
# Numpy signed integer types
check(assert_type(s.astype(np.byte), "pd.Series[int]"), pd.Series, np.byte)
check(assert_type(s.astype(np.int8), "pd.Series[int]"), pd.Series, np.int8)
check(assert_type(s.astype(np.int16), "pd.Series[int]"), pd.Series, np.int16)
check(assert_type(s.astype(np.int32), "pd.Series[int]"), pd.Series, np.int32)
check(assert_type(s.astype(np.int64), "pd.Series[int]"), pd.Series, np.int64)
check(assert_type(s.astype(np.intp), "pd.Series[int]"), pd.Series, np.intp)
check(assert_type(s.astype("byte"), "pd.Series[int]"), pd.Series, np.byte)
check(assert_type(s.astype("int8"), "pd.Series[int]"), pd.Series, np.int8)
check(assert_type(s.astype("int16"), "pd.Series[int]"), pd.Series, np.int16)
check(assert_type(s.astype("int32"), "pd.Series[int]"), pd.Series, np.int32)
check(assert_type(s.astype("int64"), "pd.Series[int]"), pd.Series, np.int64)
check(assert_type(s.astype("intp"), "pd.Series[int]"), pd.Series, np.intp)
# Numpy unsigned integer types
check(assert_type(s.astype(np.ubyte), "pd.Series[int]"), pd.Series, np.ubyte)
check(assert_type(s.astype(np.uint8), "pd.Series[int]"), pd.Series, np.uint8)
check(assert_type(s.astype(np.uint16), "pd.Series[int]"), pd.Series, np.uint16)
check(assert_type(s.astype(np.uint32), "pd.Series[int]"), pd.Series, np.uint32)
check(assert_type(s.astype(np.uint64), "pd.Series[int]"), pd.Series, np.uint64)
check(assert_type(s.astype(np.intp), "pd.Series[int]"), pd.Series, np.intp)
check(assert_type(s.astype(np.uintp), "pd.Series[int]"), pd.Series, np.uintp)
check(assert_type(s.astype(np.byte), "pd.Series[int]"), pd.Series, np.byte)
check(assert_type(s.astype(np.ubyte), "pd.Series[int]"), pd.Series, np.ubyte)
check(assert_type(s.astype("ubyte"), "pd.Series[int]"), pd.Series, np.ubyte)
check(assert_type(s.astype("uint8"), "pd.Series[int]"), pd.Series, np.uint8)
check(assert_type(s.astype("uint16"), "pd.Series[int]"), pd.Series, np.uint16)
check(assert_type(s.astype("uint32"), "pd.Series[int]"), pd.Series, np.uint32)
check(assert_type(s.astype("uint64"), "pd.Series[int]"), pd.Series, np.uint64)
check(assert_type(s.astype("uintp"), "pd.Series[int]"), pd.Series, np.uintp)

# String types

# Builtin str types
check(assert_type(s.astype(str), "pd.Series[str]"), pd.Series, str)
check(assert_type(s.astype(pd.StringDtype()), "pd.Series[str]"), pd.Series, str)
check(assert_type(s.astype("str"), "pd.Series[str]"), pd.Series, str)
# Pandas nullable string types
check(assert_type(s.astype(pd.StringDtype()), "pd.Series[str]"), pd.Series, str)
check(assert_type(s.astype("string"), "pd.Series[str]"), pd.Series, str)

# Bytes types

check(assert_type(s.astype(bytes), "pd.Series[bytes]"), pd.Series, bytes)

# Float types

# Builtin float types
check(assert_type(s.astype(float), "pd.Series[float]"), pd.Series, float)
check(assert_type(s.astype("float"), "pd.Series[float]"), pd.Series, float)
# Pandas nullable float types
check(
assert_type(s.astype(pd.Float32Dtype()), "pd.Series[float]"),
pd.Series,
Expand All @@ -1485,12 +1534,22 @@ def test_updated_astype() -> None:
pd.Series,
np.float64,
)
check(assert_type(s.astype("Float32"), "pd.Series[float]"), pd.Series, np.float32)
check(assert_type(s.astype("Float64"), "pd.Series[float]"), pd.Series, np.float64)
# Numpy float types
check(assert_type(s.astype(np.float16), "pd.Series[float]"), pd.Series, np.float16)
check(assert_type(s.astype(np.float32), "pd.Series[float]"), pd.Series, np.float32)
check(assert_type(s.astype(np.float64), "pd.Series[float]"), pd.Series, np.float64)
check(assert_type(s.astype(float), "pd.Series[float]"), pd.Series, float)
check(assert_type(s.astype("float"), "pd.Series[float]"), pd.Series, float)
check(assert_type(s.astype("float16"), "pd.Series[float]"), pd.Series, np.float16)
check(assert_type(s.astype("float32"), "pd.Series[float]"), pd.Series, np.float32)
check(assert_type(s.astype("float64"), "pd.Series[float]"), pd.Series, np.float64)

# Complex types

# Builtin complex types
check(assert_type(s.astype(complex), "pd.Series[complex]"), pd.Series, complex)
check(assert_type(s.astype("complex"), "pd.Series[complex]"), pd.Series, complex)
# Numpy complex types
check(
assert_type(s.astype(np.complex64), "pd.Series[complex]"),
pd.Series,
Expand All @@ -1501,17 +1560,16 @@ def test_updated_astype() -> None:
pd.Series,
np.complex128,
)
check(assert_type(s.astype(complex), "pd.Series[complex]"), pd.Series, complex)
check(assert_type(s.astype("complex"), "pd.Series[complex]"), pd.Series, complex)

check(
assert_type(s1.astype(pd.BooleanDtype()), "pd.Series[bool]"),
assert_type(s.astype("complex64"), "pd.Series[complex]"),
pd.Series,
np.bool_,
np.complex64,
)
check(
assert_type(s.astype("complex128"), "pd.Series[complex]"),
pd.Series,
np.complex128,
)
check(assert_type(s.astype("bool"), "pd.Series[bool]"), pd.Series, np.bool_)
check(assert_type(s.astype(bool), "pd.Series[bool]"), pd.Series, np.bool_)
check(assert_type(s.astype(np.bool_), "pd.Series[bool]"), pd.Series, np.bool_)

check(
assert_type(s.astype("timedelta64[ns]"), TimedeltaSeries),
Expand Down