(Python) Refactor Series constructor (#1032)

- Moved Series construction logic to construction.py.
- Added some type conversion utils to datatypes.py for use in Series construction
- Deprecated Series.from_arrow. Usage of pl.from_arrow is encouraged, which calls Series._from_arrow.

py-polars/polars/convert.py

@@ -133,7 +133,7 @@ def from_arrow(
     if isinstance(a, pa.Table):
         return pl.DataFrame.from_arrow(a, rechunk)
     elif isinstance(a, pa.Array):
-        return pl.Series.from_arrow("", a)
+        return pl.Series._from_arrow("", a)
     else:
         raise ValueError(f"expected arrow table / array, got {a}")
 

py-polars/polars/datatypes.py

@@ -1,9 +1,18 @@
 import ctypes
 import typing as tp
-from typing import Any, Dict, Type
+from typing import Any, Callable, Dict, Sequence, Type
 
+import numpy as np
+import pyarrow as pa
 from _ctypes import _SimpleCData
 
+try:
+    from polars.polars import PySeries
+except ImportError:
+    import warnings
+
+    warnings.warn("binary files missing")
+
 __pdoc__ = {
     "dtype_to_ctype": False,
     "dtype_to_int": False,
@@ -271,3 +280,96 @@ def pytype_to_polars_type(data_type: Type[Any]) -> Type[DataType]:
     else:
         polars_type = data_type
     return polars_type
+
+
+_POLARS_TYPE_TO_CONSTRUCTOR = {
+    Float32: PySeries.new_opt_f32,
+    Float64: PySeries.new_opt_f64,
+    Int8: PySeries.new_opt_i8,
+    Int16: PySeries.new_opt_i16,
+    Int32: PySeries.new_opt_i32,
+    Int64: PySeries.new_opt_i64,
+    UInt8: PySeries.new_opt_u8,
+    UInt16: PySeries.new_opt_u16,
+    UInt32: PySeries.new_opt_u32,
+    UInt64: PySeries.new_opt_u64,
+    Date32: PySeries.new_opt_i32,
+    Date64: PySeries.new_opt_i32,
+    Boolean: PySeries.new_opt_bool,
+    Utf8: PySeries.new_str,
+    Object: PySeries.new_object,
+}
+
+
+def polars_type_to_constructor(
+    dtype: Type[DataType],
+) -> Callable[[str, Sequence[Any]], "PySeries"]:
+    """
+    Get the right PySeries constructor for the given Polars dtype.
+    """
+    try:
+        return _POLARS_TYPE_TO_CONSTRUCTOR[dtype]
+    except KeyError:
+        raise ValueError(f"Cannot construct PySeries for type {dtype}.")
+
+
+_NUMPY_TYPE_TO_CONSTRUCTOR = {
+    np.float32: PySeries.new_f32,
+    np.float64: PySeries.new_f64,
+    np.int8: PySeries.new_i8,
+    np.int16: PySeries.new_i16,
+    np.int32: PySeries.new_i32,
+    np.int64: PySeries.new_i64,
+    np.uint8: PySeries.new_u8,
+    np.uint16: PySeries.new_u16,
+    np.uint32: PySeries.new_u32,
+    np.uint64: PySeries.new_u64,
+    np.str_: PySeries.new_str,
+    bool: PySeries.new_bool,
+}
+
+
+def numpy_type_to_constructor(dtype: Type[np.dtype]) -> Callable[..., "PySeries"]:
+    """
+    Get the right PySeries constructor for the given Polars dtype.
+    """
+    try:
+        return _NUMPY_TYPE_TO_CONSTRUCTOR[dtype]
+    except KeyError:
+        return PySeries.new_object
+
+
+_PY_TYPE_TO_CONSTRUCTOR = {
+    float: PySeries.new_opt_f64,
+    int: PySeries.new_opt_i64,
+    str: PySeries.new_str,
+    bool: PySeries.new_opt_bool,
+}
+
+
+def py_type_to_constructor(dtype: Type[Any]) -> Callable[..., "PySeries"]:
+    """
+    Get the right PySeries constructor for the given Python dtype.
+    """
+    try:
+        return _PY_TYPE_TO_CONSTRUCTOR[dtype]
+    except KeyError:
+        return PySeries.new_object
+
+
+_PY_TYPE_TO_ARROW_TYPE = {
+    float: pa.float64(),
+    int: pa.int64(),
+    str: pa.large_utf8(),
+    bool: pa.bool_(),
+}
+
+
+def py_type_to_arrow_type(dtype: Type[Any]) -> pa.lib.DataType:
+    """
+    Convert a Python dtype to an Arrow dtype.
+    """
+    try:
+        return _PY_TYPE_TO_ARROW_TYPE[dtype]
+    except KeyError:
+        raise ValueError(f"Cannot parse dtype {dtype} into arrow dtype.")

py-polars/polars/eager/frame.py

@@ -262,7 +262,7 @@ def _from_dict(
         -------
         DataFrame
         """
-        return cls(data, columns=columns, nullable=nullable)
+        return cls._from_pydf(dict_to_pydf(data, columns=columns, nullable=nullable))
 
     @classmethod
     def _from_records(
@@ -294,7 +294,15 @@ def _from_records(
         -------
         DataFrame
         """
-        return cls(data, columns=columns, orient=orient, nullable=nullable)
+        if isinstance(data, np.ndarray):
+            pydf = numpy_to_pydf(
+                data, columns=columns, orient=orient, nullable=nullable
+            )
+        else:
+            pydf = sequence_to_pydf(
+                data, columns=columns, orient=orient, nullable=nullable
+            )
+        return cls._from_pydf(pydf)
 
     @classmethod
     def from_arrow(cls, table: pa.Table, rechunk: bool = True) -> "DataFrame":
@@ -373,7 +381,7 @@ def _from_pandas(
         ╰─────┴─────┴─────╯
         ```
         """
-        return cls(data, columns=columns, nullable=nullable)
+        return cls._from_pydf(pandas_to_pydf(data, columns=columns, nullable=nullable))
 
     @classmethod
     def from_rows(