Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.8/site-packages/pandas/io/spss.py: 45%

1from __future__ import annotations 

2 

3from pathlib import Path 

4from typing import ( 

5 TYPE_CHECKING, 

6 Sequence, 

7) 

8 

9from pandas._libs import lib 

10from pandas.compat._optional import import_optional_dependency 

11from pandas.util._validators import check_dtype_backend 

12 

13from pandas.core.dtypes.inference import is_list_like 

14 

15from pandas.io.common import stringify_path 

16 

17if TYPE_CHECKING: 

18 from pandas._typing import DtypeBackend 

19 

20 from pandas import DataFrame 

21 

22 

23def read_spss( 

24 path: str | Path, 

25 usecols: Sequence[str] | None = None, 

26 convert_categoricals: bool = True, 

27 dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default, 

28) -> DataFrame: 

29 """ 

30 Load an SPSS file from the file path, returning a DataFrame. 

31 

32 Parameters 

33 ---------- 

34 path : str or Path 

35 File path. 

36 usecols : list-like, optional 

37 Return a subset of the columns. If None, return all columns. 

38 convert_categoricals : bool, default is True 

39 Convert categorical columns into pd.Categorical. 

40 dtype_backend : {"numpy_nullable", "pyarrow"}, defaults to NumPy backed DataFrames 

41 Which dtype_backend to use, e.g. whether a DataFrame should have NumPy 

42 arrays, nullable dtypes are used for all dtypes that have a nullable 

43 implementation when "numpy_nullable" is set, pyarrow is used for all 

44 dtypes if "pyarrow" is set. 

45 

46 The dtype_backends are still experimential. 

47 

48 .. versionadded:: 2.0 

49 

50 Returns 

51 ------- 

52 DataFrame 

53 """ 

54 pyreadstat = import_optional_dependency("pyreadstat") 

55 check_dtype_backend(dtype_backend) 

56 

57 if usecols is not None: 

58 if not is_list_like(usecols): 

59 raise TypeError("usecols must be list-like.") 

60 usecols = list(usecols) # pyreadstat requires a list 

61 

62 df, _ = pyreadstat.read_sav( 

63 stringify_path(path), usecols=usecols, apply_value_formats=convert_categoricals 

64 ) 

65 if dtype_backend is not lib.no_default: 

66 df = df.convert_dtypes(dtype_backend=dtype_backend) 

67 return df