DEVOCEAN

import pandas as pd
dt = pd.Series([1, 8, 6, 7, 9])
print(dt.sort_values())
#>>[1, 6, 7, 8, 9]
print(dt)
#>>[1, 8, 6, 7, 9]

import pandas as pd
dt = pd.Series([1, 8, 6, 7, 9])
print(dt.sort_values(inplace = True))
#>>None
print(dt)
#>>[1, 6, 7, 8, 9]

#https://github.com/pandas-dev/pandas/blob/v2.1.1/pandas/core/series.py#L3545
class Series(base.IndexOpsMixin, NDFrame):  # type: ignore[misc]
    ...
    def sort_values(
        self,
        *,
        axis: Axis = 0,
        ascending: bool | Sequence[bool] = True,
        inplace: bool = False,
        kind: SortKind = "quicksort",
        na_position: NaPosition = "last",
        ignore_index: bool = False,
        key: ValueKeyFunc | None = None,
    ) -> Series | None:
        inplace = validate_bool_kwarg(inplace, "inplace")
        # Validate the axis parameter
        self._get_axis_number(axis)

        # inplace를 실행하기 위해서는 copy본을 꼭 만들어줘야됨
        # 단순하게 view를 통해 접근한 data에는 불가
        if inplace and self._is_cached:
            raise ValueError(
                "This Series is a view of some other array, to "
                "sort in-place you must create a copy"
            )

#https://github.com/pandas-dev/pandas/blob/v2.1.1/pandas/core/series.py#L3725
class Series(base.IndexOpsMixin, NDFrame):  # type: ignore[misc]
        ...        
        else:
            #특정 값 기준으로 sorting하기 위해서
            #sorting을 할 target data를 설정
            values_to_sort = self._values
        #value를 기반으로 fancy indexing을 진행할 fancy index를 생성함
        sorted_index = nargsort(values_to_sort, kind, bool(ascending), na_position)
        
        #비교결과 정렬할 필요가 없을 경우
        if is_range_indexer(sorted_index, len(sorted_index)):
            if inplace:
                #self._update_inplace(self)를 실행하고
                #None을 반환하게 되어있음.
                #self._update_inplace(self)
                #return None
                #과 동일
                return self._update_inplace(self)
            return self.copy(deep=None)

#https://github.com/pandas-dev/pandas/blob/v2.1.1/pandas/core/series.py#L3734
        #중요!
        #정렬이 필요한 경우 정렬 후 value와 index를 fancy indexing한 뒤
        #새로운 object를 생성해줌
        result = self._constructor(
            self._values[sorted_index], index=self.index[sorted_index], copy=False
        )
        #아래 문장과 동일함
        #result = Series(
        #    self._values[sorted_index], index=self.index[sorted_index], copy=False
        #)
        #이때 copy = False를 사용해서 기존 Data를 완전 복제하지 않음
        ...
        
        if not inplace:
            return result.__finalize__(self, method="sort_values")
        self._update_inplace(result)
        return None

import numpy as np
arr = np.arange(100)
arr_copy = arr[0:10]
arr[0] = 100
print(arr[0])
#>>100
print(arr_copy[0])
#>>100

arr = np.arange(100)
arr_copy = arr[[i for i in range(100)]]
arr[0] = 100
print(arr[0])
#>>100
print(arr_copy[0])
#>>0

import pandas as pd
import numpy as np
import psutil
cnt = 100000000
dt = np.arange(cnt)
p = psutil.Process()
rss = p.memory_info().rss/2**20
temp = pd.Series(dt, copy = False)
print(f'mem usage : {p.memory_info().rss/2**20-rss:10.5f}MB')
#copy = False일 경우 메모리가 증가하지 않음
#>>mem usage :    0.06250MB
p = psutil.Process()
rss = p.memory_info().rss/2**20
temp = pd.Series(dt, copy = True)
print(f'mem usage : {p.memory_info().rss/2**20-rss:10.5f}MB')
#copy = True일 경우 메모리가 증가함
#>>mem usage :  381.47266MB

import pandas as pd
import numpy as np

arr = np.arange(10)
tg = pd.Series(arr, copy = False)
print(tg[0])
#>> 0
#Series Object가 아니라 Series Object를 만들때 사용된 Data를 변경
arr[0] = 200
print(tg[0])
#>> 200

#https://github.com/pandas-dev/pandas/blob/v2.1.1/pandas/core/series.py#L3741
    ... 
        if not inplace:
            return result.__finalize__(self, method="sort_values")
        self._update_inplace(result)
        return None

#https://github.com/pandas-dev/pandas/blob/v2.1.1/pandas/core/generic.py#L6148
class NDFrame(PandasObject, indexing.IndexingMixin):
    ...
    def __finalize__(self, other, method: str | None = None, **kwargs) -> Self:        
        if isinstance(other, NDFrame):
            for name in other.attrs:
                self.attrs[name] = other.attrs[name]

            self.flags.allows_duplicate_labels = other.flags.allows_duplicate_labels
            # For subclasses using _metadata.
            for name in set(self._metadata) & set(other._metadata):
                assert isinstance(name, str)
                object.__setattr__(self, name, getattr(other, name, None))

        if method == "concat":
            ...
        return self

#https://github.com/pandas-dev/pandas/blob/e86ed377639948c64c429059127bcf5b359ab6be/pandas/core/generic.py#L238
class NDFrame(PandasObject, indexing.IndexingMixin):
    ...
    @property
    def attrs(self) -> dict[Hashable, Any]:
        """
        Dictionary of global attributes of this dataset.
        .. warning::
           attrs is experimental and may change without warning.
        """
        ...

#https://github.com/pandas-dev/pandas/blob/v2.1.1/pandas/core/generic.py#L4807
class NDFrame(PandasObject, indexing.IndexingMixin):
    ...
    @final
    def _update_inplace(self, result, verify_is_copy: bool_t = True) -> None:        
        self._reset_cache()
        self._clear_item_cache()
        self._mgr = result._mgr
        self._maybe_update_cacher(verify_is_copy=verify_is_copy, inplace=True)

#https://github.com/pandas-dev/pandas/blob/e86ed377639948c64c429059127bcf5b359ab6be/pandas/core/series.py#L5525C1-L5618C26
class Series(base.IndexOpsMixin, NDFrame):  # type: ignore[misc]
    ...
    def dropna(
        self,
        *,
        axis: Axis = 0,
        inplace: bool = False,
        how: AnyAll | None = None,
        ignore_index: bool = False,
    ) -> Series | None:
        ...
        #매개변수의 String Literal 확인
        inplace = validate_bool_kwarg(inplace, "inplace")
        ignore_index = validate_bool_kwarg(ignore_index, "ignore_index")
        # Validate the axis parameter
        self._get_axis_number(axis or 0)
        
        #na를 제거한 새로운 Series를 반환받음
        if self._can_hold_na:
            result = remove_na_arraylike(self)
        ...

        #inplace여부에따라 분기를 나눠서 result로 덮어씌움
        if inplace:
            return self._update_inplace(result)
        else:
            return result