diff --git a/pandas/core/indexes/multi.py b/pandas/core/indexes/multi.py index 8955329a7afbf..c5cab225fa7b1 100644 --- a/pandas/core/indexes/multi.py +++ b/pandas/core/indexes/multi.py @@ -107,7 +107,10 @@ lexsort_indexer, ) -from pandas.io.formats.printing import pprint_thing +from pandas.io.formats.printing import ( + get_adjustment, + pprint_thing, +) if TYPE_CHECKING: from pandas import ( @@ -1437,8 +1440,6 @@ def format( ) if adjoin: - from pandas.io.formats.format import get_adjustment - adj = get_adjustment() return adj.adjoin(space, *result_levels).split("\n") else: diff --git a/pandas/io/common.py b/pandas/io/common.py index f255ea8197304..d08612f4f09f6 100644 --- a/pandas/io/common.py +++ b/pandas/io/common.py @@ -68,8 +68,8 @@ is_integer, is_list_like, ) +from pandas.core.dtypes.generic import ABCMultiIndex -from pandas.core.indexes.api import MultiIndex from pandas.core.shared_docs import _shared_docs _VALID_URLS = set(uses_relative + uses_netloc + uses_params) @@ -91,6 +91,8 @@ WriteBuffer, ) + from pandas import MultiIndex + @dataclasses.dataclass class IOArgs: @@ -1228,7 +1230,7 @@ def is_potential_multi_index( return bool( len(columns) - and not isinstance(columns, MultiIndex) + and not isinstance(columns, ABCMultiIndex) and all(isinstance(c, tuple) for c in columns if c not in list(index_col)) ) diff --git a/pandas/io/formats/format.py b/pandas/io/formats/format.py index 1efd6f18dc0af..cac83e2a48972 100644 --- a/pandas/io/formats/format.py +++ b/pandas/io/formats/format.py @@ -7,7 +7,6 @@ from collections.abc import ( Generator, Hashable, - Iterable, Mapping, Sequence, ) @@ -26,7 +25,6 @@ Final, cast, ) -from unicodedata import east_asian_width import numpy as np @@ -226,7 +224,7 @@ def __init__( float_format = get_option("display.float_format") self.float_format = float_format self.dtype = dtype - self.adj = get_adjustment() + self.adj = printing.get_adjustment() self._chk_truncate() @@ -347,69 +345,6 @@ def to_string(self) -> str: return str("".join(result)) -class _TextAdjustment: - def __init__(self) -> None: - self.encoding = get_option("display.encoding") - - def len(self, text: str) -> int: - return len(text) - - def justify(self, texts: Any, max_len: int, mode: str = "right") -> list[str]: - return printing.justify(texts, max_len, mode=mode) - - def adjoin(self, space: int, *lists, **kwargs) -> str: - return printing.adjoin( - space, *lists, strlen=self.len, justfunc=self.justify, **kwargs - ) - - -class _EastAsianTextAdjustment(_TextAdjustment): - def __init__(self) -> None: - super().__init__() - if get_option("display.unicode.ambiguous_as_wide"): - self.ambiguous_width = 2 - else: - self.ambiguous_width = 1 - - # Definition of East Asian Width - # https://unicode.org/reports/tr11/ - # Ambiguous width can be changed by option - self._EAW_MAP = {"Na": 1, "N": 1, "W": 2, "F": 2, "H": 1} - - def len(self, text: str) -> int: - """ - Calculate display width considering unicode East Asian Width - """ - if not isinstance(text, str): - return len(text) - - return sum( - self._EAW_MAP.get(east_asian_width(c), self.ambiguous_width) for c in text - ) - - def justify( - self, texts: Iterable[str], max_len: int, mode: str = "right" - ) -> list[str]: - # re-calculate padding space per str considering East Asian Width - def _get_pad(t): - return max_len - self.len(t) + len(t) - - if mode == "left": - return [x.ljust(_get_pad(x)) for x in texts] - elif mode == "center": - return [x.center(_get_pad(x)) for x in texts] - else: - return [x.rjust(_get_pad(x)) for x in texts] - - -def get_adjustment() -> _TextAdjustment: - use_east_asian_width = get_option("display.unicode.east_asian_width") - if use_east_asian_width: - return _EastAsianTextAdjustment() - else: - return _TextAdjustment() - - def get_dataframe_repr_params() -> dict[str, Any]: """Get the parameters used to repr(dataFrame) calls using DataFrame.to_string. @@ -529,7 +464,7 @@ def __init__( self.tr_frame = self.frame self.truncate() - self.adj = get_adjustment() + self.adj = printing.get_adjustment() def get_strcols(self) -> list[list[str]]: """ @@ -1789,13 +1724,13 @@ def _make_fixed_width( strings: list[str], justify: str = "right", minimum: int | None = None, - adj: _TextAdjustment | None = None, + adj: printing._TextAdjustment | None = None, ) -> list[str]: if len(strings) == 0 or justify == "all": return strings if adj is None: - adjustment = get_adjustment() + adjustment = printing.get_adjustment() else: adjustment = adj diff --git a/pandas/io/formats/printing.py b/pandas/io/formats/printing.py index b57797b7ec717..2cc9368f8846a 100644 --- a/pandas/io/formats/printing.py +++ b/pandas/io/formats/printing.py @@ -15,11 +15,14 @@ TypeVar, Union, ) +from unicodedata import east_asian_width from pandas._config import get_option from pandas.core.dtypes.inference import is_sequence +from pandas.io.formats.console import get_console_size + EscapeChars = Union[Mapping[str, str], Iterable[str]] _KT = TypeVar("_KT") _VT = TypeVar("_VT") @@ -42,7 +45,7 @@ def adjoin(space: int, *lists: list[str], **kwargs) -> str: function used to justify str. Needed for unicode handling. """ strlen = kwargs.pop("strlen", len) - justfunc = kwargs.pop("justfunc", justify) + justfunc = kwargs.pop("justfunc", _adj_justify) newLists = [] lengths = [max(map(strlen, x)) + space for x in lists[:-1]] @@ -57,7 +60,7 @@ def adjoin(space: int, *lists: list[str], **kwargs) -> str: return "\n".join("".join(lines) for lines in toJoin) -def justify(texts: Iterable[str], max_len: int, mode: str = "right") -> list[str]: +def _adj_justify(texts: Iterable[str], max_len: int, mode: str = "right") -> list[str]: """ Perform ljust, center, rjust against string or list-like """ @@ -314,9 +317,6 @@ def format_object_summary( ------- summary string """ - from pandas.io.formats.console import get_console_size - from pandas.io.formats.format import get_adjustment - display_width, _ = get_console_size() if display_width is None: display_width = get_option("display.width") or 80 @@ -501,3 +501,72 @@ class PrettyDict(dict[_KT, _VT]): def __repr__(self) -> str: return pprint_thing(self) + + +class _TextAdjustment: + def __init__(self) -> None: + self.encoding = get_option("display.encoding") + + def len(self, text: str) -> int: + return len(text) + + def justify(self, texts: Any, max_len: int, mode: str = "right") -> list[str]: + """ + Perform ljust, center, rjust against string or list-like + """ + if mode == "left": + return [x.ljust(max_len) for x in texts] + elif mode == "center": + return [x.center(max_len) for x in texts] + else: + return [x.rjust(max_len) for x in texts] + + def adjoin(self, space: int, *lists, **kwargs) -> str: + return adjoin(space, *lists, strlen=self.len, justfunc=self.justify, **kwargs) + + +class _EastAsianTextAdjustment(_TextAdjustment): + def __init__(self) -> None: + super().__init__() + if get_option("display.unicode.ambiguous_as_wide"): + self.ambiguous_width = 2 + else: + self.ambiguous_width = 1 + + # Definition of East Asian Width + # https://unicode.org/reports/tr11/ + # Ambiguous width can be changed by option + self._EAW_MAP = {"Na": 1, "N": 1, "W": 2, "F": 2, "H": 1} + + def len(self, text: str) -> int: + """ + Calculate display width considering unicode East Asian Width + """ + if not isinstance(text, str): + return len(text) + + return sum( + self._EAW_MAP.get(east_asian_width(c), self.ambiguous_width) for c in text + ) + + def justify( + self, texts: Iterable[str], max_len: int, mode: str = "right" + ) -> list[str]: + # re-calculate padding space per str considering East Asian Width + def _get_pad(t): + return max_len - self.len(t) + len(t) + + if mode == "left": + return [x.ljust(_get_pad(x)) for x in texts] + elif mode == "center": + return [x.center(_get_pad(x)) for x in texts] + else: + return [x.rjust(_get_pad(x)) for x in texts] + + +def get_adjustment() -> _TextAdjustment: + use_east_asian_width = get_option("display.unicode.east_asian_width") + if use_east_asian_width: + return _EastAsianTextAdjustment() + else: + return _TextAdjustment() diff --git a/pandas/tests/io/formats/test_format.py b/pandas/tests/io/formats/test_format.py index 585958765f0b6..57f1f082708ae 100644 --- a/pandas/tests/io/formats/test_format.py +++ b/pandas/tests/io/formats/test_format.py @@ -212,7 +212,7 @@ def test_repr_truncation(self): r = repr(df) r = r[r.find("\n") + 1 :] - adj = fmt.get_adjustment() + adj = printing.get_adjustment() for line, value in zip(r.split("\n"), df["B"]): if adj.len(value) + 1 > max_len: diff --git a/pandas/tests/io/formats/test_printing.py b/pandas/tests/io/formats/test_printing.py index 492657587ae0e..78198bce71460 100644 --- a/pandas/tests/io/formats/test_printing.py +++ b/pandas/tests/io/formats/test_printing.py @@ -8,7 +8,6 @@ import pandas as pd from pandas.io.formats import printing -import pandas.io.formats.format as fmt def test_adjoin(): @@ -48,7 +47,7 @@ def test_adjoin_unicode(self): adjoined = printing.adjoin(2, *data) assert adjoined == expected - adj = fmt._EastAsianTextAdjustment() + adj = printing._EastAsianTextAdjustment() expected = """あ dd ggg b ええ hhh @@ -73,7 +72,7 @@ def test_adjoin_unicode(self): assert adj.len(cols[2]) == 26 def test_justify(self): - adj = fmt._EastAsianTextAdjustment() + adj = printing._EastAsianTextAdjustment() def just(x, *args, **kwargs): # wrapper to test single str @@ -95,7 +94,7 @@ def just(x, *args, **kwargs): assert just("パンダ", 10, mode="right") == " パンダ" def test_east_asian_len(self): - adj = fmt._EastAsianTextAdjustment() + adj = printing._EastAsianTextAdjustment() assert adj.len("abc") == 3 assert adj.len("abc") == 3 @@ -106,11 +105,11 @@ def test_east_asian_len(self): assert adj.len("ﾊﾟﾝﾀﾞpanda") == 10 def test_ambiguous_width(self): - adj = fmt._EastAsianTextAdjustment() + adj = printing._EastAsianTextAdjustment() assert adj.len("¡¡ab") == 4 with cf.option_context("display.unicode.ambiguous_as_wide", True): - adj = fmt._EastAsianTextAdjustment() + adj = printing._EastAsianTextAdjustment() assert adj.len("¡¡ab") == 6 data = [["あ", "b", "c"], ["dd", "ええ", "ff"], ["ggg", "¡¡ab", "いいい"]]