s452625/projektAI
2
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
5e1e575b90
projektAI/venv/Lib/site-packages/pandas/io/formats/format.py

2056 lines
64 KiB
Python
Raw Normal View History

Działa
2021-06-06 22:13:05 +02:00
"""
Internal module for formatting output data in csv, html,
and latex files. This module also applies to display formatting.
"""
from contextlib import contextmanager
from csv import QUOTE_NONE, QUOTE_NONNUMERIC
import decimal
from functools import partial
from io import StringIO
import math
import re
from shutil import get_terminal_size
from typing import (
IO,
TYPE_CHECKING,
Any,
Callable,
Dict,
Iterable,
List,
Mapping,
Optional,
Sequence,
Tuple,
Type,
Union,
cast,
)
from unicodedata import east_asian_width
import numpy as np
from pandas._config.config import get_option, set_option
from pandas._libs import lib
from pandas._libs.missing import NA
from pandas._libs.tslibs import NaT, Timedelta, Timestamp, iNaT
from pandas._libs.tslibs.nattype import NaTType
from pandas._typing import (
ArrayLike,
CompressionOptions,
FilePathOrBuffer,
FloatFormatType,
IndexLabel,
Label,
StorageOptions,
)
from pandas.core.dtypes.common import (
is_categorical_dtype,
is_complex_dtype,
is_datetime64_dtype,
is_datetime64tz_dtype,
is_extension_array_dtype,
is_float,
is_float_dtype,
is_integer,
is_integer_dtype,
is_list_like,
is_numeric_dtype,
is_scalar,
is_timedelta64_dtype,
)
from pandas.core.dtypes.missing import isna, notna
from pandas.core.arrays.datetimes import DatetimeArray
from pandas.core.arrays.timedeltas import TimedeltaArray
from pandas.core.base import PandasObject
import pandas.core.common as com
from pandas.core.construction import extract_array
from pandas.core.indexes.api import Index, MultiIndex, PeriodIndex, ensure_index
from pandas.core.indexes.datetimes import DatetimeIndex
from pandas.core.indexes.timedeltas import TimedeltaIndex
from pandas.core.reshape.concat import concat
from pandas.io.common import stringify_path
from pandas.io.formats.printing import adjoin, justify, pprint_thing
if TYPE_CHECKING:
from pandas import Categorical, DataFrame, Series
FormattersType = Union[
List[Callable], Tuple[Callable, ...], Mapping[Union[str, int], Callable]
]
ColspaceType = Mapping[Label, Union[str, int]]
ColspaceArgType = Union[
str, int, Sequence[Union[str, int]], Mapping[Label, Union[str, int]]
]
common_docstring = """
Parameters
----------
buf : str, Path or StringIO-like, optional, default None
Buffer to write to. If None, the output is returned as a string.
columns : sequence, optional, default None
The subset of columns to write. Writes all columns by default.
col_space : %(col_space_type)s, optional
%(col_space)s.
header : %(header_type)s, optional
%(header)s.
index : bool, optional, default True
Whether to print index (row) labels.
na_rep : str, optional, default 'NaN'
String representation of ``NaN`` to use.
formatters : list, tuple or dict of one-param. functions, optional
Formatter functions to apply to columns' elements by position or
name.
The result of each function must be a unicode string.
List/tuple must be of length equal to the number of columns.
float_format : one-parameter function, optional, default None
Formatter function to apply to columns' elements if they are
floats. This function must return a unicode string and will be
applied only to the non-``NaN`` elements, with ``NaN`` being
handled by ``na_rep``.
.. versionchanged:: 1.2.0
sparsify : bool, optional, default True
Set to False for a DataFrame with a hierarchical index to print
every multiindex key at each row.
index_names : bool, optional, default True
Prints the names of the indexes.
justify : str, default None
How to justify the column labels. If None uses the option from
the print configuration (controlled by set_option), 'right' out
of the box. Valid values are
* left
* right
* center
* justify
* justify-all
* start
* end
* inherit
* match-parent
* initial
* unset.
max_rows : int, optional
Maximum number of rows to display in the console.
min_rows : int, optional
The number of rows to display in the console in a truncated repr
(when number of rows is above `max_rows`).
max_cols : int, optional
Maximum number of columns to display in the console.
show_dimensions : bool, default False
Display DataFrame dimensions (number of rows by number of columns).
decimal : str, default '.'
Character recognized as decimal separator, e.g. ',' in Europe.
"""
_VALID_JUSTIFY_PARAMETERS = (
"left",
"right",
"center",
"justify",
"justify-all",
"start",
"end",
"inherit",
"match-parent",
"initial",
"unset",
)
return_docstring = """
Returns
-------
str or None
If buf is None, returns the result as a string. Otherwise returns
None.
"""
class CategoricalFormatter:
def __init__(
self,
categorical: "Categorical",
buf: Optional[IO[str]] = None,
length: bool = True,
na_rep: str = "NaN",
footer: bool = True,
):
self.categorical = categorical
self.buf = buf if buf is not None else StringIO("")
self.na_rep = na_rep
self.length = length
self.footer = footer
self.quoting = QUOTE_NONNUMERIC
def _get_footer(self) -> str:
footer = ""
if self.length:
if footer:
footer += ", "
footer += f"Length: {len(self.categorical)}"
level_info = self.categorical._repr_categories_info()
# Levels are added in a newline
if footer:
footer += "\n"
footer += level_info
return str(footer)
def _get_formatted_values(self) -> List[str]:
return format_array(
self.categorical._internal_get_values(),
None,
float_format=None,
na_rep=self.na_rep,
quoting=self.quoting,
)
def to_string(self) -> str:
categorical = self.categorical
if len(categorical) == 0:
if self.footer:
return self._get_footer()
else:
return ""
fmt_values = self._get_formatted_values()
fmt_values = [i.strip() for i in fmt_values]
values = ", ".join(fmt_values)
result = ["[" + values + "]"]
if self.footer:
footer = self._get_footer()
if footer:
result.append(footer)
return str("\n".join(result))
class SeriesFormatter:
def __init__(
self,
series: "Series",
buf: Optional[IO[str]] = None,
length: Union[bool, str] = True,
header: bool = True,
index: bool = True,
na_rep: str = "NaN",
name: bool = False,
float_format: Optional[str] = None,
dtype: bool = True,
max_rows: Optional[int] = None,
min_rows: Optional[int] = None,
):
self.series = series
self.buf = buf if buf is not None else StringIO()
self.name = name
self.na_rep = na_rep
self.header = header
self.length = length
self.index = index
self.max_rows = max_rows
self.min_rows = min_rows
if float_format is None:
float_format = get_option("display.float_format")
self.float_format = float_format
self.dtype = dtype
self.adj = get_adjustment()
self._chk_truncate()
def _chk_truncate(self) -> None:
self.tr_row_num: Optional[int]
min_rows = self.min_rows
max_rows = self.max_rows
# truncation determined by max_rows, actual truncated number of rows
# used below by min_rows
is_truncated_vertically = max_rows and (len(self.series) > max_rows)
series = self.series
if is_truncated_vertically:
max_rows = cast(int, max_rows)
if min_rows:
# if min_rows is set (not None or 0), set max_rows to minimum
# of both
max_rows = min(min_rows, max_rows)
if max_rows == 1:
row_num = max_rows
series = series.iloc[:max_rows]
else:
row_num = max_rows // 2
series = concat((series.iloc[:row_num], series.iloc[-row_num:]))
self.tr_row_num = row_num
else:
self.tr_row_num = None
self.tr_series = series
self.is_truncated_vertically = is_truncated_vertically
def _get_footer(self) -> str:
name = self.series.name
footer = ""
if getattr(self.series.index, "freq", None) is not None:
assert isinstance(
self.series.index, (DatetimeIndex, PeriodIndex, TimedeltaIndex)
)
footer += f"Freq: {self.series.index.freqstr}"
if self.name is not False and name is not None:
if footer:
footer += ", "
series_name = pprint_thing(name, escape_chars=("\t", "\r", "\n"))
footer += f"Name: {series_name}"
if self.length is True or (
self.length == "truncate" and self.is_truncated_vertically
):
if footer:
footer += ", "
footer += f"Length: {len(self.series)}"
if self.dtype is not False and self.dtype is not None:
dtype_name = getattr(self.tr_series.dtype, "name", None)
if dtype_name:
if footer:
footer += ", "
footer += f"dtype: {pprint_thing(dtype_name)}"
# level infos are added to the end and in a new line, like it is done
# for Categoricals
if is_categorical_dtype(self.tr_series.dtype):
level_info = self.tr_series._values._repr_categories_info()
if footer:
footer += "\n"
footer += level_info
return str(footer)
def _get_formatted_index(self) -> Tuple[List[str], bool]:
index = self.tr_series.index
if isinstance(index, MultiIndex):
have_header = any(name for name in index.names)
fmt_index = index.format(names=True)
else:
have_header = index.name is not None
fmt_index = index.format(name=True)
return fmt_index, have_header
def _get_formatted_values(self) -> List[str]:
return format_array(
self.tr_series._values,
None,
float_format=self.float_format,
na_rep=self.na_rep,
leading_space=self.index,
)
def to_string(self) -> str:
series = self.tr_series
footer = self._get_footer()
if len(series) == 0:
return f"{type(self.series).__name__}([], {footer})"
fmt_index, have_header = self._get_formatted_index()
fmt_values = self._get_formatted_values()
if self.is_truncated_vertically:
n_header_rows = 0
row_num = self.tr_row_num
row_num = cast(int, row_num)
width = self.adj.len(fmt_values[row_num - 1])
if width > 3:
dot_str = "..."
else:
dot_str = ".."
# Series uses mode=center because it has single value columns
# DataFrame uses mode=left
dot_str = self.adj.justify([dot_str], width, mode="center")[0]
fmt_values.insert(row_num + n_header_rows, dot_str)
fmt_index.insert(row_num + 1, "")
if self.index:
result = self.adj.adjoin(3, *[fmt_index[1:], fmt_values])
else:
result = self.adj.adjoin(3, fmt_values)
if self.header and have_header:
result = fmt_index[0] + "\n" + result
if footer:
result += "\n" + footer
return str("".join(result))
class TextAdjustment:
def __init__(self):
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 justify(texts, max_len, mode=mode)
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):
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()
class DataFrameFormatter:
"""Class for processing dataframe formatting options and data."""
__doc__ = __doc__ if __doc__ else ""
__doc__ += common_docstring + return_docstring
def __init__(
self,
frame: "DataFrame",
columns: Optional[Sequence[str]] = None,
col_space: Optional[ColspaceArgType] = None,
header: Union[bool, Sequence[str]] = True,
index: bool = True,
na_rep: str = "NaN",
formatters: Optional[FormattersType] = None,
justify: Optional[str] = None,
float_format: Optional[FloatFormatType] = None,
sparsify: Optional[bool] = None,
index_names: bool = True,
max_rows: Optional[int] = None,
min_rows: Optional[int] = None,
max_cols: Optional[int] = None,
show_dimensions: Union[bool, str] = False,
decimal: str = ".",
bold_rows: bool = False,
escape: bool = True,
):
self.frame = frame
self.columns = self._initialize_columns(columns)
self.col_space = self._initialize_colspace(col_space)
self.header = header
self.index = index
self.na_rep = na_rep
self.formatters = self._initialize_formatters(formatters)
self.justify = self._initialize_justify(justify)
self.float_format = float_format
self.sparsify = self._initialize_sparsify(sparsify)
self.show_index_names = index_names
self.decimal = decimal
self.bold_rows = bold_rows
self.escape = escape
self.max_rows = max_rows
self.min_rows = min_rows
self.max_cols = max_cols
self.show_dimensions = show_dimensions
self.max_cols_fitted = self._calc_max_cols_fitted()
self.max_rows_fitted = self._calc_max_rows_fitted()
self.tr_frame = self.frame
self.truncate()
self.adj = get_adjustment()
def get_strcols(self) -> List[List[str]]:
"""
Render a DataFrame to a list of columns (as lists of strings).
"""
strcols = self._get_strcols_without_index()
if self.index:
str_index = self._get_formatted_index(self.tr_frame)
strcols.insert(0, str_index)
return strcols
@property
def should_show_dimensions(self) -> bool:
return self.show_dimensions is True or (
self.show_dimensions == "truncate" and self.is_truncated
)
@property
def is_truncated(self) -> bool:
return bool(self.is_truncated_horizontally or self.is_truncated_vertically)
@property
def is_truncated_horizontally(self) -> bool:
return bool(self.max_cols_fitted and (len(self.columns) > self.max_cols_fitted))
@property
def is_truncated_vertically(self) -> bool:
return bool(self.max_rows_fitted and (len(self.frame) > self.max_rows_fitted))
@property
def dimensions_info(self) -> str:
return f"\n\n[{len(self.frame)} rows x {len(self.frame.columns)} columns]"
@property
def has_index_names(self) -> bool:
return _has_names(self.frame.index)
@property
def has_column_names(self) -> bool:
return _has_names(self.frame.columns)
@property
def show_row_idx_names(self) -> bool:
return all((self.has_index_names, self.index, self.show_index_names))
@property
def show_col_idx_names(self) -> bool:
return all((self.has_column_names, self.show_index_names, self.header))
@property
def max_rows_displayed(self) -> int:
return min(self.max_rows or len(self.frame), len(self.frame))
def _initialize_sparsify(self, sparsify: Optional[bool]) -> bool:
if sparsify is None:
return get_option("display.multi_sparse")
return sparsify
def _initialize_formatters(
self, formatters: Optional[FormattersType]
) -> FormattersType:
if formatters is None:
return {}
elif len(self.frame.columns) == len(formatters) or isinstance(formatters, dict):
return formatters
else:
raise ValueError(
f"Formatters length({len(formatters)}) should match "
f"DataFrame number of columns({len(self.frame.columns)})"
)
def _initialize_justify(self, justify: Optional[str]) -> str:
if justify is None:
return get_option("display.colheader_justify")
else:
return justify
def _initialize_columns(self, columns: Optional[Sequence[str]]) -> Index:
if columns is not None:
cols = ensure_index(columns)
self.frame = self.frame[cols]
return cols
else:
return self.frame.columns
def _initialize_colspace(
self, col_space: Optional[ColspaceArgType]
) -> ColspaceType:
result: ColspaceType
if col_space is None:
result = {}
elif isinstance(col_space, (int, str)):
result = {"": col_space}
result.update({column: col_space for column in self.frame.columns})
elif isinstance(col_space, Mapping):
for column in col_space.keys():
if column not in self.frame.columns and column != "":
raise ValueError(
f"Col_space is defined for an unknown column: {column}"
)
result = col_space
else:
if len(self.frame.columns) != len(col_space):
raise ValueError(
f"Col_space length({len(col_space)}) should match "
f"DataFrame number of columns({len(self.frame.columns)})"
)
result = dict(zip(self.frame.columns, col_space))
return result
def _calc_max_cols_fitted(self) -> Optional[int]:
"""Number of columns fitting the screen."""
if not self._is_in_terminal():
return self.max_cols
width, _ = get_terminal_size()
if self._is_screen_narrow(width):
return width
else:
return self.max_cols
def _calc_max_rows_fitted(self) -> Optional[int]:
"""Number of rows with data fitting the screen."""
max_rows: Optional[int]
if self._is_in_terminal():
_, height = get_terminal_size()
if self.max_rows == 0:
# rows available to fill with actual data
return height - self._get_number_of_auxillary_rows()
if self._is_screen_short(height):
max_rows = height
else:
max_rows = self.max_rows
else:
max_rows = self.max_rows
return self._adjust_max_rows(max_rows)
def _adjust_max_rows(self, max_rows: Optional[int]) -> Optional[int]:
"""Adjust max_rows using display logic.
See description here:
https://pandas.pydata.org/docs/dev/user_guide/options.html#frequently-used-options
GH #37359
"""
if max_rows:
if (len(self.frame) > max_rows) and self.min_rows:
# if truncated, set max_rows showed to min_rows
max_rows = min(self.min_rows, max_rows)
return max_rows
def _is_in_terminal(self) -> bool:
"""Check if the output is to be shown in terminal."""
return bool(self.max_cols == 0 or self.max_rows == 0)
def _is_screen_narrow(self, max_width) -> bool:
return bool(self.max_cols == 0 and len(self.frame.columns) > max_width)
def _is_screen_short(self, max_height) -> bool:
return bool(self.max_rows == 0 and len(self.frame) > max_height)
def _get_number_of_auxillary_rows(self) -> int:
"""Get number of rows occupied by prompt, dots and dimension info."""
dot_row = 1
prompt_row = 1
num_rows = dot_row + prompt_row
if self.show_dimensions:
num_rows += len(self.dimensions_info.splitlines())
if self.header:
num_rows += 1
return num_rows
def truncate(self) -> None:
"""
Check whether the frame should be truncated. If so, slice the frame up.
"""
if self.is_truncated_horizontally:
self._truncate_horizontally()
if self.is_truncated_vertically:
self._truncate_vertically()
def _truncate_horizontally(self) -> None:
"""Remove columns, which are not to be displayed and adjust formatters.
Attributes affected:
- tr_frame
- formatters
- tr_col_num
"""
assert self.max_cols_fitted is not None
col_num = self.max_cols_fitted // 2
if col_num >= 1:
left = self.tr_frame.iloc[:, :col_num]
right = self.tr_frame.iloc[:, -col_num:]
self.tr_frame = concat((left, right), axis=1)
# truncate formatter
if isinstance(self.formatters, (list, tuple)):
self.formatters = [
*self.formatters[:col_num],
*self.formatters[-col_num:],
]
else:
col_num = cast(int, self.max_cols)
self.tr_frame = self.tr_frame.iloc[:, :col_num]
self.tr_col_num = col_num
def _truncate_vertically(self) -> None:
"""Remove rows, which are not to be displayed.
Attributes affected:
- tr_frame
- tr_row_num
"""
assert self.max_rows_fitted is not None
row_num = self.max_rows_fitted // 2
if row_num >= 1:
head = self.tr_frame.iloc[:row_num, :]
tail = self.tr_frame.iloc[-row_num:, :]
self.tr_frame = concat((head, tail))
else:
row_num = cast(int, self.max_rows)
self.tr_frame = self.tr_frame.iloc[:row_num, :]
self.tr_row_num = row_num
def _get_strcols_without_index(self) -> List[List[str]]:
strcols: List[List[str]] = []
if not is_list_like(self.header) and not self.header:
for i, c in enumerate(self.tr_frame):
fmt_values = self.format_col(i)
fmt_values = _make_fixed_width(
strings=fmt_values,
justify=self.justify,
minimum=int(self.col_space.get(c, 0)),
adj=self.adj,
)
strcols.append(fmt_values)
return strcols
if is_list_like(self.header):
# cast here since can't be bool if is_list_like
self.header = cast(List[str], self.header)
if len(self.header) != len(self.columns):
raise ValueError(
f"Writing {len(self.columns)} cols "
f"but got {len(self.header)} aliases"
)
str_columns = [[label] for label in self.header]
else:
str_columns = self._get_formatted_column_labels(self.tr_frame)
if self.show_row_idx_names:
for x in str_columns:
x.append("")
for i, c in enumerate(self.tr_frame):
cheader = str_columns[i]
header_colwidth = max(
int(self.col_space.get(c, 0)), *(self.adj.len(x) for x in cheader)
)
fmt_values = self.format_col(i)
fmt_values = _make_fixed_width(
fmt_values, self.justify, minimum=header_colwidth, adj=self.adj
)
max_len = max(max(self.adj.len(x) for x in fmt_values), header_colwidth)
cheader = self.adj.justify(cheader, max_len, mode=self.justify)
strcols.append(cheader + fmt_values)
return strcols
def format_col(self, i: int) -> List[str]:
frame = self.tr_frame
formatter = self._get_formatter(i)
return format_array(
frame.iloc[:, i]._values,
formatter,
float_format=self.float_format,
na_rep=self.na_rep,
space=self.col_space.get(frame.columns[i]),
decimal=self.decimal,
leading_space=self.index,
)
def _get_formatter(self, i: Union[str, int]) -> Optional[Callable]:
if isinstance(self.formatters, (list, tuple)):
if is_integer(i):
i = cast(int, i)
return self.formatters[i]
else:
return None
else:
if is_integer(i) and i not in self.columns:
i = self.columns[i]
return self.formatters.get(i, None)
def _get_formatted_column_labels(self, frame: "DataFrame") -> List[List[str]]:
from pandas.core.indexes.multi import sparsify_labels
columns = frame.columns
if isinstance(columns, MultiIndex):
fmt_columns = columns.format(sparsify=False, adjoin=False)
fmt_columns = list(zip(*fmt_columns))
dtypes = self.frame.dtypes._values
# if we have a Float level, they don't use leading space at all
restrict_formatting = any(level.is_floating for level in columns.levels)
need_leadsp = dict(zip(fmt_columns, map(is_numeric_dtype, dtypes)))
def space_format(x, y):
if (
y not in self.formatters
and need_leadsp[x]
and not restrict_formatting
):
return " " + y
return y
str_columns = list(
zip(*[[space_format(x, y) for y in x] for x in fmt_columns])
)
if self.sparsify and len(str_columns):
str_columns = sparsify_labels(str_columns)
str_columns = [list(x) for x in zip(*str_columns)]
else:
fmt_columns = columns.format()
dtypes = self.frame.dtypes
need_leadsp = dict(zip(fmt_columns, map(is_numeric_dtype, dtypes)))
str_columns = [
[" " + x if not self._get_formatter(i) and need_leadsp[x] else x]
for i, (col, x) in enumerate(zip(columns, fmt_columns))
]
# self.str_columns = str_columns
return str_columns
def _get_formatted_index(self, frame: "DataFrame") -> List[str]:
# Note: this is only used by to_string() and to_latex(), not by
# to_html(). so safe to cast col_space here.
col_space = {k: cast(int, v) for k, v in self.col_space.items()}
index = frame.index
columns = frame.columns
fmt = self._get_formatter("__index__")
if isinstance(index, MultiIndex):
fmt_index = index.format(
sparsify=self.sparsify,
adjoin=False,
names=self.show_row_idx_names,
formatter=fmt,
)
else:
fmt_index = [index.format(name=self.show_row_idx_names, formatter=fmt)]
fmt_index = [
tuple(
_make_fixed_width(
list(x), justify="left", minimum=col_space.get("", 0), adj=self.adj
)
)
for x in fmt_index
]
adjoined = self.adj.adjoin(1, *fmt_index).split("\n")
# empty space for columns
if self.show_col_idx_names:
col_header = [str(x) for x in self._get_column_name_list()]
else:
col_header = [""] * columns.nlevels
if self.header:
return col_header + adjoined
else:
return adjoined
def _get_column_name_list(self) -> List[str]:
names: List[str] = []
columns = self.frame.columns
if isinstance(columns, MultiIndex):
names.extend("" if name is None else name for name in columns.names)
else:
names.append("" if columns.name is None else columns.name)
return names
class DataFrameRenderer:
"""Class for creating dataframe output in multiple formats.
Called in pandas.core.generic.NDFrame:
- to_csv
- to_latex
Called in pandas.core.frame.DataFrame:
- to_html
- to_string
Parameters
----------
fmt : DataFrameFormatter
Formatter with the formating options.
"""
def __init__(self, fmt: DataFrameFormatter):
self.fmt = fmt
def to_latex(
self,
buf: Optional[FilePathOrBuffer[str]] = None,
column_format: Optional[str] = None,
longtable: bool = False,
encoding: Optional[str] = None,
multicolumn: bool = False,
multicolumn_format: Optional[str] = None,
multirow: bool = False,
caption: Optional[str] = None,
label: Optional[str] = None,
position: Optional[str] = None,
) -> Optional[str]:
"""
Render a DataFrame to a LaTeX tabular/longtable environment output.
"""
from pandas.io.formats.latex import LatexFormatter
latex_formatter = LatexFormatter(
self.fmt,
longtable=longtable,
column_format=column_format,
multicolumn=multicolumn,
multicolumn_format=multicolumn_format,
multirow=multirow,
caption=caption,
label=label,
position=position,
)
string = latex_formatter.to_string()
return save_to_buffer(string, buf=buf, encoding=encoding)
def to_html(
self,
buf: Optional[FilePathOrBuffer[str]] = None,
encoding: Optional[str] = None,
classes: Optional[Union[str, List, Tuple]] = None,
notebook: bool = False,
border: Optional[int] = None,
table_id: Optional[str] = None,
render_links: bool = False,
) -> Optional[str]:
"""
Render a DataFrame to a html table.
Parameters
----------
buf : str, Path or StringIO-like, optional, default None
Buffer to write to. If None, the output is returned as a string.
encoding : str, default “utf-8”
Set character encoding.
classes : str or list-like
classes to include in the `class` attribute of the opening
``<table>`` tag, in addition to the default "dataframe".
notebook : {True, False}, optional, default False
Whether the generated HTML is for IPython Notebook.
border : int
A ``border=border`` attribute is included in the opening
``<table>`` tag. Default ``pd.options.display.html.border``.
table_id : str, optional
A css id is included in the opening `<table>` tag if specified.
render_links : bool, default False
Convert URLs to HTML links.
"""
from pandas.io.formats.html import HTMLFormatter, NotebookFormatter
Klass = NotebookFormatter if notebook else HTMLFormatter
html_formatter = Klass(
self.fmt,
classes=classes,
border=border,
table_id=table_id,
render_links=render_links,
)
string = html_formatter.to_string()
return save_to_buffer(string, buf=buf, encoding=encoding)
def to_string(
self,
buf: Optional[FilePathOrBuffer[str]] = None,
encoding: Optional[str] = None,
line_width: Optional[int] = None,
) -> Optional[str]:
"""
Render a DataFrame to a console-friendly tabular output.
Parameters
----------
buf : str, Path or StringIO-like, optional, default None
Buffer to write to. If None, the output is returned as a string.
encoding: str, default “utf-8”
Set character encoding.
line_width : int, optional
Width to wrap a line in characters.
"""
from pandas.io.formats.string import StringFormatter
string_formatter = StringFormatter(self.fmt, line_width=line_width)
string = string_formatter.to_string()
return save_to_buffer(string, buf=buf, encoding=encoding)
def to_csv(
self,
path_or_buf: Optional[FilePathOrBuffer[str]] = None,
encoding: Optional[str] = None,
sep: str = ",",
columns: Optional[Sequence[Label]] = None,
index_label: Optional[IndexLabel] = None,
mode: str = "w",
compression: CompressionOptions = "infer",
quoting: Optional[int] = None,
quotechar: str = '"',
line_terminator: Optional[str] = None,
chunksize: Optional[int] = None,
date_format: Optional[str] = None,
doublequote: bool = True,
escapechar: Optional[str] = None,
errors: str = "strict",
storage_options: StorageOptions = None,
) -> Optional[str]:
"""
Render dataframe as comma-separated file.
"""
from pandas.io.formats.csvs import CSVFormatter
if path_or_buf is None:
created_buffer = True
path_or_buf = StringIO()
else:
created_buffer = False
csv_formatter = CSVFormatter(
path_or_buf=path_or_buf,
line_terminator=line_terminator,
sep=sep,
encoding=encoding,
errors=errors,
compression=compression,
quoting=quoting,
cols=columns,
index_label=index_label,
mode=mode,
chunksize=chunksize,
quotechar=quotechar,
date_format=date_format,
doublequote=doublequote,
escapechar=escapechar,
storage_options=storage_options,
formatter=self.fmt,
)
csv_formatter.save()
if created_buffer:
assert isinstance(path_or_buf, StringIO)
content = path_or_buf.getvalue()
path_or_buf.close()
return content
return None
def save_to_buffer(
string: str,
buf: Optional[FilePathOrBuffer[str]] = None,
encoding: Optional[str] = None,
) -> Optional[str]:
"""
Perform serialization. Write to buf or return as string if buf is None.
"""
with get_buffer(buf, encoding=encoding) as f:
f.write(string)
if buf is None:
return f.getvalue()
return None
@contextmanager
def get_buffer(buf: Optional[FilePathOrBuffer[str]], encoding: Optional[str] = None):
"""
Context manager to open, yield and close buffer for filenames or Path-like
objects, otherwise yield buf unchanged.
"""
if buf is not None:
buf = stringify_path(buf)
else:
buf = StringIO()
if encoding is None:
encoding = "utf-8"
elif not isinstance(buf, str):
raise ValueError("buf is not a file name and encoding is specified.")
if hasattr(buf, "write"):
yield buf
elif isinstance(buf, str):
with open(buf, "w", encoding=encoding, newline="") as f:
# GH#30034 open instead of codecs.open prevents a file leak
# if we have an invalid encoding argument.
# newline="" is needed to roundtrip correctly on
# windows test_to_latex_filename
yield f
else:
raise TypeError("buf is not a file name and it has no write method")
# ----------------------------------------------------------------------
# Array formatters
def format_array(
values: Any,
formatter: Optional[Callable],
float_format: Optional[FloatFormatType] = None,
na_rep: str = "NaN",
digits: Optional[int] = None,
space: Optional[Union[str, int]] = None,
justify: str = "right",
decimal: str = ".",
leading_space: Optional[bool] = True,
quoting: Optional[int] = None,
) -> List[str]:
"""
Format an array for printing.
Parameters
----------
values
formatter
float_format
na_rep
digits
space
justify
decimal
leading_space : bool, optional, default True
Whether the array should be formatted with a leading space.
When an array as a column of a Series or DataFrame, we do want
the leading space to pad between columns.
When formatting an Index subclass
(e.g. IntervalIndex._format_native_types), we don't want the
leading space since it should be left-aligned.
Returns
-------
List[str]
"""
fmt_klass: Type[GenericArrayFormatter]
if is_datetime64_dtype(values.dtype):
fmt_klass = Datetime64Formatter
elif is_datetime64tz_dtype(values.dtype):
fmt_klass = Datetime64TZFormatter
elif is_timedelta64_dtype(values.dtype):
fmt_klass = Timedelta64Formatter
elif is_extension_array_dtype(values.dtype):
fmt_klass = ExtensionArrayFormatter
elif is_float_dtype(values.dtype) or is_complex_dtype(values.dtype):
fmt_klass = FloatArrayFormatter
elif is_integer_dtype(values.dtype):
fmt_klass = IntArrayFormatter
else:
fmt_klass = GenericArrayFormatter
if space is None:
space = get_option("display.column_space")
if float_format is None:
float_format = get_option("display.float_format")
if digits is None:
digits = get_option("display.precision")
fmt_obj = fmt_klass(
values,
digits=digits,
na_rep=na_rep,
float_format=float_format,
formatter=formatter,
space=space,
justify=justify,
decimal=decimal,
leading_space=leading_space,
quoting=quoting,
)
return fmt_obj.get_result()
class GenericArrayFormatter:
def __init__(
self,
values: Any,
digits: int = 7,
formatter: Optional[Callable] = None,
na_rep: str = "NaN",
space: Union[str, int] = 12,
float_format: Optional[FloatFormatType] = None,
justify: str = "right",
decimal: str = ".",
quoting: Optional[int] = None,
fixed_width: bool = True,
leading_space: Optional[bool] = True,
):
self.values = values
self.digits = digits
self.na_rep = na_rep
self.space = space
self.formatter = formatter
self.float_format = float_format
self.justify = justify
self.decimal = decimal
self.quoting = quoting
self.fixed_width = fixed_width
self.leading_space = leading_space
def get_result(self) -> List[str]:
fmt_values = self._format_strings()
return _make_fixed_width(fmt_values, self.justify)
def _format_strings(self) -> List[str]:
if self.float_format is None:
float_format = get_option("display.float_format")
if float_format is None:
precision = get_option("display.precision")
float_format = lambda x: _trim_zeros_single_float(
f"{x: .{precision:d}f}"
)
else:
float_format = self.float_format
if self.formatter is not None:
formatter = self.formatter
else:
quote_strings = self.quoting is not None and self.quoting != QUOTE_NONE
formatter = partial(
pprint_thing,
escape_chars=("\t", "\r", "\n"),
quote_strings=quote_strings,
)
def _format(x):
if self.na_rep is not None and is_scalar(x) and isna(x):
try:
# try block for np.isnat specifically
# determine na_rep if x is None or NaT-like
if x is None:
return "None"
elif x is NA:
return str(NA)
elif x is NaT or np.isnat(x):
return "NaT"
except (TypeError, ValueError):
# np.isnat only handles datetime or timedelta objects
pass
return self.na_rep
elif isinstance(x, PandasObject):
return str(x)
else:
# object dtype
return str(formatter(x))
vals = extract_array(self.values, extract_numpy=True)
is_float_type = (
lib.map_infer(vals, is_float)
# vals may have 2 or more dimensions
& np.all(notna(vals), axis=tuple(range(1, len(vals.shape))))
)
leading_space = self.leading_space
if leading_space is None:
leading_space = is_float_type.any()
fmt_values = []
for i, v in enumerate(vals):
if not is_float_type[i] and leading_space:
fmt_values.append(f" {_format(v)}")
elif is_float_type[i]:
fmt_values.append(float_format(v))
else:
if leading_space is False:
# False specifically, so that the default is
# to include a space if we get here.
tpl = "{v}"
else:
tpl = " {v}"
fmt_values.append(tpl.format(v=_format(v)))
return fmt_values
class FloatArrayFormatter(GenericArrayFormatter):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
# float_format is expected to be a string
# formatter should be used to pass a function
if self.float_format is not None and self.formatter is None:
# GH21625, GH22270
self.fixed_width = False
if callable(self.float_format):
self.formatter = self.float_format
self.float_format = None
def _value_formatter(
self,
float_format: Optional[FloatFormatType] = None,
threshold: Optional[Union[float, int]] = None,
) -> Callable:
"""Returns a function to be applied on each value to format it"""
# the float_format parameter supersedes self.float_format
if float_format is None:
float_format = self.float_format
# we are going to compose different functions, to first convert to
# a string, then replace the decimal symbol, and finally chop according
# to the threshold
# when there is no float_format, we use str instead of '%g'
# because str(0.0) = '0.0' while '%g' % 0.0 = '0'
if float_format:
def base_formatter(v):
assert float_format is not None # for mypy
# pandas\io\formats\format.py:1411: error: "str" not callable
# [operator]
# pandas\io\formats\format.py:1411: error: Unexpected keyword
# argument "value" for "__call__" of "EngFormatter" [call-arg]
return (
float_format(value=v) # type: ignore[operator,call-arg]
if notna(v)
else self.na_rep
)
else:
def base_formatter(v):
return str(v) if notna(v) else self.na_rep
if self.decimal != ".":
def decimal_formatter(v):
return base_formatter(v).replace(".", self.decimal, 1)
else:
decimal_formatter = base_formatter
if threshold is None:
return decimal_formatter
def formatter(value):
if notna(value):
if abs(value) > threshold:
return decimal_formatter(value)
else:
return decimal_formatter(0.0)
else:
return self.na_rep
return formatter
def get_result_as_array(self) -> np.ndarray:
"""
Returns the float values converted into strings using
the parameters given at initialisation, as a numpy array
"""
def format_with_na_rep(values: ArrayLike, formatter: Callable, na_rep: str):
mask = isna(values)
formatted = np.array(
[
formatter(val) if not m else na_rep
for val, m in zip(values.ravel(), mask.ravel())
]
).reshape(values.shape)
return formatted
if self.formatter is not None:
return format_with_na_rep(self.values, self.formatter, self.na_rep)
if self.fixed_width:
threshold = get_option("display.chop_threshold")
else:
threshold = None
# if we have a fixed_width, we'll need to try different float_format
def format_values_with(float_format):
formatter = self._value_formatter(float_format, threshold)
# default formatter leaves a space to the left when formatting
# floats, must be consistent for left-justifying NaNs (GH #25061)
if self.justify == "left":
na_rep = " " + self.na_rep
else:
na_rep = self.na_rep
# separate the wheat from the chaff
values = self.values
is_complex = is_complex_dtype(values)
values = format_with_na_rep(values, formatter, na_rep)
if self.fixed_width:
if is_complex:
result = _trim_zeros_complex(values, self.decimal)
else:
result = _trim_zeros_float(values, self.decimal)
return np.asarray(result, dtype="object")
return values
# There is a special default string when we are fixed-width
# The default is otherwise to use str instead of a formatting string
float_format: Optional[FloatFormatType]
if self.float_format is None:
if self.fixed_width:
if self.leading_space is True:
fmt_str = "{value: .{digits:d}f}"
else:
fmt_str = "{value:.{digits:d}f}"
float_format = partial(fmt_str.format, digits=self.digits)
else:
float_format = self.float_format
else:
float_format = lambda value: self.float_format % value
formatted_values = format_values_with(float_format)
if not self.fixed_width:
return formatted_values
# we need do convert to engineering format if some values are too small
# and would appear as 0, or if some values are too big and take too
# much space
if len(formatted_values) > 0:
maxlen = max(len(x) for x in formatted_values)
too_long = maxlen > self.digits + 6
else:
too_long = False
with np.errstate(invalid="ignore"):
abs_vals = np.abs(self.values)
# this is pretty arbitrary for now
# large values: more that 8 characters including decimal symbol
# and first digit, hence > 1e6
has_large_values = (abs_vals > 1e6).any()
has_small_values = (
(abs_vals < 10 ** (-self.digits)) & (abs_vals > 0)
).any()
if has_small_values or (too_long and has_large_values):
if self.leading_space is True:
fmt_str = "{value: .{digits:d}e}"
else:
fmt_str = "{value:.{digits:d}e}"
float_format = partial(fmt_str.format, digits=self.digits)
formatted_values = format_values_with(float_format)
return formatted_values
def _format_strings(self) -> List[str]:
return list(self.get_result_as_array())
class IntArrayFormatter(GenericArrayFormatter):
def _format_strings(self) -> List[str]:
if self.leading_space is False:
formatter_str = lambda x: f"{x:d}".format(x=x)
else:
formatter_str = lambda x: f"{x: d}".format(x=x)
formatter = self.formatter or formatter_str
fmt_values = [formatter(x) for x in self.values]
return fmt_values
class Datetime64Formatter(GenericArrayFormatter):
def __init__(
self,
values: Union[np.ndarray, "Series", DatetimeIndex, DatetimeArray],
nat_rep: str = "NaT",
date_format: None = None,
**kwargs,
):
super().__init__(values, **kwargs)
self.nat_rep = nat_rep
self.date_format = date_format
def _format_strings(self) -> List[str]:
""" we by definition have DO NOT have a TZ """
values = self.values
if not isinstance(values, DatetimeIndex):
values = DatetimeIndex(values)
if self.formatter is not None and callable(self.formatter):
return [self.formatter(x) for x in values]
fmt_values = values._data._format_native_types(
na_rep=self.nat_rep, date_format=self.date_format
)
return fmt_values.tolist()
class ExtensionArrayFormatter(GenericArrayFormatter):
def _format_strings(self) -> List[str]:
values = extract_array(self.values, extract_numpy=True)
formatter = self.formatter
if formatter is None:
formatter = values._formatter(boxed=True)
if is_categorical_dtype(values.dtype):
# Categorical is special for now, so that we can preserve tzinfo
array = values._internal_get_values()
else:
array = np.asarray(values)
fmt_values = format_array(
array,
formatter,
float_format=self.float_format,
na_rep=self.na_rep,
digits=self.digits,
space=self.space,
justify=self.justify,
decimal=self.decimal,
leading_space=self.leading_space,
quoting=self.quoting,
)
return fmt_values
def format_percentiles(
percentiles: Union[
np.ndarray, List[Union[int, float]], List[float], List[Union[str, float]]
]
) -> List[str]:
"""
Outputs rounded and formatted percentiles.
Parameters
----------
percentiles : list-like, containing floats from interval [0,1]
Returns
-------
formatted : list of strings
Notes
-----
Rounding precision is chosen so that: (1) if any two elements of
``percentiles`` differ, they remain different after rounding
(2) no entry is *rounded* to 0% or 100%.
Any non-integer is always rounded to at least 1 decimal place.
Examples
--------
Keeps all entries different after rounding:
>>> format_percentiles([0.01999, 0.02001, 0.5, 0.666666, 0.9999])
['1.999%', '2.001%', '50%', '66.667%', '99.99%']
No element is rounded to 0% or 100% (unless already equal to it).
Duplicates are allowed:
>>> format_percentiles([0, 0.5, 0.02001, 0.5, 0.666666, 0.9999])
['0%', '50%', '2.0%', '50%', '66.67%', '99.99%']
"""
percentiles = np.asarray(percentiles)
# It checks for np.NaN as well
with np.errstate(invalid="ignore"):
if (
not is_numeric_dtype(percentiles)
or not np.all(percentiles >= 0)
or not np.all(percentiles <= 1)
):
raise ValueError("percentiles should all be in the interval [0,1]")
percentiles = 100 * percentiles
int_idx = np.isclose(percentiles.astype(int), percentiles)
if np.all(int_idx):
out = percentiles.astype(int).astype(str)
return [i + "%" for i in out]
unique_pcts = np.unique(percentiles)
to_begin = unique_pcts[0] if unique_pcts[0] > 0 else None
to_end = 100 - unique_pcts[-1] if unique_pcts[-1] < 100 else None
# Least precision that keeps percentiles unique after rounding
prec = -np.floor(
np.log10(np.min(np.ediff1d(unique_pcts, to_begin=to_begin, to_end=to_end)))
).astype(int)
prec = max(1, prec)
out = np.empty_like(percentiles, dtype=object)
out[int_idx] = percentiles[int_idx].astype(int).astype(str)
out[~int_idx] = percentiles[~int_idx].round(prec).astype(str)
return [i + "%" for i in out]
def is_dates_only(
values: Union[np.ndarray, DatetimeArray, Index, DatetimeIndex]
) -> bool:
# return a boolean if we are only dates (and don't have a timezone)
if not isinstance(values, Index):
values = values.ravel()
values = DatetimeIndex(values)
if values.tz is not None:
return False
values_int = values.asi8
consider_values = values_int != iNaT
one_day_nanos = 86400 * 1e9
even_days = (
np.logical_and(consider_values, values_int % int(one_day_nanos) != 0).sum() == 0
)
if even_days:
return True
return False
def _format_datetime64(x: Union[NaTType, Timestamp], nat_rep: str = "NaT") -> str:
if x is NaT:
return nat_rep
return str(x)
def _format_datetime64_dateonly(
x: Union[NaTType, Timestamp],
nat_rep: str = "NaT",
date_format: Optional[str] = None,
) -> str:
if x is NaT:
return nat_rep
if date_format:
return x.strftime(date_format)
else:
return x._date_repr
def get_format_datetime64(
is_dates_only: bool, nat_rep: str = "NaT", date_format: Optional[str] = None
) -> Callable:
if is_dates_only:
return lambda x: _format_datetime64_dateonly(
x, nat_rep=nat_rep, date_format=date_format
)
else:
return lambda x: _format_datetime64(x, nat_rep=nat_rep)
def get_format_datetime64_from_values(
values: Union[np.ndarray, DatetimeArray, DatetimeIndex], date_format: Optional[str]
) -> Optional[str]:
""" given values and a date_format, return a string format """
if isinstance(values, np.ndarray) and values.ndim > 1:
# We don't actually care about the order of values, and DatetimeIndex
# only accepts 1D values
values = values.ravel()
ido = is_dates_only(values)
if ido:
return date_format or "%Y-%m-%d"
return date_format
class Datetime64TZFormatter(Datetime64Formatter):
def _format_strings(self) -> List[str]:
""" we by definition have a TZ """
values = self.values.astype(object)
ido = is_dates_only(values)
formatter = self.formatter or get_format_datetime64(
ido, date_format=self.date_format
)
fmt_values = [formatter(x) for x in values]
return fmt_values
class Timedelta64Formatter(GenericArrayFormatter):
def __init__(
self,
values: Union[np.ndarray, TimedeltaIndex],
nat_rep: str = "NaT",
box: bool = False,
**kwargs,
):
super().__init__(values, **kwargs)
self.nat_rep = nat_rep
self.box = box
def _format_strings(self) -> List[str]:
formatter = self.formatter or get_format_timedelta64(
self.values, nat_rep=self.nat_rep, box=self.box
)
return [formatter(x) for x in self.values]
def get_format_timedelta64(
values: Union[np.ndarray, TimedeltaIndex, TimedeltaArray],
nat_rep: str = "NaT",
box: bool = False,
) -> Callable:
"""
Return a formatter function for a range of timedeltas.
These will all have the same format argument
If box, then show the return in quotes
"""
values_int = values.astype(np.int64)
consider_values = values_int != iNaT
one_day_nanos = 86400 * 1e9
even_days = (
np.logical_and(consider_values, values_int % one_day_nanos != 0).sum() == 0
)
if even_days:
format = None
else:
format = "long"
def _formatter(x):
if x is None or (is_scalar(x) and isna(x)):
return nat_rep
if not isinstance(x, Timedelta):
x = Timedelta(x)
result = x._repr_base(format=format)
if box:
result = f"'{result}'"
return result
return _formatter
def _make_fixed_width(
strings: List[str],
justify: str = "right",
minimum: Optional[int] = None,
adj: Optional[TextAdjustment] = None,
) -> List[str]:
if len(strings) == 0 or justify == "all":
return strings
if adj is None:
adjustment = get_adjustment()
else:
adjustment = adj
max_len = max(adjustment.len(x) for x in strings)
if minimum is not None:
max_len = max(minimum, max_len)
conf_max = get_option("display.max_colwidth")
if conf_max is not None and max_len > conf_max:
max_len = conf_max
def just(x: str) -> str:
if conf_max is not None:
if (conf_max > 3) & (adjustment.len(x) > max_len):
x = x[: max_len - 3] + "..."
return x
strings = [just(x) for x in strings]
result = adjustment.justify(strings, max_len, mode=justify)
return result
def _trim_zeros_complex(str_complexes: np.ndarray, decimal: str = ".") -> List[str]:
"""
Separates the real and imaginary parts from the complex number, and
executes the _trim_zeros_float method on each of those.
"""
trimmed = [
"".join(_trim_zeros_float(re.split(r"([j+-])", x), decimal))
for x in str_complexes
]
# pad strings to the length of the longest trimmed string for alignment
lengths = [len(s) for s in trimmed]
max_length = max(lengths)
padded = [
s[: -((k - 1) // 2 + 1)] # real part
+ (max_length - k) // 2 * "0"
+ s[-((k - 1) // 2 + 1) : -((k - 1) // 2)] # + / -
+ s[-((k - 1) // 2) : -1] # imaginary part
+ (max_length - k) // 2 * "0"
+ s[-1]
for s, k in zip(trimmed, lengths)
]
return padded
def _trim_zeros_single_float(str_float: str) -> str:
"""
Trims trailing zeros after a decimal point,
leaving just one if necessary.
"""
str_float = str_float.rstrip("0")
if str_float.endswith("."):
str_float += "0"
return str_float
def _trim_zeros_float(
str_floats: Union[np.ndarray, List[str]], decimal: str = "."
) -> List[str]:
"""
Trims the maximum number of trailing zeros equally from
all numbers containing decimals, leaving just one if
necessary.
"""
trimmed = str_floats
number_regex = re.compile(fr"^\s*[\+-]?[0-9]+\{decimal}[0-9]*$")
def is_number_with_decimal(x):
return re.match(number_regex, x) is not None
def should_trim(values: Union[np.ndarray, List[str]]) -> bool:
"""
Determine if an array of strings should be trimmed.
Returns True if all numbers containing decimals (defined by the
above regular expression) within the array end in a zero, otherwise
returns False.
"""
numbers = [x for x in values if is_number_with_decimal(x)]
return len(numbers) > 0 and all(x.endswith("0") for x in numbers)
while should_trim(trimmed):
trimmed = [x[:-1] if is_number_with_decimal(x) else x for x in trimmed]
# leave one 0 after the decimal points if need be.
result = [
x + "0" if is_number_with_decimal(x) and x.endswith(decimal) else x
for x in trimmed
]
return result
def _has_names(index: Index) -> bool:
if isinstance(index, MultiIndex):
return com.any_not_none(*index.names)
else:
return index.name is not None
class EngFormatter:
"""
Formats float values according to engineering format.
Based on matplotlib.ticker.EngFormatter
"""
# The SI engineering prefixes
ENG_PREFIXES = {
-24: "y",
-21: "z",
-18: "a",
-15: "f",
-12: "p",
-9: "n",
-6: "u",
-3: "m",
0: "",
3: "k",
6: "M",
9: "G",
12: "T",
15: "P",
18: "E",
21: "Z",
24: "Y",
}
def __init__(self, accuracy: Optional[int] = None, use_eng_prefix: bool = False):
self.accuracy = accuracy
self.use_eng_prefix = use_eng_prefix
def __call__(self, num: Union[int, float]) -> str:
"""
Formats a number in engineering notation, appending a letter
representing the power of 1000 of the original number. Some examples:
>>> format_eng(0) # for self.accuracy = 0
' 0'
>>> format_eng(1000000) # for self.accuracy = 1,
# self.use_eng_prefix = True
' 1.0M'
>>> format_eng("-1e-6") # for self.accuracy = 2
# self.use_eng_prefix = False
'-1.00E-06'
@param num: the value to represent
@type num: either a numeric value or a string that can be converted to
a numeric value (as per decimal.Decimal constructor)
@return: engineering formatted string
"""
dnum = decimal.Decimal(str(num))
if decimal.Decimal.is_nan(dnum):
return "NaN"
if decimal.Decimal.is_infinite(dnum):
return "inf"
sign = 1
if dnum < 0: # pragma: no cover
sign = -1
dnum = -dnum
if dnum != 0:
pow10 = decimal.Decimal(int(math.floor(dnum.log10() / 3) * 3))
else:
pow10 = decimal.Decimal(0)
pow10 = pow10.min(max(self.ENG_PREFIXES.keys()))
pow10 = pow10.max(min(self.ENG_PREFIXES.keys()))
int_pow10 = int(pow10)
if self.use_eng_prefix:
prefix = self.ENG_PREFIXES[int_pow10]
else:
if int_pow10 < 0:
prefix = f"E-{-int_pow10:02d}"
else:
prefix = f"E+{int_pow10:02d}"
mant = sign * dnum / (10 ** pow10)
if self.accuracy is None: # pragma: no cover
format_str = "{mant: g}{prefix}"
else:
format_str = f"{{mant: .{self.accuracy:d}f}}{{prefix}}"
formatted = format_str.format(mant=mant, prefix=prefix)
return formatted
def set_eng_float_format(accuracy: int = 3, use_eng_prefix: bool = False) -> None:
"""
Alter default behavior on how float is formatted in DataFrame.
Format float in engineering format. By accuracy, we mean the number of
decimal digits after the floating point.
See also EngFormatter.
"""
set_option("display.float_format", EngFormatter(accuracy, use_eng_prefix))
set_option("display.column_space", max(12, accuracy + 9))
def get_level_lengths(
levels: Any, sentinel: Union[bool, object, str] = ""
) -> List[Dict[int, int]]:
"""
For each index in each level the function returns lengths of indexes.
Parameters
----------
levels : list of lists
List of values on for level.
sentinel : string, optional
Value which states that no new index starts on there.
Returns
-------
Returns list of maps. For each level returns map of indexes (key is index
in row and value is length of index).
"""
if len(levels) == 0:
return []
control = [True] * len(levels[0])
result = []
for level in levels:
last_index = 0
lengths = {}
for i, key in enumerate(level):
if control[i] and key == sentinel:
pass
else:
control[i] = False
lengths[last_index] = i - last_index
last_index = i
lengths[last_index] = len(level) - last_index
result.append(lengths)
return result
def buffer_put_lines(buf: IO[str], lines: List[str]) -> None:
"""
Appends lines to a buffer.
Parameters
----------
buf
The buffer to write to
lines
The lines to append.
"""
if any(isinstance(x, str) for x in lines):
lines = [str(x) for x in lines]
buf.write("\n".join(lines))
Reference in New Issue Copy Permalink