Pandas 数据框样式：根据格式列突出显示一些单元格

Question

Problem description

I have a DataFrame in which last column is a format column. The purpose of this column is to contain the format of the DataFrame row. Here is an example of such a dataframe:

df = pd.DataFrame({'ID': [1, 24, 31, 37],
                   'Status': ['to analyze', 'to analyze','to analyze','analyzed'],
                   'priority' : ['P1','P1','P2','P1'],
                   'format' : ['n;y;n','n;n;n','n;y;y','y;n;y']}

Each df['format'] row contains a string intended to be taken as a list (when split) to give the format of the row.

Symbols meaning:

• n means "no highlight"
• y means "to highlight in yellow"

df['format'].to_list()[0] = 'n;y;n' means for example:

• n: first column ID item "1" not highlighted
• y: second column Status item "to analyze" to be highlighted
• n: third column Priority item "P1" not highlighted

So that expected outcome is:

What I've tried

I've tried to use df.format to get a list of lists containing the format needed. Here is my code:

import pandas as pd
import numpy as np

def highlight_case(df):
    list_of_format_lists = []
    for format_line in df['format']:
        format_line_list = format_line.split(';')
        format_list = []
        for form in format_line_list:
            if 'y' in form:
                format_list.append('background-color: yellow')
            else:
                format_list.append('')
        list_of_format_lists.append(format_list)
    list_of_format_lists = list(map(list, zip(*list_of_format_lists)))#transpose
    print(list_of_format_lists)
    return list_of_format_lists


highlight_style = highlight_case(df)
df.style.apply(highlight_style)

It doesn't work, and I get this output:

TypeError                                 Traceback (most recent call last)
c:\python38\lib\site-packages\IPython\core\formatters.py in __call__(self, obj)
    343             method = get_real_method(obj, self.print_method)
    344             if method is not None:
--> 345                 return method()
    346             return None
    347         else:

c:\python38\lib\site-packages\pandas\io\formats\style.py in _repr_html_(self)
    191         Hooks into Jupyter notebook rich display system.
    192         """
--> 193         return self.render()
    194 
    195     @doc(NDFrame.to_excel, klass="Styler")

c:\python38\lib\site-packages\pandas\io\formats\style.py in render(self, **kwargs)
    538         * table_attributes
    539         """
--> 540         self._compute()
    541         # TODO: namespace all the pandas keys
    542         d = self._translate()

c:\python38\lib\site-packages\pandas\io\formats\style.py in _compute(self)
    623         r = self
    624         for func, args, kwargs in self._todo:
--> 625             r = func(self)(*args, **kwargs)
    626         return r
    627 

c:\python38\lib\site-packages\pandas\io\formats\style.py in _apply(self, func, axis, subset, **kwargs)
    637         data = self.data.loc[subset]
    638         if axis is not None:
--> 639             result = data.apply(func, axis=axis, result_type="expand", **kwargs)
    640             result.columns = data.columns
    641         else:

c:\python38\lib\site-packages\pandas\core\frame.py in apply(self, func, axis, raw, result_type, args, **kwds)
   7543             kwds=kwds,
   7544         )
-> 7545         return op.get_result()
   7546 
   7547     def applymap(self, func) -> "DataFrame":

c:\python38\lib\site-packages\pandas\core\apply.py in get_result(self)
    142         # dispatch to agg
    143         if is_list_like(self.f) or is_dict_like(self.f):
--> 144             return self.obj.aggregate(self.f, axis=self.axis, *self.args, **self.kwds)
    145 
    146         # all empty

c:\python38\lib\site-packages\pandas\core\frame.py in aggregate(self, func, axis, *args, **kwargs)
   7353         axis = self._get_axis_number(axis)
   7354 
-> 7355         relabeling, func, columns, order = reconstruct_func(func, **kwargs)
   7356 
   7357         result = None

c:\python38\lib\site-packages\pandas\core\aggregation.py in reconstruct_func(func, **kwargs)
     74 
     75     if not relabeling:
---> 76         if isinstance(func, list) and len(func) > len(set(func)):
     77 
     78             # GH 28426 will raise error if duplicated function names are used and

TypeError: unhashable type: 'list'

Answer 1

Since the formats are encoded for each row, it makes sense apply row-wise:

def format_row(r):
    formats = r['format'].split(';')
    return ['background-color: yellow' if y=='y' else '' for y in formats] + ['']

df.style.apply(format_row, axis=1)

Output: