如何在Pandas applymap中依据Policy列设置Snapshot Time单元格样式

现有如下表格数据存储在Pandas DataFrame中：

当前代码用applymap实现了Snapshot Time超过24小时就高亮单元格，但需求要改成：PROD_BACKUP策略对应24小时阈值，NONPROD_BACKUP策略对应72小时阈值，需要调整代码让高亮逻辑能获取当前行的Policy值来判断阈值。

因为applymap是逐单元格处理，无法直接获取同一行的其他列数据，所以需要改用apply方法按行处理，这样就能拿到当前行的Policy值来选择对应阈值。具体修改如下：

1. 调整高亮函数 #

重写highlight_snap函数，让它接收整行数据，先根据Policy列的值确定阈值，再调用原有的highlight_cells函数生成样式：

def highlight_snap(row, check_time):
    # 根据当前行的Policy选择对应阈值
    limit = 24 if row['Policy'] == 'PROD_BACKUP' else 72
    return highlight_cells(row['Snapshot Time'], check_time, limit)

2. 替换applymap为按行处理的apply #

把原来的applymap调用换成apply，指定axis=1（按行处理），同时确保只给Snapshot Time单元格应用样式：

df_styler = df_styler.apply(
    lambda row: ['', '', highlight_snap(row, report_start_time)],
    axis=1
)

这里lambda返回的列表和列数对应，前两列（DB、Policy）返回空样式，仅第三列（Snapshot Time）应用高亮逻辑。

import pandas as pd
from datetime import datetime,timedelta

report_start_time = datetime.now()


def highlight_cells(val, check_time, limit):
    time_diff = check_time - val
    total_hours = (time_diff.days*24) + (time_diff.seconds / (60*60))

    if total_hours >= limit:
        format_code = '''background-color: #B00202;
           font-weight: bold'''
    else:
        format_code = ''

    return format_code

def highlight_snap(row, check_time):
    # 根据Policy选择对应阈值
    limit = 24 if row['Policy'] == 'PROD_BACKUP' else 72
    return highlight_cells(row['Snapshot Time'], check_time, limit)

d = {'DB': ['A-DB', 'B-DB', 'C-DB'], 
     'Policy': ['PROD_BACKUP','PROD_BACKUP','NONPROD_BACKUP'], 
     'Snapshot Time': ['10/17/2022 12:00:00','10/16/2022 10:00:00','10/15/2022 16:00:00']}
df = pd.DataFrame(data=d)
df['Snapshot Time'] = pd.to_datetime(df['Snapshot Time'])

table_styler = [
    {   "selector" : "table",
        "props":[   ("border", "3px solid black"),
                    ("border-collapse","separate"),
                    ("width", "100%")
                ]
     },
     {   "selector" : "th",
         "props":[   ("color","black"),
                     ("background-color", "#F5F3F3"),
                     ("border", "1px solid gray"),
                     ("padding", "3px 3px"),
                     ("border-collapse","separate"),
                     ("font-size", "16px")
                 ]
     },
     {   "selector" : "td",
         "props":[   ("color","black"),
                     ("border", "1px solid gray"),
                     ("padding", "1px 3px"),
                     ("font-size", "14px")
                 ]
     }
]

df_styler = df.style.set_table_styles(table_styler)

# 替换applymap为按行处理的apply
df_styler = df_styler.apply(
    lambda row: ['', '', highlight_snap(row, report_start_time)],
    axis=1
)

print(df_styler.to_html())

关键说明 #

• apply(axis=1)让函数可以访问整行数据，从而动态获取Policy列的值来设置阈值。
• 返回的样式列表和列数一一对应，确保只有Snapshot Time列被应用高亮规则。

内容的提问来源于stack exchange，提问作者Nick S