如何在Python Click CLI中传递过滤后的Pandas数据至保存函数？

我是Python编程新手，正在使用Pandas和Click开发一款CLI工具，支持用户通过预设命令筛选CSV数据。目前已实现查看原始数据的list()函数和按颜色筛选数据的color()函数，功能符合预期。我希望添加save()函数，将筛选后的数据导出为新CSV文件，但无法实现将color()函数中过滤得到的数据传递至save()函数。我尝试将new_df设为全局变量，也尝试返回new_df，但均未成功。以下是我的代码：

#!/usr/bin/env python
import click #imports click for CLI commands 
import pandas as pd #import for pandas 

df = pd.read_csv('dogs.csv') #reads the original data set.
@click.group() #creates a group 
@click.help_option('--help', help='displays the list of available commands') #adds help option

def cli():
    pass

##this shows the original list of cats 
@click.command() 
def list():
    """Simple command that lists all original dogs."""
    click.echo(f"{df}")

### this lists the filtered list of colored cats 
@click.command() 
@click.option("--color", prompt="Enter the desired color", help="The color of the dogs")
def color(color):
    """Simple command that filters dogs based on COLOR."""
    color = color.lower()
    new_df = df[df['Color'].str.contains(color)]
    click.echo(f"{new_df}")
    
### i want this to save the changes to a new csv file 
@click.command()
@click.option("--save",  help="Saves the file")
def save(save):
    """Simple command that saves the filtered list of dogs."""
    # i am lost here 
    save = new_df
    click.echo(f"{save}")
    
    
### add all commands to group 
cli.add_command(list)
cli.add_command(color)
cli.add_command(save)

if __name__ == "__main__":
    cli()

可行的解决方案 #

Click的命令默认是独立执行的，每次运行命令都是一个独立进程，所以直接在color和save之间传递内存中的DataFrame行不通。下面是几种实用的解决思路：

方法1：让color命令直接支持保存（最简洁推荐） #

修改color命令，添加--save选项，让用户在筛选时直接指定保存路径，一步完成筛选+保存：

#!/usr/bin/env python
import click
import pandas as pd

df = pd.read_csv('dogs.csv')

@click.group()
@click.help_option('--help', help='displays the list of available commands')
def cli():
    pass

@click.command()
def list():
    """Simple command that lists all original dogs."""
    click.echo(f"{df}")

@click.command()
@click.option("--color", prompt="Enter the desired color", help="The color of the dogs")
@click.option("--save", type=click.Path(), help="Path to save the filtered CSV file")
def color(color, save):
    """Simple command that filters dogs based on COLOR, optionally saves the result."""
    color = color.lower()
    new_df = df[df['Color'].str.contains(color)]
    click.echo(f"{new_df}")
    
    if save:
        new_df.to_csv(save, index=False)
        click.echo(f"Filtered data saved to {save}")

cli.add_command(list)
cli.add_command(color)

if __name__ == "__main__":
    cli()

使用示例：

# 只筛选不保存
python your_script.py color --color brown

# 筛选并保存到指定文件
python your_script.py color --color brown --save filtered_dogs.csv

方法2：使用全局变量（适合简单场景） #

如果一定要拆分color和save命令，可以用全局变量存储筛选结果，但仅在同一个进程会话中有效：

#!/usr/bin/env python
import click
import pandas as pd

df = pd.read_csv('dogs.csv')
filtered_df = None  # 全局变量存储筛选结果

@click.group()
@click.help_option('--help', help='displays the list of available commands')
def cli():
    pass

@click.command()
def list():
    """Simple command that lists all original dogs."""
    click.echo(f"{df}")

@click.command()
@click.option("--color", prompt="Enter the desired color", help="The color of the dogs")
def color(color):
    """Simple command that filters dogs based on COLOR."""
    global filtered_df
    color = color.lower()
    filtered_df = df[df['Color'].str.contains(color)]
    click.echo(f"{filtered_df}")

@click.command()
@click.option("--output", type=click.Path(), required=True, help="Path to save the filtered CSV file")
def save(output):
    """Simple command that saves the last filtered list of dogs."""
    global filtered_df
    if filtered_df is None:
        click.echo("Error: No filtered data found. Run 'color' command first.")
        return
    
    filtered_df.to_csv(output, index=False)
    click.echo(f"Filtered data saved to {output}")

cli.add_command(list)
cli.add_command(color)
cli.add_command(save)

if __name__ == "__main__":
    cli()

使用示例（需在同一个命令会话中执行）：

python your_script.py color --color brown && python your_script.py save --output filtered_dogs.csv

方法3：通过临时文件传递数据（更健壮） #

让color命令将筛选结果写入临时文件，save命令读取该文件并保存，适合跨进程传递数据：

#!/usr/bin/env python
import click
import pandas as pd
import tempfile
import os

df = pd.read_csv('dogs.csv')
TEMP_FILE = os.path.join(tempfile.gettempdir(), 'filtered_dogs_temp.csv')

@click.group()
@click.help_option('--help', help='displays the list of available commands')
def cli():
    pass

@click.command()
def list():
    """Simple command that lists all original dogs."""
    click.echo(f"{df}")

@click.command()
@click.option("--color", prompt="Enter the desired color", help="The color of the dogs")
@click.option("--keep-temp", is_flag=True, help="Keep the temporary file after filtering")
def color(color, keep_temp):
    """Simple command that filters dogs based on COLOR, saves result to temp file."""
    color = color.lower()
    new_df = df[df['Color'].str.contains(color)]
    click.echo(f"{new_df}")
    
    new_df.to_csv(TEMP_FILE, index=False)
    click.echo(f"Filtered data stored in temporary file: {TEMP_FILE}")
    
    if not keep_temp:
        # 注册脚本退出时删除临时文件
        import atexit
        atexit.register(lambda: os.remove(TEMP_FILE) if os.path.exists(TEMP_FILE) else None)

@click.command()
@click.option("--output", type=click.Path(), required=True, help="Path to save the filtered CSV file")
def save(output):
    """Simple command that saves the filtered data from temp file."""
    if not os.path.exists(TEMP_FILE):
        click.echo("Error: No filtered data found. Run 'color' command first.")
        return
    
    filtered_df = pd.read_csv(TEMP_FILE)
    filtered_df.to_csv(output, index=False)
    click.echo(f"Filtered data saved to {output}")

cli.add_command(list)
cli.add_command(color)
cli.add_command(save)

if __name__ == "__main__":
    cli()

使用示例：

# 筛选并生成临时文件
python your_script.py color --color brown

# 从临时文件保存到目标文件
python your_script.py save --output filtered_dogs.csv

总结 #

优先选择方法1，它符合CLI工具的设计逻辑，流程简洁直接。如果必须拆分命令，方法3比方法2更健壮，避免了全局变量的状态依赖问题。

内容的提问来源于stack exchange，提问作者chuggles