ENH: pd.concat with keys and ignore_index=True should raise

[WillAyd]

Feature Type

• Adding new functionality to pandas

• Changing existing functionality in pandas

• Removing existing functionality in pandas

Problem Description

You can happily pass both of these arguments together, but the ignore_index=True call drops the keys, so it ends up being a moot point:

In [1]: import pandas as pd

In [2]: df1 = pd.DataFrame([[0]])

In [3]: df2 = pd.DataFrame([[42]])

In [4]: pd.concat([df1, df2])
Out[4]: 
    0
0   0
0  42

In [5]: pd.concat([df1, df2], keys=["df1", "df2"])
Out[5]: 
        0
df1 0   0
df2 0  42

In [6]: pd.concat([df1, df2], keys=["df1", "df2"], ignore_index=True)
Out[6]: 
    0
0   0
1  42

Feature Description

pandas can raise that this combination of values does not make sense

Alternative Solutions

status quo

Additional Context

No response

[Animesh-Shukla]

take

[Prabhat-Thapa45]

@WillAyd so kinda warning should be raised to warn this as counterproductive action

[MarcoGorelli]

@Prabhat-Thapa45 rather than warning, I think the suggestion is that an informative error be raised

[jahn96]

take