[ENH] Wrap pandas string methods

[ericmjl]

Brief Description

One thing that might be nice is to wrap all of the string methods such that they are now method-chainable.

If this could be done programmatically, that would be awesome. Otherwise, manually wrapping ones of interest, and building up a submodule of string method operations, would be a superb alternative.

Example API

The example API can be found in this notebook.

[szuckerman]

What about this?

Instead of:

@pf.register_dataframe_method
def str_join(df, column_name: str, sep: str, *args, **kwargs):
    """
    Wrapper around `df.str.join`
    Joins items in a list.    
    """

    df[column_name] = df[column_name].str.join(sep)
    return df

Make the function generic:

@pf.register_dataframe_method
def str(df, method_type, column, *arg, **kwargs):
    df[column_name] = getattr(df[column_name].str, method_type)(sep)
    return df

df.str('join', 'column_name', sep=',')

I've been looking for a way to make the function name generic, too, (i.e. so we could use df.str_join
and in pyjanitor the function name would be something like df.str_ + method_name) but that looks like you have to use setattr within a class and might not work with @pf.register_dataframe_method.

[eyaltrabelsi]

@szuckerman @ericmjl I would like o work on this, what testing strategy do you think i should take?

[samukweku]

@ericmjl , Kindly assign to me if this is still available. @eyaltrabelsi asked first though, so please let’s check that he/she is still interested in it. Thanks

[ericmjl]

Oh my, it looks like I totally let this one drop. @eyaltrabelsi, I'm so sorry.

@samukweku, given the time interval, I think you can go ahead without any issues. Please do so 😄.

[ericmjl]

🎉