valid parameters check while reading data

[shantanuo]

Pandas seems to ignore the extra (invalid) parameters. For e.g.

Code Sample, a copy-pastable example if possible

import pandas as pd
df=pd.read_excel('myfile.xlsx', some_dummy_param=True)

Note that some_dummy_param does not throw an error.

Problem description

Is there any way to make sure only valid parameters are passed to read_excel method?

Expected Output

Since there is no such parameter called "some_dummy_param", I should get an error:

TypeError: init() got an unexpected keyword argument 'some_dummy_param'

[TomAugspurger]

I thought we had another issue for this (#17994 was aobut sheet_name vs. sheetname).

[TomAugspurger]

The most difficult part is that kwds is passed through to the underlying engine. We would want to reconcile the common keyword arguments, and then perhaps provide an engine_kwargs parameter that's a dict of engine-specific kwargs.

[shantanuo]

That thread ends with your note "the goal is to remove the **kwargs from the signature". And the issue is closed! May be a decorator should be considered.

[TomAugspurger]

Yes, that issue was specifically for sheetname. I'm not sure why a decorator would be useful, could you explain?

…

On Fri, Aug 3, 2018 at 8:54 AM Shantanu Oak ***@***.***> wrote: That thread ends with your note "the goal is to remove the **kwargs from the signature". And the issue is closed! May be a decorator should be considered. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22189 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABQHImYCI1JJcEPUNIlQ3wDOs5DYByaNks5uNFYNgaJpZM4Vt6-x> .

[shantanuo]

I just saw a decorator called "deprecate_kwarg" to deprecate a keyword argument of a function. I thought similar concept can be used for "validate_kwarg". Ignore this comment if that is not the case.

[shantanuo]

May be this function in pandas/pandas/util/_decorators.py

def validate_kwarg():
    def _validate_kwarg(func):
        @wraps(func)
        def wrapper(*args, **kwargs):
            expected_keys=['io', 'sheet_name','header', 'names', 'index_col', 'usecols', 'squeeze', 'dtype', 'engine', 
               'converters', 'true_values',  'false_values', 'skiprows', 'nrows', 'na_values', 'verbose', 'parse_dates',
               'date_parser', 'thousands', 'comment', 'skipfooter', 'convert_float']

            if set(kwargs.keys()).difference(set(expected_keys)): 
                raise ValueError('invalid parameter found')

            return func(*args, **kwargs)
        return wrapper
    return _validate_kwarg

and a decorater added in the file pandas/io/excel.py

@validate_kwarg()
@appender(_read_excel_doc)
@deprecate_kwarg("parse_cols", "usecols")
@deprecate_kwarg("skip_footer", "skipfooter")
def read_excel(io,

[TomAugspurger]

I don't think a decorator is necessary here, since this is only being applied to a single function. I think the best path forward is to remove the need for **kwargs in read_excel.

…

On Fri, Aug 3, 2018 at 3:01 PM Shantanu Oak ***@***.***> wrote: May be this function def validate_kwarg(): def _validate_kwarg(func): @wraps(func) def wrapper(*args, **kwargs): expected_keys=['io', 'sheet_name','header', 'names', 'index_col', 'usecols', 'squeeze', 'dtype', 'engine', 'converters', 'true_values', 'false_values', 'skiprows', 'nrows', 'na_values', 'verbose', 'parse_dates', 'date_parser', 'thousands', 'comment', 'skipfooter', 'convert_float'] if set(kwargs.keys()).difference(set(expected_keys)): raise ValueError('invalid parameter found') return func(*args, **kwargs) return wrapper return _validate_kwarg and a decorater added in the file pandas/io/excel.py @validate_kwarg() @appender(_read_excel_doc) @deprecate_kwarg("parse_cols", "usecols") @deprecate_kwarg("skip_footer", "skipfooter") def read_excel(io, — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22189 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABQHIkYJTJmBk5RfO4VVkAP4Q7jYPIGlks5uNKwQgaJpZM4Vt6-x> .

[mroeschke]

engine_kwargs is getting added in 1.3 to read_excel which I think fits the spirit of this issue.

https://pandas.pydata.org/pandas-docs/dev/whatsnew/v1.3.0.html#other-api-changes

Closing, but happy to reopen if there is more to this issue