Overview over all function objects

user_14042465 0
Viewed 61

I have a question which is more about a general understanding of loops and case-structures. Therefor I will not deliver a runable code.

Is there somehow an option to iterate over objects of a function? I use a if-elif differentiation to call diffent files of a dataset.

Is it possible for the user of this function to make a request about all possible cases the programmer/data base manager offers in this function?

I try to better explain my question in an example:

Example of available datasets given in a if-else-structure (the user might not be aware of all possible choices that can be called):

def data(date):
    if date == '20220318':
        file = 'dataset_a'
        return file
    elif date == '20220317':
        file = 'dataset_b'
        return file
    elif date == '20210712':
        file = 'dataset_c'
        return file
    elif date == '20210708':
        file = 'dataset_d'
        return file
    elif date == '20210621':
        file = 'dataset_e'
        return file      
    else:
        raise Exception("Invalid date")

The result I am aiming at is a list of all "date"-objects like:

20220318
20220317
20210712
20210708
20210621

A direct iteration seems not to be possible like:

for date in data:
    print(date)

This leads to TypeError: 'function' object is not iterable. Is there a way to call all possible choises?

Thanks for helping to understand how loops really work.

for-loop python-3.x python
1 Answers

In answer to the question:

"Is it possible for the user of this function to make a request about all possible cases the programmer/data base manager offers in this function?"

The answer is yes. Below are listed two methods in particular, one 'messy' way, and another more robust way - but will require function re-factoring.

The messy way:

(Please do not use this in production code)

The advantage of this approach, is that you only require access to the function, and re-factoring is not required.

This approach uses the inspect module to parse the function into text. Then, a regex pattern is used to extract the possible date options, and returns these as a list.

import inspect
import re

# Define the regex pattern to extract the required values.
rexp = re.compile('date\s==\s\'(\d+)\'')
# Get source code in plain text.
src = inspect.getsource(data)
# Extract date strings.
dates = rexp.findall(src)

Output:

['20220318', '20220317', '20210712', '20210708', '20210621']

The more robust way:

This approach will require re-factoring the function, but will likely serve better in the long term. (recommended)

Simply place all possible dates and their associated file in a dict. Use the dict.get() method to return the requested value, or a default string if the requested value is not found.

The function also features a return_dates_only argument, which if passed as True, returns the keys to the lookup dictionary and provides the user with the valid options.

def data2(*, date: str=None, return_dates_only: bool=False):
    lookup = {'20220318': 'dataset_a',
              '20220317': 'dataset_b',
              '20210712': 'dataset_c',
              '20210708': 'dataset_d',
              '20210621': 'dataset_e'}
    if return_dates_only:
        return list(lookup.keys())
    else:
        return lookup.get(date, '[Error]: Invalid date')

Note: The asterisk in the function declaration tells the function to accept keyword arguments only.

Usage:

>>> data(date='20220318')
`dataset_a`

>>> data(date='20220101')
'[Error]: Invalid date'

>>> data()
'[Error]: Invalid date'

>>> data(return_dates_only=True)
['20220318', '20220317', '20210712', '20210708', '20210621']

Comments:

In answer to the question:

"Can someone help me to understand why a concept like for date in data ... cannot work?"

Per the __iter__ method's documentation:

This [an __iter__ method] is required to allow both containers and iterators to be used with the for and in statements.

... which the data function does not (nor can it) have.

user_63404960
Related