Python Switch/Case Statement Adaptation

Question

Ok, at the risk of being ridiculed for not 'trying harder', I have a scenario that I've been attempting to adapt to a pythonic switch case statement. I know python has the new match method in 3.10 but I'm limited to 3.8.10 in my AWS use case. I've been reading up on switch cases in other languages and I want to find a pythonic way to convert the following jumbled mess of if/elif/else statements to a clean switch case. I am wondering what others would do in this scenario

OBJECTIVE: I have a filename that will be passed into this sequence of code and I need to return the first three items (i.e. transaction_recipient_verification, transaction_account_tokenization, etc). Occasionally the code will receive a filename containing "field_results" or "issuers" and I need to make sure that the trimmed return string contains the respective case.

import random

sampleKeys = [
    'transaction_recipient_notification_status_sent/transaction_recipient_notification_status_sent_2021_10_29_12_02_14.snappy',
    'transaction_recipient_payment_status_success/transaction_recipient_payment_status_success_2021_10_29_12_02_14.snappy',
    'transaction_recipient_verification_rvdm_failure/transaction_recipient_verification_rvdm_failure_2021_10_29_12_02_14.snappy',
    'transaction_recipient_verification_rvdm_failure_field_results/transaction_recipient_verification_rvdm_failure_2021_10_29_12_02_14.snappy',
    'transaction_recipient_authentication_status_success/transaction_recipient_authentication_status_success_2021_10_29_12_02_14.snappy',
    'transaction_recipient_authentication_status_success_field_results/transaction_recipient_authentication_status_success_2021_10_29_12_02_14.snappy',
    'transaction_account_tokenization_success/transaction_account_tokenization_success_2021_10_29_12_02_14.snappy',
    'transaction_account_tokenization_success_issuers/transaction_account_tokenization_success_2021_10_29_12_02_14.snappy',
    'transaction_recipient_payment_status_terminated/transaction_recipient_payment_status_terminated_2021_10_29_12_02_14.snappy',
    'transaction_recipient_verification_rvdm_success/transaction_recipient_verification_rvdm_success_2021_10_29_12_02_14.snappy',
    'transaction_recipient_verification_rvdm_success_field_results/transaction_recipient_verification_rvdm_success_2021_10_29_12_02_14.snappy',
    'transaction_recipient_notification_status_received/transaction_recipient_notification_status_received_2021_10_29_12_02_14.snappy',
    'transaction_recipient_authentication_status_success/transaction_recipient_authentication_status_success_2021_10_29_11_17_45.snappy'
    
]

key = random.choice(sampleKeys)

array_data = any(substring in key for substring in ['_issuers', '_field_results'])
if not array_data:
    if 'transaction_recipient_notification' in key:
        keySubject = 'transaction_recipient_notification'
    elif 'transaction_recipient_authentication' in key:
        keySubject = 'transaction_recipient_authentication'
    elif 'transaction_recipient_verification' in key:
        keySubject = 'transaction_recipient_verification'
    elif 'transaction_account_verification' in key:
        keySubject = 'transaction_account_verification'
    elif 'transaction_account_tokenization' in key:
        keySubject = 'transaction_account_tokenization'
    elif 'transaction_recipient_payment' in key:
        keySubject = 'transaction_recipient_payment'
else:
    if '_issuers' in key:
        if 'transaction_recipient_notification' in key:
            keySubject = 'transaction_recipient_notification_issuers'
        elif 'transaction_recipient_authentication' in key:
            keySubject = 'transaction_recipient_authentication_issuers'
        elif 'transaction_recipient_verification' in key:
            keySubject = 'transaction_recipient_verification_issuers'
        elif 'transaction_account_verification' in key:
            keySubject = 'transaction_account_verification_issuers'
        elif 'transaction_account_tokenization' in key:
            keySubject = 'transaction_account_tokenization_issuers'
        elif 'transaction_recipient_payment' in key:
            keySubject = 'transaction_recipient_payment_issuers'
    elif '_field_results' in key:
        if 'transaction_recipient_notification' in key:
            keySubject = 'transaction_recipient_notification_field_results'
        elif 'transaction_recipient_authentication' in key:
            keySubject = 'transaction_recipient_authentication_field_results'
        elif 'transaction_recipient_verification' in key:
            keySubject = 'transaction_recipient_verification_field_results'
        elif 'transaction_account_verification' in key:
            keySubject = 'transaction_account_verification_field_results'
        elif 'transaction_account_tokenization' in key:
            keySubject = 'transaction_account_tokenization_field_results'
        elif 'transaction_recipient_payment' in key:
            keySubject = 'transaction_recipient_payment_field_results'
print(f'BEFORE ===> {key}')
print(f'AFTER  ===> {keySubject}')

Possible Direction:

import re
    
class MainKeyHandleSwitch:
    
    ARRAY_OPTIONS = ['_issuers', '_field_results']
        
    def __init__(self,key):
        self._original_key = key
        self._array_data = any(substring in key for substring in self.ARRAY_OPTIONS)
        self._trimmed_dict = self.trimmed_dict()
    
    @property
    def get_trimmed_dict(self):
        return self._trimmed_dict
    
    @property
    def get_trimmed_key(self):
        return self.__get_key_subject__()
    
    def trimmed_dict(self):
        trim_dict = dict()
        trim_dict['case_one'] = re.search('transaction_recipient_notification+', self._original_key)
        trim_dict['case_two'] = re.search('transaction_recipient_authentication+', self._original_key)
        trim_dict['case_three'] = re.search('transaction_recipient_verification+', self._original_key)
        trim_dict['case_four'] = re.search('transaction_account_verification+', self._original_key)
        trim_dict['case_five'] = re.search('transaction_account_tokenization+', self._original_key)
        trim_dict['case_six'] = re.search('transaction_recipient_payment+', self._original_key)
        return trim_dict
    
    def __get_key_subject__(self):
        obj = next(item for item in list(self._trimmed_dict.values()) if item is not None)
        if not self._array_data:
            return obj.group(0)
        else:
            if '_issuers' in self._original_key:
                return f'{obj.group(0)}_issuers'
            
            elif '_field_results' in self._original_key:
                return f'{obj.group(0)}_field_results'        

And the code to test the class:

import random

key = random.choice(sampleKeys)
print(f'before ===> {key}')
a = MainKeyHandleSwitch(key)
trimmed_key = a.get_trimmed_key
print(f'after  ===> {trimmed_key}')

Answer 1

I see a lot of repetition in your code, so the first thing I'll think of would be: "can I use a loop to simply this code?" and the answer is yes!

Since your code repeatedly used the six subjectTypes and the keySubject depends on the subject type, creating a list of the six types then use next() with a generator expression should simplify the over abundance of if's (If there weren't any correlations, a dictionary would work instead). Also, instead of array_data, you can use an if-elif-else clause to prevent an extra block level.

sampleKeys = [...]
key = random.choice(sampleKeys)

subjectTypes = ['transaction_recipient_notification', 'transaction_recipient_authentication',
                'transaction_recipient_verification', 'transaction_account_verification',
                'transaction_account_tokenization', 'transaction_recipient_payment']

if '_issuers' in key:
    keySubject = next(t + '_issuers' for t in subjectTypes if t in key)
elif '_field_results' in key:
    keySubject = next(t + '_field_results' for t in subjectTypes if t in key)
else:
    keySubject = next(t for t in subjectTypes if t in key)

print(f'BEFORE ===> {key}')
print(f'AFTER  ===> {keySubject}')

Answer 2

If you're looking for an approach to implement a switch statement, here's how I tackled it:

First create a helper function for the switch like this (I placed mine in a "myTools.py" module that I use in all my projects):

def switch(v): yield lambda *c:v in c

The way it works is by returning a function (lambda) that captured the switch value and returns True if the value is in any of its argument. The single yield makes switch() a generator usable in a for-loop. The for-loop will only perform one iteration but, given that it is a loop nonetheless, it will support break statements and the else clause at the end (when no breaks have been executed).

This allows a one iteration for-loop to be used in a form very similar to the switch statement in other languages (such as C++):

value = 5
for case in switch(value):               # case is actually a function
    if case(0):     print("None"); break
    if case(2,4,6): print("Even"); break
    if case(3,5,7): print("Odd");  break
else:
    print("invalid value")

With this approach you can make your switch function perform any kind of pattern matching:

def switchAllIn(v): yield lambda *c:all(s in v for s in c)

This version of the switch function checks if all case arguments are in the switch value:

value = 'The quick brown fox, jumped over the lazy dogs'

for case in switchAllIn(value):
    if case('quick','lazy'): print('oxymoron'); break
    if case('quick','fast'): print('redundant'); break
else:
    print('nothing special')

You can even make it use regular expressions:

def switchMatch(v): yield lambda *c:any(re.match(p,v) for p in c)         

value = 'The quick brown fox, jumped over the lazy dogs'
for case in switchMatch(value):
    if case(r'\bdog\b',r'\bfox\b') and not case('lazy'):
       print('lively canine')
       break
    if case(r'\bquick\b.+\bfox\b'):
       print('veloce vulpe')
       break

There is a lot of flexibility in this solution.

• you don't actually have to use break, so you can pass through multiple cases or even perform some logic between the if case(...): blocks

• you can combine cases with and/or operators such as if case(1) or case(3):

• when using breaks, you can nest the for case in switch without having to use a different name for the case function

for example:

for case in switch(letter):
    if case('a','b'):
        for case in switch(number):
            if case(1,2,3): print('vitamin'); break
            ...
        break
    if case('c','d'):
        ...
        break

• You can combine multiple switch calls with zip()

For example:

for caseL,caseN in zip(switch(L),switch(N)):
    if caseL('a','b') and caseN(1,2,3):
       print('vitamin')
       break
    ...

• You can apply the switch to every element of a list using map()

In which case the for-loop will run more than once and you must use continue instead of break:

L = [1,2,3,4,5]
for case in map(switch,L):
    if case(1,2,3):
       print('low')
       continue             # using continue instead of break 
    if case(4,5,6):
       print('medium')
       continue
    print('high')

For your specific scenario however, a switch statement is not necessarily the best solution. There seems to be a combinatory pattern of prefixes and suffixes with a limited set of keywords in between. You could use a loop on regular expressions composed of the prefixes, keywords and suffixes to get the keySubject:

import re
prefixes = ('transaction_recipient_','transaction_account_')
suffixes = ('_issuers','_field_results','') # in priority order
keywords = r'notification|authentication|verification|tokenization|payment'
for suffix in suffixes:
    for prefix in prefixes:
        pattern = r'\b('+prefix+keywords+suffix+')\b'
        m = re.match(pattern,key)
        if not m: continue
        keySubject = m.group()
        break
    else: continue; break
print(f'AFTER  ===> {keySubject}')