Matching partial cases to Python dictionary

Question

At first glance I thought this to be a simple issue yet I cannot find an answer that accurately fits...

I have a dictionary of state names and abbreviations like so;

{(' ak', ',ak', ', ak', 'juneau', ',alaska', ', alaska'): 'alaska',
 (' al', ',al', ', al', 'montgomery', ',alabama', ', alabama'): 'alabama',
 (' ar', ',ar', ', ar', 'little rock', ',arkansas', ', arkansas'): 'arkansas',
 (' az', ',az', ', az', 'phoenix', ',arizona', ', arizona'): 'arizona',

I am attempting to map this dictionary over various cases of self reported Twitter location that I have in a pandas dataframe to look for partial matches. For example, if one case read 'anchorage,ak' it would change the value to Alaska. I could see this being quite simple if it were a list, yet there must be another way to do this without looping. Any help is greatly appreciated!

Answer 1

I think timgeb has the right idea above. I would add two things:

1) You can also remove all whitespace from the given case before processing-- thus, there will be no need to include ' ak' , ',ak' , and ', ak' all as keys-- a simple 'ak' key would suffice.

2) Instead of repeating the state values in the dictionary, I would create an extra hash from integers to states ie {0: 'alaska, 1: 'alabama' ...} and store the corresponding integer key in your original dictionary.

Thus your resulting dictionary should look something like this:

A = {'ak': 0, 'juneau': 0, 'alaska': 0, 'al': 1, 'montgomery': 1, 'alabama': 1, ...}

And to access state names from integer values, you should have another dictionary like this for all 50 states:

B = {0: 'alaska', 1: 'alabama',  ...}

so given a case...

case = 'anchorage,ak'
case_list = case.replace(' ', '').split(',')  # remove all whitespace and split case by comma
for elem in case_list:
    if elem in A:
        # insert code to replace case with B[A[elem]]
        break