Python - List of tuples into oddly shaped other list of tuples

Question

Confused as to how to create this final part of my function.

The function below needs to read in data, start year and end year.

Consider the following:

def summary_statistics(data, year_start, year_end):
    earthquake_count_by_year = []
    total_damages_by_year = []
    casualties_by_year = []
    year_count = {}
    dmg_sum = {}
    casualties = {}
    
    years = []
    year_start = int(year_start)
    year_end = int(year_end)
    
    if year_end >= year_start:
        
        #store year range into list
        years = list(range(year_start, year_end+1))
        
        for index, tuple in enumerate(data):
            #values to be summed for each year in dmg_sum
            yr = tuple[0]
            dmgs = tuple[9]
            deaths = tuple[6]
            missing = tuple[7]
            injured = tuple[8]
            
            #if year in range of years (year_start - year_end)
            if tuple[0] in years:
                #if year does not exist in dict, set it to 0
                if tuple[0] not in year_count:
                    year_count[tuple[0]] = 0
                #count number of occurences for each year
                year_count[tuple[0]] += 1
                
                if tuple[0] not in dmg_sum:
                    dmg_sum[tuple[0]] = dmgs
                else:
                    dmg_sum[tuple[0]] += dmgs
                    
                if tuple[0] not in casualties:
                    casualties[tuple[0]] = list(deaths + ',' + missing + ',' + injured)
                else:
                    casualties[tuple[0]] += list(deaths + ',' + missing + ',' + injured)

        earthquake_count_by_year = list(year_count.items())
        total_damages_by_year = list(dmg_sum.items())
        casualties_by_year = list(casualties.items())

    
    L = [[earthquake_count_by_year], [total_damages_by_year], [casualties_by_year]]
    print(L)
    return L

The part I am having trouble with is:

                if tuple[0] not in casualties:
                    casualties[tuple[0]] = list(deaths + ',' + missing + ',' + injured)
                else:
                    casualties[tuple[0]] += list(deaths + ',' + missing + ',' + injured)

This is the expected output I need for casualties_by_year[]:

[(2020, (deaths, missing, injured)), (2019, (deaths, missing, injured)), (2018, (deaths, missing, injured))]

So what I'm trying to do is construct a list of tuples, in the arrangement shown above, for the final list going into L. it's tuple[0] for year, tuple[6] for deaths, tuple[7] for missing, tuple[8] for injured.

How can I correct that final if/else to get the list of tuples I'm looking for?

List comprehension?

EDIT- This is what "data" looks as it's input into the function:

[(2020, 1, 6.0, 'CHINA:  XINJIANG PROVINCE', 39.831, 77.106, 1, 0, 2, 0), (2020, 1, 6.7, 'TURKEY:  ELAZIG AND MALATYA PROVINCES', 38.39, 39.081, 41, 0, 1600, 0), (2020, 1, 7.7, 'CUBA: GRANMA;  CAYMAN IS;  JAMAICA', 19.44, -78.755, 0, 0, 0, 0), (2020, 2, 6.0, 'TURKEY: VAN;  IRAN', 38.482, 44.367, 10, 0, 60, 0), (2020, 3, 5.4, 'BALKANS NW:  CROATIA:  ZAGREB', 45.897, 15.966, 1, 0, 27, 6000.0)]

Do you have a test case that produces the wrong answer? [mcve] — Kenny Ostrom, Nov 03 '20 at 22:24
If you just put in the effort to write the data, I can help more, but since you didn't, I'm just going to tell you the short answer here in a comment: You need to make a tuple with normal python syntax: tuple(stuff) ... except you can't because you masked the type name with your own variable named "tuple" Don't do that. Rename your variable. If you give me a data dict with valid syntax I'll fill out the answer a little more. — Kenny Ostrom, Nov 03 '20 at 22:37
@kennyOstrom my apologies, Can't believe I forgot the data input. The data is input as a list of tuples, it's not a dict. I just updated my question with the input. — boog, Nov 04 '20 at 01:44

Kenny Ostrom · Accepted Answer · 2020-11-04T02:57:27.163

If you want a tuple, make a tuple. It's that simple. You can still use parentheses notation normally, but sometimes you need the constructor.

But first, never name a variable the same as some important python type. You want to create an instance of type tuple, but you have a variable named tuple, so now "tuple" means that local variable, and you can't access the tuple constructor, because you've hidden its name. I renamed it to "year_data"

Now, when you add the tuples together, that gets slightly more interesting. See Python element-wise tuple operations like sum

The rest is already there. You just have to use it. I deleted some irrelevant stuff.

import operator

data = [
    (2020, 1, 6.0, 'CHINA:  XINJIANG PROVINCE', 39.831, 77.106, 1, 0, 2, 0), 
    (2020, 1, 6.7, 'TURKEY:  ELAZIG AND MALATYA PROVINCES', 38.39, 39.081, 41, 0, 1600, 0), 
    (2020, 1, 7.7, 'CUBA: GRANMA;  CAYMAN IS;  JAMAICA', 19.44, -78.755, 0, 0, 0, 0), 
    (2020, 2, 6.0, 'TURKEY: VAN;  IRAN', 38.482, 44.367, 10, 0, 60, 0), 
    (2020, 3, 5.4, 'BALKANS NW:  CROATIA:  ZAGREB', 45.897, 15.966, 1, 0, 27, 6000.0)
]

def summary_statistics(data):
    casualties = {}
    for index, year_data in enumerate(data):
        year = year_data[0]
        deaths = year_data[6]
        missing = year_data[7]
        injured = year_data[8]
        casualties_data = (deaths, missing, injured)

        if year not in casualties:
            casualties[year] = casualties_data
        else:
            # https://stackoverflow.com/a/497894/1766544
            casualties[year] = tuple(map(operator.add, casualties[year], (deaths, missing, injured)))

    casualties_by_year = list(casualties.items())
    return casualties_by_year
    
result = summary_statistics(data)
print(result)

[(2020, (53, 0, 1689))]

Thank you so much Kenny. Really helped me out with this one. — boog, Nov 04 '20 at 05:48

Python - List of tuples into oddly shaped other list of tuples

1 Answers1