How to group same values of a particular key in a list of dictionary together

Question

list = [
{'roll':101 , 'class':10},
{'roll':102 , 'class':10},
{'roll':103 , 'class':10},
{'roll':104 , 'class':11},
{'roll':105 , 'class':11},
{'roll':106 , 'class':11},
{'roll':107 , 'class':12},
{'roll':108 , 'class':12},
{'roll':109 , 'class':12} ]

Should be converted to

new_list = [
{'roll':[101,102,103] , 'class':10},
{'roll':[104,105,106] , 'class':11},
{'roll':[107,108,109] , 'class':12} ]

The initial list values will be from database . The values of roll will always be different where as class can be same . Cannot seem to figure out how to proceed.

Answer 1

This solution iterates through the list from the database (you shouldn't use list as a variable name), checks for existance in the new list and either appends to the roll list it exists, or creates the new dictionary.

db_list = [
{'roll':101 , 'class':10},
{'roll':102 , 'class':10},
{'roll':103 , 'class':10},
{'roll':104 , 'class':11},
{'roll':105 , 'class':11},
{'roll':106 , 'class':11},
{'roll':107 , 'class':12},
{'roll':108 , 'class':12},
{'roll':109 , 'class':12} ]

new_list = []
for item in db_list:
    for new_item in new_list:
        if new_item['class'] == item['class']:
            new_item['roll'].append(item['roll'])
            break
    else:
        new_list.append({'roll': [item['roll']], 'class': item['class']})

print(new_list)
# output 
# [{'roll': [101, 102, 103], 'class': 10}, {'roll': [104, 105, 106], 'class': 11}, {'roll': [107, 108, 109], 'class': 12}]

Alternatively, you can use an intermediate dictionary to build the new lists, which significantly improves the code readiblity,

from collections import defaultdict
class_roll = defaultdict(list)
for item in db_list:
    class_roll[item['class']].append(item['roll'])

new_list = [{'roll':class_roll[class_], 'class':class_} for class_ in class_roll]