Source: 📖 Python Cookbook ch1.15 p25

Grouping objects together based on a field with itertools.groupby()

If you have a sequence of objects that you would like to be grouped by a particular field or attribute, the gorupby() function from itertools is a great fit. There is a preliminary step before handing the objects over to groupby()—they must be sorted by the desired field or attribute, as groupby() works by iterating through the sequence and grouping together sequential runs that contain identical data in a desired field.

from itertools import groupby


meals_list = [
	{'bfast': 'cereal', 'date': '2021-04-16'},
	{'lunch': 'sandwich', 'date': '2021-04-16'},
	{'dinner': 'soup', 'date': '2021-04-16'},
	{'bfast': 'bacon & eggs', 'date': '2021-04-15'},
	{'lunch': 'sandwich', 'date': '2021-04-15'},
	{'dinner': 'chicken & potatoes', 'date': '2021-04-15'},
]
# already sorted by date, no need to sort in this case.

for date, meals in groupby(meals_list, key=lambda x: x['date']):
	print(date)
	for meal in meals:
		print('    ', meal)

2021-04-16
	{'date': '2021-04-16', 'bfast': 'cereal'}
	{'date': '2021-04-16', 'luncht': 'sandwich'}
	{'date': '2021-04-16', 'dinner': 'soup'}
2021-04-15
	{'date': '2021-04-15', 'bfast': 'bacon & eggs'}
	{'date': '2021-04-15', 'bfast': 'sandwich'}
	{'date': '2021-04-15', 'bfast': 'chicken & potatoes'}

On each iteration, the groupby() function returns the value it grouped by along with an iterator containing all the contents that matched the grouping criteria.