Why List Comprehension with groupby get weird results?

Python Help
1 
from itertools import groupby
dic = [{"name" : "A", "age": 15}, {"name" : "B", "age": 15},
 {"name" : "C", "age": 25}, {"name" : "D", "age": 25}]
print(len([(p, list(ps)) for _, ps in groupby(dic, key = lambda x: x["age"]) for p in ps])) #2
print(len([(p, ps) for _, ps in groupby(dic, key = lambda x: x["age"]) for p in ps])) #4

dic_g = groupby(dic, key = lambda x: x["age"])
print(len([(p, list(ps)) for _, ps in dic_g for p in ps])) #2
print(len([(p, ps) for _, ps in dic_g for p in ps])) # 0

dic_g = groupby(dic, key = lambda x: x["age"])
print(len([(p, ps) for _, ps in dic_g for p in ps])) #4
print(len([(p, list(ps)) for _, ps in dic_g for p in ps])) #0

dic_g = [(k, list(v)) for k, v in groupby(dic, key = lambda x: x["age"])]
print(len([(p, ps) for _, ps in dic_g for p in ps])) #4
print(len([(p, list(ps)) for _, ps in dic_g for p in ps])) #4

matrix = [(1, 2, 3), (4, 5, 6), (7, 8, 9)]
print(len([(e, list(r)) for r in matrix for e in r])) #9
print(len([(e, r) for r in matrix for e in r])) #9

I have use List Comprehension with groupby, but it will get weird results. It’s such behavior expected?

2

Yes, all of these look expected.

The central issue you are missing is that iterating through an object can be a destructive operation, and therefore iterating through it twice will not give you the same result. groupby, and all the “values” generators it yields are such object.

a = groupby([1,2,3])
for k, vs in a:
    print("First loop:", k, list(vs)) # runs three times

for k, vs in a:
    print("Second loop:", k, list(vs)) # doesn't run at all
2 Likes