从csv文件创建字典，其中特定列是键，行是Python中的值

Question

我有一个csv文件，格式如下:

jersey number,   position,   name,   birth_date,    birth_city,   years_in_nba,   team
23           ,     SF    ,   Lebron,  12/30/84 ,     Akron    ,      19      ,   Lakers
30           ,     PG    ,   Curry,   03/14/88 ,     Akron    ,      13      ,   Warriors
34           ,     PF    ,   Giannis, 08/26/89 ,     Athens    ,      8      ,   Bucks

我的目标是能够按位置分组行，其中键是不同的位置，并且值是具有相同位置的球员信息列表，如下所示:

{
"SF": [{..player1..}, {..player2..}],
"PF": [{..player1..}, {..player2..}],
"SG": [{..player1..}, {..player2..}],
"PG": [{..player1..}, {..player2..}],
}

这是我目前为止写的:

positions = {}
def players_position():
with open(filename, 'r') as file_obj:
dict_reader = DictReader(file_obj, delimiter=",")
for row in dict_reader:
positions[row["position"]] = row
return positions
print(players_position())

的代码不变，当前值将始终覆盖前一个值。我不想那样。我想有一个列表，我们可以继续添加播放器信息，如上面所示。

Answer 1

给你。我最初尝试以您所拥有的为基础，但由于在CSV原始数据中分割的数据最终以一堆空白填充结束，因此我做了一些额外的工作来清理键/值。

import csv
CSV = """
# jersey number,   position,   name,   birth_date,    birth_city,   years_in_nba,   team
# 23           ,     SF    ,   Lebron,  12/30/84 ,     Akron    ,      19      ,   Lakers
# 30           ,     PG    ,   Curry,   03/14/88 ,     Akron    ,      13      ,   Warriors
# 34           ,     PF    ,   Giannis, 08/26/89 ,     Athens    ,      8      ,   Bucks
"""

def main():
players = [player for player in csv.DictReader(CSV.splitlines())]
# gives me:
"""
{None: ['# jersey number', '   position', '   name', '   birth_date', '    birth_city', '   years_in_nba', '   team']}
{None: ['# 23           ', '     SF    ', '   Lebron', '  12/30/84 ', '     Akron    ', '      19      ', '   Lakers']}
{None: ['# 30           ', '     PG    ', '   Curry', '   03/14/88 ', '     Akron    ', '      13      ', '   Warriors']}
{None: ['# 34           ', '     PF    ', '   Giannis', ' 08/26/89 ', '     Athens    ', '      8      ', '   Bucks']}
"""
# strip the keys because extra whitespace, and because of how DictReader makes arrays, we have to do this:
keys = [k.strip() for k in [v for v in players.pop(0).values()][0]]
# then let's gather the players and strip the values so they don't have unnecessary spaces
players = [[p.strip() for p in list(p.values())[0]] for p in [v for v in players]]
# and combine (zip) the keys and values to make a dictionary
players_list = [dict(zip(keys, p)) for p in players]
# this gives me:
"""
{'# jersey number': '# 23', 'position': 'SF', 'name': 'Lebron', 'birth_date': '12/30/84', 'birth_city': 'Akron', 'years_in_nba': '19', 'team': 'Lakers'}
{'# jersey number': '# 30', 'position': 'PG', 'name': 'Curry', 'birth_date': '03/14/88', 'birth_city': 'Akron', 'years_in_nba': '13', 'team': 'Warriors'}
{'# jersey number': '# 34', 'position': 'PF', 'name': 'Giannis', 'birth_date': '08/26/89', 'birth_city': 'Athens', 'years_in_nba': '8', 'team': 'Bucks'}
"""
# last, create a dictionary of lists grouping by position using a list comprehension
# you could do this in a for loop, but we're writing python, after all...
players_by_position = {
position: [
player for player in players_list if player['position'] == position
] for position in set([player['position'] for player in players_list])
}

return players_by_position
"""
{
'SF': [
{
'# jersey number': '# 23', 
'position': 'SF', 
'name': 'Lebron', 
'birth_date': '12/30/84', 
'birth_city': 'Akron', 
'years_in_nba': '19', 
'team': 'Lakers'
}
], 
'PG': [
{
'# jersey number': '# 30', 
'position': 'PG', 
'name': 'Curry', 
'birth_date': '03/14/88', 
'birth_city': 'Akron', 
'years_in_nba': '13', 
'team': 'Warriors'
}
], 
'PF': [
{
'# jersey number': '# 34', 
'position': 'PF', 
'name': 'Giannis', 
'birth_date': '08/26/89', 
'birth_city': 'Athens', 
'years_in_nba': '8', 
'team': 'Bucks'
}
]
}
"""
if __name__ == "__main__":
print(main())