将多列csv文件读取到嵌套字典

Question

我正在尝试从csv文件读取并将数据存储在嵌套字典中。

CSV file content

Type, ID, Frequency
Advanced,AAA,30 secs
Advanced,AAA,60 secs
Advanced,BBB,30 secs
Basic,CCC,30 secs
Basic,CCC,60 secs
Basic,DDD,30 secs

Expected output where the 'type' is the higher level key with the values as another dictionary with the ID and frequency as the key/value pair.

{'Advanced': {'AAA':['30 secs', '60 secs'], 'BBB':['30 secs']}, 'Basic': {'CCC':['30 secs', '60 secs'], 'DDD':['30 secs']}}

我有两列，它使用defaultdict容器来工作。

symbols = co.defaultdict(list)
with open(filename, 'r') as f:
    lines = csv.DictReader(f)
    for line in lines:
        print(line)
        symbols[line['Type']].append(line['ID'])

Answer 1

您可以改用dict.setdefault ：

symbols = {}
with open(filename, 'r') as f:
    for row in csv.DictReader(f, skipinitialspace=True):
        symbols.setdefault(row['Type'], {}).setdefault(row['ID'], []).append(row['Frequency'])

symbols变为：

{'Advanced': {'AAA': ['30 secs', '60 secs'], 'BBB': ['30 secs']}, 'Basic': {'CCC': ['30 secs', '60 secs'], 'DDD': ['30 secs']}}

或者，如果您更喜欢使用collections.defaultdict ，则应使symbols成为列表的defaultdict的defaultdict ：

symbols = defaultdict(lambda: defaultdict(list))
for row in csv.DictReader(f, skipinitialspace=True):
    symbols[row['Type']][row['ID']].append(row['Frequency'])