我需要有关以以下格式展平嵌套字典的功能的帮助:
dict_test = {
"id" : "5d4c2c0fd89234260ec81",
"Reference Number" : "JA-L800D-191",
"entities_discovered" : {
"OTHER_ID" : [
"L800DFAG02191"
],
"CODE_ID" : [
"160472708",
"276954773"
]
},
"label_field" : [
"ELECTRONICS",
"HDMI"
],
"numeric_field" : [
491,
492
],
}
我正在使用的函数将字典按我的意愿展平到一个维度(键:值),但是没有在同一键迭代中将值连接在一起。
def flatten(d):
agg = {}
def _flatten(d, prev_key=''):
if isinstance(d, list):
for i, item in enumerate(d):
new_k = '%s.%s' % (prev_key, i) if prev_key else i
_flatten(item, prev_key=new_k)
elif isinstance(d, dict):
for k, v in d.items():
new_k = '%s.%s' % (prev_key, k) if prev_key else k
_flatten(v, prev_key=new_k)
else:
agg[prev_key] = d
_flatten(d)
return agg
我当前的输出是:
{
"id" : "5d4c2c0fd89234260ec81",
"Reference Number" : "JA-L800D-191",
"entities_discovered.OTHER_ID.0" : "L800DFAG02191",
"entities_discovered.CODE_ID.0" : "160472708",
"entities_discovered.CODE_ID.1" : "276954773",
"label_field.0" : "ELECTRONICS",
"label_field.1" : "HDMI",
"numeric_field.0" : 491,
"numeric_field.1" : 492
}
但是实际上我正在寻找类似的东西(将值连接到相同的字符串中,并用或|分隔):
{
"id" : "5d4c2c0fd89234260ec81",
"Reference Number" : "JA-L800D-191",
"OTHER_ID" : "L800DFAG02191",
"CODE_ID" : "160472708, 276954773",
"label_field" : "ELECTRONICS, HDMI",
"numeric_field" : ¨491, 492¨
}
答案 0 :(得分:2)
您可以使用join()
内置方法将值连接在一起。
def do():
dict_test = {
"id": "5d4c2c0fd89234260ec81",
"Reference Number": "JA-L800D-191",
"entities_discovered": {
"OTHER_ID": [
"L800DFAG02191"
],
"CODE_ID": [
"160472708",
"276954773"
]
},
"label_field": [
"ELECTRONICS",
"HDMI"
],
"numeric_field": [
491,
492
],
}
new_dict = {}
for key, value in dict_test.items():
if isinstance(value, dict):
for _key, _value in value.items():
if isinstance(_value, list):
new_dict.update({_key: ', '.join([str(item) for item in _value])})
elif isinstance(value, list):
new_dict.update({key: ', '.join([str(item) for item in value])})
else:
new_dict.update({key: value})
return new_dict
if __name__ == '__main__':
print(do())
输出:
{
'id': '5d4c2c0fd89234260ec81',
'Reference Number': 'JA-L800D-191',
'OTHER_ID': 'L800DFAG02191',
'CODE_ID': '160472708, 276954773',
'label_field': 'ELECTRONICS, HDMI',
'numeric_field': '491, 492'
}
答案 1 :(得分:0)
def recursive_flatten_dict(tmp, dict_test):
for i,v in dict_test.items():
if type(v) == type({}):
recursive_flatten_dict(tmp,v)
else:
tmp[i] = v
return tmp
recursive_flatten_dict({},dict_test)
答案 2 :(得分:0)
使用生成器的简单递归:
def flatten(d):
for a, b in d.items():
if isinstance(b, dict):
yield from flatten(b)
else:
yield (a, b if not isinstance(b, list) else ', '.join(map(str, b)))
print(dict(flatten(dict_test)))
输出:
{
'id': '5d4c2c0fd89234260ec81',
'Reference Number': 'JA-L800D-191',
'OTHER_ID': 'L800DFAG02191',
'CODE_ID': '160472708, 276954773',
'label_field': 'ELECTRONICS, HDMI',
'numeric_field': '491, 492'
}
答案 3 :(得分:0)
def flatten(dict_test):
for key in ['label_field', 'numeric_field']:
dict_test[key]= ', '.join([str(c) for c in dict_test[key]])
for c in dict_test['entities_discovered'].keys():
dict_test[c]= ', '.join(dict_test['entities_discovered'][c])
return dict_test
以上功能可以完成这项工作。我希望这是您想要的?