我知道我希望我的Python词典列表看起来像什么，但是在将电子表格数据拉入数据结构时遇到了问题。我遇到的问题是，一行可能有数据填充父字典值以及1个子字符。对于后续行，如果父级列中的值为空，则假定子级的列属于上一个父级。如果我们遇到父数据不为空的新行，请将其视为要添加到列表中的新父项。
这是电子表格的示例：
[
{
"name": "NTP_Policy1",
"descr": "NTP Policy 1",
"adminSt": "enabled",
"authSt": "disabled",
"servers": [
{
"hostname": "10.10.10.10",
"descr": "NTP1 Server",
"preferred": true,
"server_EPG": "oob-default",
"minPoll": 4,
"maxPoll": 6
},
{
"hostname": "20.10.10.10",
"descr": "NTP2 Server",
"preferred": false,
"server_EPG": "oob-default",
"minPoll": 4,
"maxPoll": 6
}
]
},
{
"name": "NTP_Policy2",
"descr": "NTP Policy 2",
"adminSt": "enabled",
"authSt": "disabled",
"servers": [
{
"hostname": "30.10.10.10",
"descr": "NTP3 Server",
"preferred": true,
"server_EPG": "oob-default",
"minPoll": 4,
"maxPoll": 6
},
{
"hostname": "40.10.10.10",
"descr": "NTP4 Server",
"preferred": false,
"server_EPG": "oob-default",
"minPoll": 4,
"maxPoll": 6
}
]
}
]
我希望数据结构如下所示：
>>> import pyexcel
>>> from pprint import pprint
>>> def excel_to_dict(sheet):
... rows = sheet.iter_rows()
... keys = next(rows)
... dict_list = []
... # For each row in the spreadsheet,
... # Create an iterator pair so that the key is iterated over at the same time as its matching cell in the row
... # Then save that pairing as descriptors of the switch
... for row in rows:
... dict = {}
... dict['servers'] = []
... server_atts = {}
... for key,cell in zip(keys, row):
... if str(cell.value) != 'None' and str(key.value) == 'name':
... dict[str(key.value)] = str(cell.value)
... parentKey = str(key.value)
... elif (str(cell.value) != 'None' and str(key.value) == 'descr') or (str(cell.value) != 'None' and str(key.value) == 'adminSt') or (str(cell.value) != 'None' and str(key.value) == 'authSt'):
... dict[str(key.value)] = str(cell.value)
... elif str(cell.value) == 'None':
... continue
... else:
... server_atts[str(key.value)] = str(cell.value)
... dict['servers'].append(server_atts.copy())
... dict_list.append(dict.copy())
... return dict_list
>>> wb = openpyxl.load_workbook('aci_config.xlsx')
>>> ntpPolsSheet = wb.get_sheet_by_name('ntp_pol')
>>> ntpPols = excel_to_dict(ntpPolsSheet)
>>>
>>> pprint(ntpPols)
[{'adminSt': 'enabled',
'authSt': 'disabled',
'descr': 'Test NTP Policy',
'name': 'test1-NTPPOL',
'servers': [{'server_EPG': 'oob-default',
'server_descr': 'NTP1 server',
'server_hostname_ip': '10.10.10.10',
'server_maxPoll': '6',
'server_minPol': '4',
'server_preferred': 'yes'}]},
{'servers': [{'server_EPG': 'oob-default',
'server_descr': 'NTP2 server',
'server_hostname_ip': '10.10.10.11',
'server_maxPoll': '6',
'server_minPol': '4',
'server_preferred': 'no'}]},
{'servers': [{'server_EPG': 'oob-default',
'server_descr': 'NTP3 server',
'server_hostname_ip': '10.10.10.12',
'server_maxPoll': '6',
'server_minPol': '4',
'server_preferred': 'no'}]},
{'adminSt': 'enabled',
'authSt': 'disabled',
'descr': 'Test 2 NTP policy',
'name': 'test2-NTPPOL',
'servers': [{'server_EPG': 'oob-default',
'server_descr': 'NTP1 server',
'server_hostname_ip': '20.10.10.10',
'server_maxPoll': '6',
'server_minPol': '4',
'server_preferred': 'yes'}]},
{'servers': [{'server_EPG': 'oob-default',
'server_descr': 'NTP2 server',
'server_hostname_ip': '20.10.10.11',
'server_maxPoll': '6',
'server_minPol': '4',
'server_preferred': 'no'}]},
{'servers': [{'server_EPG': 'oob-default',
'server_descr': 'NTP3 server',
'server_hostname_ip': '20.10.10.12',
'server_maxPoll': '6',
'server_minPol': '4',
'server_preferred': 'no'}]}]
我最接近的代码看起来像这样，但是后续行将子项附加到父级别。
$cars = array (
'expensive' => $BMW,
'medium' => $Volvo,
'cheap' => $Lada
);
foreach (array_keys($cars) as $index) {
echo "$index<br/>\n";
}
代码需要什么才能正确填充dict列表？是否有更好的电子表格格式可以更容易导入数据？我试图在一张纸上完成所有操作，而不是多张纸。
答案 0 :(得分：0)
我建议以csv格式保存.xlsx文件，因为它必须更容易处理。它将以文本形式显示：
name,descr,adminSt,authSt,server_hostname_ip,server_descr,server_preferred,server_EPG,server_minPoll,
test1-NTPPOL,Test NTP Policy,enabled,disabled,10.10.10.10,NTP1 server,yes,oob-default,4,6
,,,,10.10.10.11,NTP2 server,no,oob-default,4,6
,,,,10.10.10.12,NTP3 server,no,oob-default,4,6
test2-NTPPOL,Test 2 NTP policy,enabled,disabled,20.10.10.10,NTP1 server,yes,oob-default,4,6
,,,,20.10.10.11,NTP2 server,no,oob-default,4,6
,,,,20.10.10.12,NTP3 server,no,oob-default,4,6
然后你可以使用pandas来读取csv并将其转换为json。 Pandas有一个.iloc函数，它允许您先按行索引，然后按列名称进行索引。
import pandas as pd
from beeprint import pp
def excel_to_dict(sheet):
dict_list = []
last_test_dict = None
for i in xrange(len(sheet)):
# When we find a new row with a name value, we want to insert
# the old test_dict into the dict_list and make a new test_dict.
# Also, we want to skip the first row to not append an empty dict.
if pd.notnull(sheet.iloc[i]['name']):
if i != 0:
dict_list.append(test_dict)
test_dict = {}
test_dict['name'] = sheet.iloc[i]['name']
test_dict['descr'] = sheet.iloc[i]['descr']
test_dict['adminSt'] = sheet.iloc[i]['adminSt']
test_dict['authSt'] = sheet.iloc[i]['authSt']
test_dict['servers'] = []
server_info = {}
server_info['server_hostname'] = sheet.iloc[i]['server_hostname_ip']
server_info['server_descr'] = sheet.iloc[i]['server_descr']
server_info['server_preferred'] = sheet.iloc[i]['server_preferred']
server_info['server_EPG'] = sheet.iloc[i]['server_EPG']
server_info['minPoll'] = sheet.iloc[i]['server_minPoll']
server_info['maxPoll'] = sheet.iloc[i]['server_maxPoll']
test_dict['servers'].append(server_info)
last_test_dict = test_dict # keep a handle to our new dict
else:
# Use the handle to the last test dict created to add info
# about a new server without modifying the name of the test
server_info = {}
server_info['server_hostname'] = sheet.iloc[i]['server_hostname_ip']
server_info['server_descr'] = sheet.iloc[i]['server_descr']
server_info['server_preferred'] = sheet.iloc[i]['server_preferred']
server_info['server_EPG'] = sheet.iloc[i]['server_EPG']
server_info['minPoll'] = sheet.iloc[i]['server_minPoll']
server_info['maxPoll'] = sheet.iloc[i]['server_maxPoll']
last_test_dict['servers'].append(server_info)
# In case we didn't enter the last test dict into the list
dict_list.append(last_test_dict)
return dict_list
sheet = pd.read_csv('sheet.csv', sep=',')
pp(excel_to_dict(sheet))