simple_salesforce python中的父子关系查询,从有序的dicts中提取

时间:2017-12-17 22:34:30

标签: python pandas salesforce soql ordereddictionary

我尝试使用python中的simple_salesforce包来查询来自salesforce的信息。

问题在于它的嵌套字段是父子关系的一部分,是有序字典中有序字典的一部分

我想要从Opportunity对象中找到id,以及与该记录关联的accountid。

SOQL查询可能看起来像..

query = "select id, account.id from opportunity where closedate = last_n_days:5"

在SOQL(salesforce对象查询语言)中,点表示数据库中的父子关系。所以我试图从机会对象中获取id,然后从该记录上的帐户对象获取相关的id。

由于某种原因,Id很好,但是account.id嵌套在有序字典中的有序字典中:

q = sf.query_all(query)

这会拉回有序字典..

OrderedDict([('totalSize', 455),
             ('done', True),
             ('records',
              [OrderedDict([('attributes',
                             OrderedDict([('type', 'Opportunity'),
                                          ('url',

我会拉records ordereddict来创建df

df = pd.DataFrame(q['records'])

这为我提供了3列,一个名为'attributes'的有序字典,Id和另一个名为'Account'的有序字典。我正在寻找一种从嵌套的有序字典('BillingCountry', 'United States')

中提取'Account'段的方法
[OrderedDict([('attributes',
               OrderedDict([('type', 'Opportunity'),
                            ('url',
                             '/services/data/v34.0/sobjects/Opportunity/0061B003451RhZgiHHF')])),
              ('Id', '0061B003451RhZgiHHF'),
              ('Account',
               OrderedDict([('attributes',
                             OrderedDict([('type', 'Account'),
                                          ('url',
                                           '/services/data/v34.0/sobjects/Account/001304300MviPPF3Z')])),
                            ('BillingCountry', 'United States')]))])

编辑:澄清我正在寻找的东西。

我想以一个数据框结束,每个查询字段都有一列。

当我使用'records'df = pd.DataFrame(sf.query_all(query)['records'])部分放入DataFrame时,它会给我:

attributes  Id  Account
OrderedDict([('type', 'Opportunity'), ('url', '/services/data/v34.0/sobjects/Opportunity/0061B003451RhZgiHHF')])    0061B003451RhZgiHHF OrderedDict([('attributes', OrderedDict([('type', 'Account'), ('url', '/services/data/v34.0/sobjects/Account/0013000000MvkRQQAZ')])), ('BillingCountry', 'United States')])
OrderedDict([('type', 'Opportunity'), ('url', '/services/data/v34.0/sobjects/Opportunity/0061B00001Pa52QQAR')]) 0061B00001Pa52QQAR  OrderedDict([('attributes', OrderedDict([('type', 'Account'), ('url', '/services/data/v34.0/sobjects/Account/0011300001vQPxqAAG')])), ('BillingCountry', 'United States')])
OrderedDict([('type', 'Opportunity'), ('url', '/services/data/v34.0/sobjects/Opportunity/0061B00001TRu5mQAD')]) 0061B00001TRu5mQAD  OrderedDict([('attributes', OrderedDict([('type', 'Account'), ('url', '/services/data/v34.0/sobjects/Account/0011300001rfRTrAAE')])), ('BillingCountry', 'United States')])

删除'attributes'列后,我希望输出为

Id BillingCountry
0061B003451RhZgiHHF 'United States'
0061B00001Pa52QQAR 'United States'
0061B00001TRu5mQAD 'United States'

2 个答案:

答案 0 :(得分:5)

Pandas是表格数据的绝佳工具。但它虽然可以包含Python对象,但这不是它的最佳点。我建议您在将查询数据插入pandas.Dataframe

之前从查询中提取数据

提取记录:

要将所需字段提取为字典列表,就像这样简单:

records = [dict(id=rec['Id'], country=rec['Account']['BillingCountry'])
           for rec in data['records']]

将记录插入数据框:

使用dicts列表,数据框就像以下一样简单:

df = pd.DataFrame(records)

测试代码:

import pandas as pd
from collections import OrderedDict

data = OrderedDict([
    ('totalSize', 455),
    ('done', True),
    ('records', [
        OrderedDict([
            ('attributes', OrderedDict([('type', 'Opportunity'), ('url', '/services/data/v34.0/sobjects/Opportunity/0061B003451RhZgiHHF')])),
            ('Id', '0061B003451RhZgiHHF'),
            ('Account', OrderedDict([('attributes', OrderedDict([('type', 'Account'), ('url', '/services/data/v34.0/sobjects/Account/0013000000MvkRQQAZ')])),
                                     ('BillingCountry', 'United States')])),
        ]),
        OrderedDict([
            ('attributes', OrderedDict([('type', 'Opportunity'), ('url', '/services/data/v34.0/sobjects/Opportunity/0061B00001Pa52QQAR')])),
            ('Id', '0061B00001Pa52QQAR'),
            ('Account', OrderedDict([('attributes', OrderedDict([('type', 'Account'), ('url', '/services/data/v34.0/sobjects/Account/0011300001vQPxqAAG')])),
                                     ('BillingCountry', 'United States')])),
        ]),
        OrderedDict([
            ('attributes', OrderedDict([('type', 'Opportunity'), ('url', '/services/data/v34.0/sobjects/Opportunity/0061B00001TRu5mQAD')])),
            ('Id', '0061B00001TRu5mQAD'),
            ('Account', OrderedDict([('attributes', OrderedDict([('type', 'Account'), ('url', '/services/data/v34.0/sobjects/Account/0011300001rfRTrAAE')])),
                                     ('BillingCountry', 'United States')])),
        ]),
    ])
])

records = [dict(id=rec['Id'], country=rec['Account']['BillingCountry'])
           for rec in data['records']]
for r in records:
    print(r)

print(pd.DataFrame(records))

测试结果:

{'country': 'United States', 'id': '0061B003451RhZgiHHF'}
{'country': 'United States', 'id': '0061B00001Pa52QQAR'}
{'country': 'United States', 'id': '0061B00001TRu5mQAD'}

         country                   id
0  United States  0061B003451RhZgiHHF
1  United States   0061B00001Pa52QQAR
2  United States   0061B00001TRu5mQAD

答案 1 :(得分:2)

熊猫可以阅读命令字典。

import pandas as pd
from simple_salesforce import Salesforce

sf = Salesforce(username='your_username',   
                password='your_password',
                security_token='your_token')

query = "select id, account.id from opportunity where closedate = last_n_days:5"
df = pd.DataFrame(sf.query_all(query)['records']).drop(columns='attributes')