在 Power BI 中解析 .json 列

Parsing a .json column in Power BI

我想通过 Power BI 解析 .json 列。我直接从服务器导入了数据,并且在数据中有一个 .json 列以及其他列。有没有办法解析这个 json 列?

示例:

       Key      IDNumber    Module      JsonResult  
       012      200         Dine        {"CategoryType":"dining","City":"mumbai"',"Location":"all"} 
       97       303         Fly         {"JourneyType":"Return","Origin":"Mumbai (BOM)","Destination":"Chennai (MAA)","DepartureDate":"20-Oct-2016","ReturnDate":"21-Oct-2016","FlyAdult":"1","FlyChildren":"0","FlyInfant":"0","PromoCode":""} 
       276      6303        Stay        {"Destination":"Clarion Chennai","CheckInDate":"14-Oct-2016","CheckOutDate":"15-Oct-2016","Rooms":"1","NoOfPax":"2","NoOfAdult":"2","NoOfChildren":"0"}

我希望保留其他列并获得简化的解析列。

import json
from urllib import urlopen
import string
from UserList import *
l=[]
j=[]
d_base=urlopen('https://api.thingspeak.com/channels/193888/fields/1.json?results=1')
data = json.load(d_base)
for k in data['feeds']:
        name = k['entry_id']
        value = k['field1']
        l.append(name)
        j.append(value)

print l[0]
print j[0]

**此 python 代码可能对您有用 ** **270 1035 **

像这样使用Json.Document函数

let
    ...
    your_table=imported_the_data_directly_from_the_server,
    json=Table.AddColumn(your_table, "NewColName", each Json.Document([JsonResult]))
in
    json

然后使用 Table.ExpandRecordColumn

将记录扩展到 table

或者点击这个按钮

有一种更简单的方法,在查询编辑器中您想要阅读的列上 json:

  • 右键单击列
  • Select变换>JSON

然后该列变成一条记录,您可以使用右上角的按钮将其拆分为 json 的每个 属性。

使用Json.Document()函数将字符串转换为Json数据。

let
    Source = Json.Document(Json.Document(Web.Contents("http://localhost:18091/pools/default/buckets/Aggregation/docs/AvgSumAssuredByProduct"))[json]),
    #"Converted to Table" = Record.ToTable(Source),
    #"Filtered Rows" = Table.SelectRows(#"Converted to Table", each not Text.Contains([Name], "type_")),
    #"Renamed Columns" = Table.RenameColumns(#"Filtered Rows",{{"Name", "AvgSumAssuredByProduct"}}),
    #"Changed Type" = Table.TransformColumnTypes(#"Renamed Columns",{{"Value", type number}})
in
    #"Changed Type"