查看“Pandas json”的源代码

通常情况下，我们使用的都是 pandas 中的 to_json() 函数，可以通过设置 orient 参数来转换成为我们想要的 json 格式，orient 函数有以下几个参数：
*Series可选参数为："index"（默认）, "split", "records"
*DataFrame可选参数："columns"(默认)，split"，"records"， "index"，"values"
----

=== data -> dataframe ===

==== data ====
area_proc.csv
{| class="wikitable"
!area_id!!area_name!!prod_type!!prod_name!!num!!unit!!total
|-
|10||PEK||P1||Apple||12||5.2||62.4
|-
|10||PEK||P2||Pear||20||3.3||66
|-
|11||SHA||P1||Apple||8||4.5||36
|-
|11||SHA||P2||Pear||15||2.5||37.5
|}

==== dataframe ====
 ...
 df1 = pd.read_csv(u'area_proc.csv')
    area_id area_name prod_type prod_name  num  unit  total
 0       10       PEK        P1     Apple   12   5.2   62.4
 1       10       PEK        P2      Pear   20   3.3   66.0
 2       11       SHA        P1     Apple    8   4.5   36.0
 3       11       SHA        P2      Pear   15   2.5   37.5

=== dataframe -> json ===
dataframe.to_json(orient="?", force_ascii=False)

?--> columns, split, records, index, values

====dataframe -> json : columns====
df1.to_json(orient="columns",force_ascii=False)

<nowiki>'{"area_id":{"0":10,"1":10,"2":11,"3":11},"area_name":{"0":"PEK","1":"PEK","2":"SHA","3":"SHA"},"prod_type":{"0":"P1","1":"P2","2":"P1","3":"P2"},"prod_name":{"0":"Apple","1":"Pear","2":"Apple","3":"Pear"},"num":{"0":12,"1":20,"2":8,"3":15},"unit":{"0":5.2,"1":3.3,"2":4.5,"3":2.5},"total":{"0":62.4,"1":66.0,"2":36.0,"3":37.5}}'</nowiki>


====dataframe -> json : split =====
df1.to_json(orient="split",force_ascii=False)

'{"columns":["area_id","area_name","prod_type","prod_name","num","unit","total"],"index":[0,1,2,3],"data":[[10,"PEK","P1","Apple",12,5.2,62.4],[10,"PEK","P2","Pear",20,3.3,66.0],[11,"SHA","P1","Apple",8,4.5,36.0],[11,"SHA","P2","Pear",15,2.5,37.5]]}'


====dataframe -> json : records =====
df1.to_json(orient="records",force_ascii=False)

'[{"area_id":10,"area_name":"PEK","prod_type":"P1","prod_name":"Apple","num":12,"unit":5.2,"total":62.4},{"area_id":10,"area_name":"PEK","prod_type":"P2","prod_name":"Pear","num":20,"unit":3.3,"total":66.0},{"area_id":11,"area_name":"SHA","prod_type":"P1","prod_name":"Apple","num":8,"unit":4.5,"total":36.0},{"area_id":11,"area_name":"SHA","prod_type":"P2","prod_name":"Pear","num":15,"unit":2.5,"total":37.5}]'


====dataframe -> json : index =====
df1.to_json(orient="index",force_ascii=False)

<nowiki>'{"0":{"area_id":10,"area_name":"PEK","prod_type":"P1","prod_name":"Apple","num":12,"unit":5.2,"total":62.4},"1":{"area_id":10,"area_name":"PEK","prod_type":"P2","prod_name":"Pear","num":20,"unit":3.3,"total":66.0},"2":{"area_id":11,"area_name":"SHA","prod_type":"P1","prod_name":"Apple","num":8,"unit":4.5,"total":36.0},"3":{"area_id":11,"area_name":"SHA","prod_type":"P2","prod_name":"Pear","num":15,"unit":2.5,"total":37.5}}'</nowiki>


====dataframe -> json : values =====
df1.to_json(orient="values",force_ascii=False)

'[[10,"PEK","P1","Apple",12,5.2,62.4],[10,"PEK","P2","Pear",20,3.3,66.0],[11,"SHA","P1","Apple",8,4.5,36.0],[11,"SHA","P2","Pear",15,2.5,37.5]]'


----

=== json -> dataframe ===
[[分类:Develop]]
[[分类:Python]]
[[分类:Pandas]]