合并/更新來自兩個不同json檔案的json物件

我有兩個 JSON 檔案，都有相同數量的物件，每個物件都有一個 ID 鍵"DOCN"，問題是有些物件有不同的鍵，例如在 file1 中，物件"DOCN": "000093019"有 4 個鍵，而在 file2 中，同一個物件有 5 ..

我正在嘗試創建一個包含兩個檔案中相同物件的新檔案（在 file1 和 file2 中找到丟失的物件并將它們添加到物件中）

例子：

檔案 1：

[
    {
        "DOCN": "000093019",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla"
    },
    {
        "DOCN": "000093085",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla"
    }
]

檔案2：

[
    {
        "DOCN": "000093019",
        "A": "blabla",
        "C": "blabla",
        "D": "blabla",
        "E": "blabla"
    },
    {
        "DOCN": "000093085",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla"
    }
]

我想實作的目標：file3：

[
    {
        "DOCN": "000093019",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla",
        "E": "blabla"
    },
    {
        "DOCN": "000093085",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla"
    }
]

uj5u.com熱心網友回復：

我會在兩個不同的陣列中讀取它們并將其映射以創建新的陣列。

// read file1 instead using `fs`
const arr1 = [
    {
        "DOCN": "000093019",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla"
    },
    {
        "DOCN": "000093085",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla"
    }
]
// read file2 instead
const arr2 = [
    {
        "DOCN": "000093019",
        "A": "blabla",
        "C": "blabla",
        "D": "blabla",
        "E": "blabla"
    },
    {
        "DOCN": "000093085",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla"
    }
]

const arr3 = arr1.map(
    x => {
      const val = arr2.find(y => y.DOCN === x.DOCN)
      x= {
        ...x,
        ...val
      }
      return x
    })

//write arr3 to new file
    ```

uj5u.com熱心網友回復：

我會按照以下方式進行 - 使用 Pandas、concat 資料幀、group by 加載 2 個檔案DOCN并獲取第一條記錄（這將獲取 none None 值），然后將其轉換為串列并洗掉 None 條目 -

df1 = pd.read_json("my_file1.json")
df2 = pd.read_json("my_file2.json")
df = pd.concat([df1, df2])
grp = df.groupby("DOCN").first().reset_index()
[{k: v for k, v in record if v} for record in grp.to_dict(orient='records')]

uj5u.com熱心網友回復：

嗯，這是對字典的簡單操作。我不能說這對于大型資料集的表現最好。但是您可以根據鍵“DOCN”合并字典。（可能有更好的方法！;-)）

f1 = [
    {
        "DOCN": "000093019",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla"
    },
    {
        "DOCN": "000093085",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla"
    }
]

f2 = [
    {
        "DOCN": "000093019",
        "A": "blabla",
        "C": "blabla",
        "D": "blabla",
        "E": "blabla"
    },
    {
        "DOCN": "000093085",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla"
    }
]

f1 = {item.get("DOCN"): item for item in f1}
f2 = {item.get("DOCN"): item for item in f2}

keys = set(list(f1.keys()) list(f2.keys()))

output = []
for key in keys:
    output.append({**f1.get(key), **f2.get(key)})

print(output)

輸出是：

[
    {
        "DOCN": "000093019",
        "A": "blabla",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla",
        "E": "blabla"
    },
    {
        "DOCN": "000093085",
        "B": "blabla",
        "C": "blabla",
        "D": "blabla",
        "A": "blabla"
    }
]

標籤：Python json 蟒蛇-3.x

上一篇：從陣列創建新的JSON字串

下一篇：在沒有服務器的情況下從HTMLHead讀取JSON檔案