使用pandas标识JSON中的标题

我有这个JSON文件

*[
  {
    "name": "Kitty",
    "species" : "cat",
    "foods": {
      "likes": ["fresh food"],
      "dislikes": ["stale food"]
    }
  },
  {
    "name": "Pupster",
    "species" : "dog",
    "foods": {
      "likes": ["tomatoes", "peas"],
      "dislikes": ["bread"]
    }
  },
  {
    "name": "Tux",
    "species" : "cat",
    "foods": {
      "likes": ["fancy dishes"],
      "dislikes": ["basic cat food"]
    }
  }
]*

---

我是这样读这个文件的：

    content = requests.get("https://raw.githubusercontent.com/LearnWebCode/json-example/master/animals-3.json") 
data = json.loads(content.content)
df = pd.DataFrame(data)

---

但我想要的是只读取JSON文件头并将其存储在列表中。我有一个类似的代码

 getcolumns = list(df.columns.values)

但我只知道以下几点

• 名字
• 种类
• 食物

我希望实现以下目标：

• 名字
• 种类
• 食物
• 喜欢
• 不喜欢

---

我是新使用熊猫的，任何建议都可以接受

列表理解应该有效：

j = [
  {
    "name": "Kitty",
    "species" : "cat",
    "foods": {
      "likes": ["fresh food"],
      "dislikes": ["stale food"]
    }
  },
  {
    "name": "Pupster",
    "species" : "dog",
    "foods": {
      "likes": ["tomatoes", "peas"],
      "dislikes": ["bread"]
    }
  },
  {
    "name": "Tux",
    "species" : "cat",
    "foods": {
      "likes": ["fancy dishes"],
      "dislikes": ["basic cat food"]
    }
  }
]

getcolumns = [x for x in j[0]] + [x for x in j[0]['foods']]
# ['name', 'species', 'foods', 'likes', 'dislikes']

或者，如果需要数据帧：

from pandas.io.json import json_normalize
df = json_normalize(j)

     foods.dislikes       foods.likes     name species
0      [stale food]      [fresh food]    Kitty     cat
1           [bread]  [tomatoes, peas]  Pupster     dog
2  [basic cat food]    [fancy dishes]      Tux     cat

---

清单理解应该有效：

j = [
  {
    "name": "Kitty",
    "species" : "cat",
    "foods": {
      "likes": ["fresh food"],
      "dislikes": ["stale food"]
    }
  },
  {
    "name": "Pupster",
    "species" : "dog",
    "foods": {
      "likes": ["tomatoes", "peas"],
      "dislikes": ["bread"]
    }
  },
  {
    "name": "Tux",
    "species" : "cat",
    "foods": {
      "likes": ["fancy dishes"],
      "dislikes": ["basic cat food"]
    }
  }
]

getcolumns = [x for x in j[0]] + [x for x in j[0]['foods']]
# ['name', 'species', 'foods', 'likes', 'dislikes']

或者，如果需要数据帧：

from pandas.io.json import json_normalize
df = json_normalize(j)

     foods.dislikes       foods.likes     name species
0      [stale food]      [fresh food]    Kitty     cat
1           [bread]  [tomatoes, peas]  Pupster     dog
2  [basic cat food]    [fancy dishes]      Tux     cat

---

“喜欢”和“不喜欢”属于“食物”类别，如果你进一步将这两个分开，那么“食物”一栏中就不会有任何内容。该表将包含“名称”、“种类”、“喜欢”和“不喜欢”这两个词。“喜欢”和“不喜欢”属于“食品”类别，如果你进一步将这两个词分开，那么“食品”一栏中将没有任何内容。该表将包含“名称”、“种类”、“喜欢”和“不喜欢”