Question

对于创建数据库的项目，我想使用python将.json文件转换为.sqlite3文件（目前在Windows 10上运行Python 3.6.4）。下面是用于读取json文件的代码

...

with open('C:/Documents/{}/Posts.json'.format(forum), encoding="utf8") as f:
            row = json.load(f)
            parent_id = row['_Id']
            body = format_data(row['_Body'])
            score = row['_Score']
            comment_id = row['_Id']
            comment_id_type = row['_PostTypeId']
            parent_id_type = row['_PostTypeId']
            accepted_answer_id = row['_AcceptedAnswerId']
            accepted_parent_id = row['_ParentId']

            ...

运行此代码时遇到此错误。

File "C:\Python\data base.py", line 85, in <module>
parent_id = row['_Id']
KeyError: '_Id'

我已经读过这个错误发现，根据官方python文档，异常KeyError 是

在现有密钥集中找不到映射（字典）密钥时引发。

现在我无法理解这种语法，因为json文件中存在'_Id'（如下所示）

{
   "posts": {
      "row": [
         {
            "_Id": "1",
            "_PostTypeId": "1",
            "_AcceptedAnswerId": "3",
            "_CreationDate": "2016-08-02T15:39:14.947",
            "_Score": "5",
            "_ViewCount": "254",
            "_Body": "<p>What does \"backprop\" mean? I've Googled it, but it's showing backpropagation.</p>\n\n<p>Is the \"backprop\" term basically the same as \"backpropagation\" or does it have a different meaning?</p>\n",
            "_OwnerUserId": "8",
            "_LastEditorUserId": "7488",
            "_LastEditDate": "2017-05-28T13:48:02.003",
            "_LastActivityDate": "2017-05-28T13:48:02.003",
            "_Title": "What is \"backprop\"?",
            "_Tags": "<neural-networks><definitions><terminology>",
            "_AnswerCount": "3",
            "_CommentCount": "3"
         },

（这是来自AI的json：stackexchange数据）

我请求有人帮我解决我的KeyError问题，因为我搜索过的其他来源无法帮助我

请提前谢谢你。

Answer 1

首先，您必须访问"posts"

with open('C:/Documents/{}/Posts.json'.format(forum), encoding="utf8") as f:
    j = json.load(f)
    for row in j['posts']['row']:
        parent_id = row['_Id']
        body = format_data(row['_Body'])
        # ...

Answer 2

当您请求字典中不存在的密钥时，会引发KeyError。在你的情况下，从json开始，你似乎必须像这样访问它，

<强> json['posts' ]['row'][0].

帖子是一个词典。 row是一个dicts列表。列表是有序的，这就是我们可以索引它的原因。

完整代码：

with open('C:/Documents/{}/Posts.json'.format(forum), encoding="utf8") as f:
            jsondict = json.load(f)

            # Remember, posts > row > first_index
            row = jsondict['posts']['row'][0]
            parent_id = row['_Id']
            body = format_data(row['_Body'])
            score = row['_Score']
            comment_id = row['_Id']
            comment_id_type = row['_PostTypeId']
            parent_id_type = row['_PostTypeId']
            accepted_answer_id = row['_AcceptedAnswerId']
            accepted_parent_id = row['_ParentId']
            ...

使用python读取JSON：KeyError

2 个答案: