如何追加 JSON 嵌套数组？

Question

在相当长的 Python 程序的一部分中，我有以下初始 JSON 对象：

{
  "Topics": {
    "Topic": [
      {
        "Topic Id": "",
        "Topic Title": "A short title that describes this topic",
        "Conversations": {
          "Conversation": [
            {
              "Conversation Id": "",
              "Conversation Title": "",
              "Conversation Text": ""
            }
          ]
        }
      }
    ]
  }
}

我正在迭代一系列输入并创建初始条目，如下所示：

newtopic = {
                "Topic Id": topicId,
                "Topic Title": json_data.get("Topic Title", None),
                "Conversations": json.dumps(Conversations[thisConversation])
           }

这似乎有效（尽管没有“对话”标签），但我不知道如何附加到对话数组。

# None of these work
if topicId == currentTopic["Topic Id"]:
  currentTopic["Conversations"].update(json.dumps(Conversations[thisConversation]))

if topicId == currentTopic["Topic Id"]:
  currentTopic["Conversations"].append(json.dumps(Conversations[thisConversation]))

if topicId == currentTopic["Topic Id"]:
  currentTopic["Conversations"].extend(json.dumps(Conversations[thisConversation]))

我得到同样的错误：AttributeError：'str'对象没有属性'append' - 但'append'是我尝试的任何内容（例如扩展，更新）。

对话列表是根据远程 URL 调用动态生成的，我需要通过类似的调用发送主题列表结果。为此，我希望尽可能避免自定义 Python 类。

json.dumps 调用返回在 Conversation 属性下找到的 JSON 字符串。创建后，我尝试将其他 JSON 字符串附加到 Conditions 元素。

有演员阵容或者我应该做的事情吗？

Answer 1

当提出这样的问题时，您应该包含允许人们实际重现您所看到的问题的代码。目前这个问题需要我们进行大量猜测。更重要的是，因为您似乎误解了如何在 Python 中处理 JSON 的几个重要部分。

没关系，你来这里是为了学习，但你不能指望人们凭直觉理解你的代码。这是一个尝试：

import json


initial_json_text = """
{
  "Topics": {
    "Topic": [
      {
        "Topic Id": "",
        "Topic Title": "A short title that describes this topic",
        "Conversations": {
          "Conversation": [
            {
              "Conversation Id": "",
              "Conversation Title": "",
              "Conversation Text": ""
            }
          ]
        }
      }
    ]
  }
}
"""

# this loads the object you probable have, which is a `dict` in Python
initial_json_object = json.loads(initial_json_text)

# getting the first topic, you apparently assigned some topic to json_data
json_data = initial_json_object["Topics"]["Topic"][0]

# you have *some* topic id - note that you should consider renaming this topic_id
topicId = json_data["Topic Id"]

# you reference some `Coversations` - the name suggests a class due to the capital
# but from the context it seems to be the conversations of some topic
Conversations = json_data["Conversations"]

# again, it's unclear where you get this from
thisConversation = 'Conversation'

# how you say you create a new topic
newtopic = {
    # the topic is assigned the topicId, presumable a string
    "Topic Id": topicId,
    # the title is assigned the title from json_data, presumably another topic
    "Topic Title": json_data.get("Topic Title", None),
    # here is your issue - json.dumps turns a Python object into a JSON string
    # so, you're assigning a string to the Conversations key
    "Conversations": json.dumps(Conversations[thisConversation])
}

# what you probably wanted
newtopic = {
    # the topic is assigned the topicId, presumable a string
    "Topic Id": topicId,
    # the title is assigned the title from json_data, presumably another topic
    "Topic Title": json_data.get("Topic Title", None),
    # here is your issue - json.dumps turns a Python object into a JSON string
    # so, you're assigning a string to the Conversations key
    "Conversations": Conversations[thisConversation]
}

# now this will add the conversation a second time, without error
newtopic["Conversations"].append(Conversations[thisConversation][0])

# you only want to use `.dumps()` when you need the JSON object represented
# as a string, for example when writing to a file
# for example, to add the topic to the original data, and write the result:
initial_json_object["Topics"]["New Topic"] = newtopic
result_json_text = json.dumps(newtopic)

# Note that you could print json_data as well, but that's because Python knows
# how to represent a `dict` as a string, if you're writing it to a file, you
# may want to have the string representation of the JSON object before writing
print(result_json_text)

请注意，您的数据结构有几个奇怪的特征，这可能是错误。

首先，您有

"Topics"

，它是一本由

"Topic"

等字符串索引的字典，但是字典中的每个主题都有一个

"Topic Id"

- 这可能只是字典中键的重复项？

类似地，您有

"Conversations"

这是一本字典，它似乎以

"Conversation"

之类的字符串作为键，但这种“对话”的实际值似乎是实际对话的 list ，带有单独的标识符。您是否打算将

"Conversations"

制作为一个列表，而花括号是错误的吗？

您调用

.dumps()

的次数似乎也远远超过了需要的次数，也许您不太明白它的作用 - 但这是您直接遇到的错误的核心，尽管您所做的选择可能会出现其他问题在文档结构中。

这样的事情会更有意义：

import json


initial_json_text = """
{
  "Topics": {
    "Topic 1": {
      "Topic Title": "A short title that describes this topic",
      "Conversations": {
        "Conversation": {
          "Conversation Title": "",
          "Conversation Text": ""
        },
        "Another Conversation": {
          "Conversation Title": "",
          "Conversation Text": ""
        }
      }
    }
  }
}
"""
initial_json_object = json.loads(initial_json_text)

# "Topic 1" is the key for the first topic, and serves as the ID
topic = initial_json_object["Topics"]["Topic 1"]