使用 Python 从 json API 响应中提取属性

Question

我正在使用此代码调用 NY Times API 并获取有关所选搜索查询文章的 json 数据。

import urllib
import re
import json

htmltext = urllib.urlopen('******http://call******')

data = json.load(htmltext)

print data

它打印出如下结构的结果：

{u'status': u'OK', u'response': {u'docs': [{u'type_of_material': u'Article', u'blog': [], u'news_desk': None, u'lead_paragraph': u'Hon. PRESTON KING, one of the ablest, most upright, and most influential members of the Democratic party of this State, thus expression his opinion in regard to political prospects in a letter to a friend: OGDENSBURG, Saturday, Sept. 16, 1854.', u'headline': {u'main': u'POLITICAL.; New-York Politics--Letter from Preston King.'}, u'abstract': u'Letter to Jerry Rescue Celebration', u'print_page': u'8', u'word_count': 1526, u'_id': u'4fbfd3e945c1498b0d00ddca', u'snippet': u'Hon. PRESTON KING, one of the ablest, most upright, and most influential members of the Democratic party of this State, thus expression his opinion in regard to political prospects in a letter to a friend: OGDENSBURG, Saturday, Sept. 16, 1854.', u'source': u'The New York Times', u'web_url': u'http://query.nytimes.com/gst/abstract.html?res=950CE6DE1238EE3BBC4B53DFB667838F649FDE', u'multimedia': [], u'subsection_name': None, u'keywords': [{u'name': u'persons', u'value': u'KING, PRESTON'}, {u'name': u'persons', u'value': u'SUMNER CHARLES'}, {u'name': u'persons', u'value': u'BEECHER, HENRY WARD'}], u'byline': None, u'document_type': u'article', u'pub_date': u'1854-10-03T00:03:58Z', u'section_name': None}, {u'type_of_material': u'Article', u'blog': [], u'news_desk': None, u'lead_paragraph': u'MISSISSIPPI LAWS IN WANT OF REMODELING. GOVERNOR McWILLIE, of Mississippl, has summened an extra session of the State Legislature, to assemble on the first Monday in November next, In this State, as in others which have adopted the system of biennial sessions of their Legislatures. the plan has not been found to be the best for the interests of the people.', u'headline': {u'main': u'Article 1 -- No Title', u'kicker': u'1'}, u'abstract': None, u'print_page': u'3', u'word_count': 334, u'_id': u'4fbfe29945c1498b0d04bed8', u'snippet': u'MISSISSIPPI LAWS IN WANT OF REMODELING. GOVERNOR McWILLIE, of Mississippl, has summened an extra session of the State Legislature, to assemble on the first Monday in November next, In this State, as in others which have adopted the system of biennial...', u'source': u'The New York Times', u'web_url': u'http://query.nytimes.com/gst/abstract.html?res=9F06E7D61331EE34BC4952DFBE668383649FDE', u'multimedia': [], u'subsection_name': None, u'keywords': [], u'byline': None, u'document_type': u'article', u'pub_date': u'1858-08-11T00:03:58Z', u'section_name': None}, ... u'meta': {u'hits': 150, u'offset': 0, u'time': 38}}, u'copyright': u'Copyright (c) 2013 The New York Times Company.  All Rights Reserved.'}

出于示例的目的，我在此处仅粘贴了 2 篇文章的数据（它实际上为您提供了 10 篇文章的数据）。

现在我想解析该数据并提取所有“web_url”属性。我该怎么做？

我试过那个代码：

import urllib
import re
import json

htmltext = urllib.urlopen('******http://call******')

data = json.load(htmltext)

print data['web_url']

但它给了我这个错误：

Traceback (most recent call last):
    File "json_trying.py", line 10, in <module>
      print data["web_url"]
KeyError: 'web_url'

Answer 1

花点时间看看响应的结构。

{
    u'status': u'OK',
    u'response': {
        u'docs': [
            {
                ...
                u'web_url': u'http://query.nytimes.com/...',
                ...
            }
            {
                ...
            }
        ]
    }
}

---

for doc in data['response']['docs']:
    print doc['web_url']

Answer 2

我认为 web_url 键是响应键的子项，您应该能够以 data["response"]["web_url"] 的形式访问它。

Answer 3

使用 http://jsonviewer.stack.hu/ 。粘贴 JSON 响应以首先查看结构。

那么你可能不得不使用类似的 for 循环：

对于jsonresponse中的每个obj： 解析剩余的项目。