Python 从json中提取文本<；脚本>；当存在多个JSON时标记

我试图从中提取评级，以便从HTML代码中提取“ratingValue”和“alternateName”字段：

<script type=application/ld+json>{
"@context": "http://schema.org",
"@type": "ClaimReview",
"datePublished": "2019-01-03 ",
"url": "https://www.truthorfiction.com/are-americans-annually-healthcare-undocumented/",
"author": {
    "@type": "Organization",
    "url": "https://www.truthorfiction.com/",
    "image": "https://dn.truthorfiction.com/wp-content/uploads/2018/10/25032229/truth-or-fiction-logo-tagline.png",
    "sameAs": "https://twitter.com/whatstruecom"
},
"claimReviewed": "More Americans die every year from a lack of affordable healthcare than by terrorism or at the hands of undocumented immigrants.",
"reviewRating": {
    "@type": "Rating",
    "ratingValue": -1,
    "worstRating":-1,
    "bestRating": -1,
    "alternateName": "True"
},
    "itemReviewed": {
    "@type": "CreativeWork",
    "author": {
        "@type": "Person",
        "name": "Person",
        "jobTitle": "",
        "image": "",
        "sameAs": [
            ""
        ]
    },
    "datePublished": "",
    "name": ""
}
}</script>

---

然而，tmp显示了一个“application/ld+json”项的字典，它来自我想要提取的评级之前的位，我想知道如何循环或循环到脚本中存储评级的相关部分。

您需要使用键访问元素

rating_value = tmp['reviewRating']['ratingValue'] # -1
alternate_name = tmp['reviewRating']['alternateName'] # 'True'

或

它有2个

您可以从

find\u all（）

或循环并搜索是否包含字符串

tmp = None
for ldjson in soup.find_all('script', type='application/ld+json'):
    if 'ratingValue' in ldjson.text:
        tmp = json.loads(ldjson.text)

---

我的问题是soup.find找到的第一个json被分配给tmp，而不是找到rating_值和alternate_名称的json，我不确定如何将相关json加载到tmp中

KeyError:“reviewRating

您可以使用

find_all（）

方法，然后遍历标记集合：

all_scripts=soup.find_all（'script'）

。

review_rating = tmp['reviewRating']
rating_value = review_rating['ratingValue'] # -1
alternate_name = review_rating['alternateName'] # 'True'

tmp = json.loads(soup.find_all('script', type='application/ld+json')[1].text)

tmp = None
for ldjson in soup.find_all('script', type='application/ld+json'):
    if 'ratingValue' in ldjson.text:
        tmp = json.loads(ldjson.text)