从链接创建摘要
许多页面（facebook、google+ 等）都有一个功能，可以通过链接创建带有标题、图像和一些文本的摘要。我试图找出是否有任何关于如何执行此类功能的库或指南，但我的搜索结果根本没有帮助。
我知道我可以解析页面的 html 并提取我想要的元素，但我认为应该有某种标准来说明如何做到这一点（也许还包括如何创建对此类功能友好的页面。
任何有好的链接可以为我指明正确的方向吗？Javascript 或 .Net 是我的首选，但我也可以自己实现。
Many pages (facebook, google+ etc) have a function that creates a summary with header, image and some text from a link. I have tried to find out if there are any libraries or guidelines about how to do this kind of function but my search-results havn't been helpful at all.
I know that I can parse the html of a page and extract the elements I'd like but I think there should be some kind of standard in how to do this (perhaps also how to create pages that are friendly to this kind of functionallity.
Anyone that have a good link that will point me to the right direction? Javascript or .Net is my prefered choise but I can implement it myself too.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
对于“也许还如何创建对这种功能友好的页面”。部分：
您可能正在寻找开放图协议：
我认为这是 facebook 会首先关注的地方。但 Facebook 似乎有自己的算法，可以在这些标签丢失时检测页面最相关的部分。
For the "perhaps also how to create pages that are friendly to this kind of functionallity." part:
You are probably searching for the open graph protocol:
I think this is the first place facebook will look. But facebook seems to have its own algorithms to detect the most relevant part of the page when these tags are missing.
这样的函数通常是使用某种“爬行”来构建的，这意味着您的脚本打开链接并查看其数据。就像你自己建议的那样。
标准方式是大多数搜索引擎（例如 Google）的做法。您可以从网站的标题中获取标题，从描述中获取描述（如果有）。现在大多数搜索引擎都会忽略描述元数据，而是尝试制作自己的摘要。
这通常是通过查找标题（h1、h2 等）和段落来完成的。
为了使网站对此类抓取“友好”，您需要根据网络标准 (W3C) 构建网站。
只要能够执行一些基本的 HTTP-GET 操作，语言实际上并不重要。
Such a function is usually build using some kind of "crawling", meaning your script opens the link and looks at its data. Just like you suggest yourself.
Standard way is the way most search engines do it, like Google. You get the title from the title of the website, description from description if there are any. Most search engines now days ignore description meta data and instead try to make their own summary.
This is usual done by looking for headers (h1, h2, etc) ad then the paragraphs.
And to make a website "Friendly" for these kind of crawls you build your website according to web standards (W3C).
Language really doesn't matter as long as it is capable of doing some basic HTTP-GET.