网络包含有丰富的、很有用的半结构化的信息。它们应该被提取出来。结构的类型包括页面间的超链接、超文本页面内的结构以及URL中的结构。我们已经发展了一种程序化语言 :Squeal。它使得基于结构的查询变得更加容易。特别地 ,只要网络是在一个标准的关系数据库 ,Squeal用户都可以对网络进行查询。我们描述Squeal ,并且显示在Squeal中写基于结构的信息工具是非常容易的事 The network contains a wealth of useful semi-structured information. They should be extracted. Types of structures include hyperlinks between pages, structures within hypertext pages, and structures in URLs. We have developed a procedural language: Squeal. It makes structure-based queries easier. In particular, Squeal users can query the web as long as the web is in a standard relational database. We describe Squeal and show that it is very easy to write structure-based information tools in Squeal