Topic: JSP 最佳实践：用 jsp:include 控制动态内容 |
 Print this page
|
| 1.JSP 最佳实践：用 jsp:include 控制动态内容 | Copy to clipboard |
|
Posted by: palatum Posted on: 2003-07-16 16:12 JSP 最佳实践：用 jsp:include 控制动态内容 英文原文 用于构建动态网站的简单 JSP 标记 级别：初级 Brett McLaughlin（brett@oreilly.com） 作家，O'Reilly and Associates 2003 年 7 月 本文是 Java“知情人士”Brett McLaughlin 继第一篇 JSP 最佳实践文章后的后续文章，在文中，作者向您演示了如何扩展 JSP 技术中用于动态内容的包含功能。了解静态 include 伪指令和动态 jsp:include 元素之间的差异，搞清楚如何混合搭配这二者以获取最优性能。 在新的 JSP 最佳实践系列的前一篇文章中，您了解了如何使用 JSP include 伪指令将诸如页眉、页脚和导航组件之类的静态内容包含到 Web 页面中。和服务器端包含一样，JSP include 伪指令允许某个页面从另一个页面提取内容或数据。清单 1 重温了 include 伪指令。 清单 1. JSP include 伪指令 <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>newInstance.com</title> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" /> <link href="/styles/default.css" rel="stylesheet" type="text/css" /> </head> <body> <%@ include file="header.jsp" %> <%@ include file="navigation.jsp" %> <%@ include file="bookshelf.jsp" %> <%@ include file="/mt-blogs/index.jsp" %> <%@ include file="footer.jsp" %> </body> </html> ]]> 您需要什么 本系列中的所有最佳实践都基于 JavaServer Pages 技术。要运行其中的任何最佳实践，都需要在您的本地机器或测试服务器上安装符合 JSP 技术的 Web 容器。您还需要使用文本编辑器或 IDE 来对 JSP 页面编码。 虽然 include 非常适于将静态内容并入 Web 页面，但对于动态内容却不尽如人意。我们在前一篇文章中在试图重新装入高速缓存文件时发现了这一问题。与大多数页眉文件及页脚文件不同，动态内容变化频繁，必须时刻更新。我们将首先扼要地重述一下 include 伪指令的局限性，然后我将向您演示如何用 jsp:include 标记来扩展 JSP 的包含能力。 高速缓存问题 JSP include 伪指令的不足之处有一个是：它会导致 Web 浏览器高速缓存所有页面。在处理诸如页脚、版权声明或一组静态链接之类的静态组件时，这是有意义的。这些文件不会改变，因此没有理由让 JSP 解释器不断地重新轮询其中的数据。凡是可能的地方，都应该实现高速缓存，因为它改善了应用程序的性能。 JSP 测试和开发 在构建 Web 应用程序或网站时，可能需要大量更新页眉、页脚和导航链接。仅仅为了看到对所包含文件所做的更改，而被迫不断地关闭浏览器或清除其高速缓存，这可能是件痛苦的事情。另一方面，为了结束开发周期，而不得不彻底检查一遍并修改数百个使用了 include 伪指令的页面，这也是一件痛苦的事情。我的建议是，在测试期间禁用浏览器高速缓存。在大多数情形下，这样做能够彻底解决问题。也有极少数情形，这样做并不奏效，这时可以在浏览器或服务器上不断地重新启动 Web 容器来确保不进行高速缓存。 但是，有时侯，进行高速缓存会得不偿失。如果提入的内容来自使用动态数据（如 Weblog 或数据库驱动的 JSP 文件）的程序，甚至如果所包含的内容是经常变化的 HTML（如时间戳记），那么每当装入 Web 页面时，都需要显示这些文件或程序的最新版本。遗憾的是，JSP include 伪指令并不具备这一功能。在测试和开发周期（请参阅侧栏“JSP 测试和开发”）中，在浏览器中禁用高速缓存通常能够解决这一问题。但是，对于实际使用的应用程序而言，性能是任何设计决策过程中的一项重要因素，禁用高速缓存并不是一种可行的长远之计。更好的解决方案是使用 jsp:include 标记。 jsp:include 标记 jsp:include 只不过是一个不同于 include 的伪指令而已。jsp:include 的优点在于：它总是会检查所含文件中的变化。过一会儿我们将研究这一新标记的工作方式。但首先看一下两种 include 各自的代码，以便能够看到二者之间的异同。 清单 2 显示了一个简单页面，它使用了原始的 JSP include 伪指令。 清单 2. JSP include 伪指令 <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>JSP include element test</title> </head> <body> This content is statically in the main JSP file.<br /> <%@ include file="included.html" %> </body> </html> ]]> 清单 3 是同一个页面，只不过这里转成使用 jsp:include 标记。 清单 3. 转成使用 jsp:include <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>JSP include element test</title> </head> <body> This content is statically in the main JSP file.<br /> <jsp:include page="included.html" flush="true" /> </body> </html> ]]> 您应该注意这两种代码类型之间的两大区别。首先，jsp:include 元素不使用属于 include 伪指令的 %@ 语法。实际上，jsp 前缀让 JSP 编译器知道：它应该寻找标准 JSP 标记集中的元素。其次，指定要包含的文件的属性从 file 变成了 page。如果愿意，可以自己测试一下新标记的结果。只需更改上一篇文章（请参阅参考资料）中 included.html 文件的内容，然后重新装入浏览器页面，就会立即看到新内容。 flush 属性 您可能已注意到 jsp:include 代码示例中的 flush 属性。顾名思义，flush 指示在读入包含内容之前是否清空任何现有的缓冲区。JSP 1.1 中需要 flush 属性，因此，如果代码中不用它，会得到一个错误。但是，在 JSP 1.2 中，flush 属性缺省为 false。由于清空大多数时候不是一个重要的问题，因此，我的建议是：对于 JSP 1.1，将 flush 设置为 true；而对于 JSP 1.2 及更高版本，将其设置为关闭。 jsp:include 是如何工作的 如果您有点爱刨根问底，那么可能十分想知道 jsp:include 标记的行为为什么与 include 伪指令不同。道理其实十分简单：jsp:include 包含的是所包含 URI 的响应，而不是 URI 本身。这意味着：对所指出的 URI 进行解释，因而包含的是生成的响应。如果页面是 HTML，那么将得到一点也没有变化的 HTML。但是，如果是 Perl 脚本、Java servlet 或者 CGI 程序，那么得到的将是从该程序解释而得的结果。虽然页面通常就是 HTML，但实际程序恰好是达到目的的手段。而且，由于每次请求页面的时候都会进行解释，因此从来不会象使用 include 伪指令时那样高速缓存结果。虽然这只是很小的变动，但它却导致了您所见到的行为中的全部差异。 一种混合搭配的解决方案 include 伪指令在某些网站上有其用武之地。例如，如果站点包含一些（如果有变化，也很少）几乎没有变化的页眉、页脚和导航文件，那么基本的 include 伪指令是这些组件的最佳选项。由于 include 伪指令采用了高速缓存，因此只需放入包含文件一次，其内容就会被高速缓存，其结果会是极大地提高了站点的性能。 然而，对于现在许多 Web 应用程序或站点而言，地毯式的高速缓存并不能解决问题。虽然页眉和页脚可能是静态的，但是不可能整个站点都是静态的。例如，从数据库提取导航链接是很常见的，并且许多基于 JSP 技术的站点还从其它站点或应用程序上的动态 JSP 页面提取内容。如果正在处理动态内容，那么需要采用 jsp:include 来处理该内容。 当然，最好的解决方案是经常把这两种方法混合搭配使用，将每种构造用到最恰当的地方。清单 4 是混合搭配包含解决方案的一个示例。 清单 4. 混合搭配解决方案 <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>newInstance.com</title> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" /> <link href="/styles/default.css" rel="stylesheet" type="text/css" /> </head> <body> <jsp:include page="header.jsp" flush="true"> <jsp:param name="pageTitle" value="newInstance.com"/> <jsp:param name="pageSlogan" value=" " /> </jsp:include> <%@ include file="/navigation.jsp" %> <jsp:include page="bookshelf.jsp" flush="true" /> <jsp:include page="/mt-blogs/index.jsp" flush="true" /> <%@ include file="/footer.jsp" %> </body> </html> ]]> 上面的代码显示了前一篇文章中的示例索引页面。导航链接和页脚是静态内容，一年最多更改一次。对于这些文件，我使用了 include 伪指令。内容窗格包含 Weblog 和“bookshelf”组件，它们是动态生成的。这两个组件需要一直更新，因此对它们，我使用了 jsp:include 标记。header.jsp 文件有点奇怪。这个组件是从另一个本质上是静态的 JSP 页面提取的。但是，正如您将注意到的那样，它从包含页提取页“标语”，然后将它显示出来。要处理这一共享信息，我们必须向页眉文件传入参数。而要处理那些参数，就必须使用 jsp:include 元素。 如果您想了解有关那些参数的内容，请放心，后续文章不久就将能满足您的需要。在下一篇文章中，我将解释 JSP 参数以及它们是如何与 JavaBeans 组件交互的。届时，我们网上见。 | |
| 2.Re:JSP 最佳实践：用 jsp:include 控制动态内容 [Re: palatum] | Copy to clipboard |
|
Posted by: palatum Posted on: 2003-07-16 16:13 JSP best practices: Manipulate dynamic content with jsp:include A simple JSP tag for building dynamic Web sites Level: Introductory Brett McLaughlin (brett@oreilly.com) Author, O'Reilly and Associates April 29, 2003 In this follow-on to his first JSP best practices installment, Java tipster Brett McLaughlin shows you how to extend the JSP technology inclusion functionality for dynamic content. Learn the differences between the static include directive and the dynamic jsp:include element, and find out how to mix and match the two for optimum performance. In the previous installment in the new JSP best practices series, you learned how to use the JSP include directive to include static content such as header, footer, and navigation components into your Web pages. Like server-side includes, the JSP include directive lets one page pull in content or data from another page. Listing 1 will refresh your memory of the include directive. Listing 1. The JSP include directive <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>newInstance.com</title> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" /> <link href="/styles/default.css" rel="stylesheet" type="text/css" /> </head> <body> <%@ include file="header.jsp" %> <%@ include file="navigation.jsp" %> <%@ include file="bookshelf.jsp" %> <%@ include file="/mt-blogs/index.jsp" %> <%@ include file="footer.jsp" %> </body> </html> ]]> What you'll need All of the best practices in this series are based on JavaServer Pages technology. To run any of them, you'll need to set up a JSP technology-compliant Web container, either on your local machine or on a test server. You'll also need to use a text editor or IDE to code your JSP pages. While include works very well for incorporating static content into your Web pages, it doesn't work so well for dynamic content. We discovered this in the last installment when we attempted to reload a cached file. Unlike most header and footer files, dynamic content changes frequently and must always be up to date. We'll start out with a quick recap of the limitations of the include directive, then I'll show you how to extend the JSP inclusion capability with the jsp:include tag. The trouble with caching One of the downsides of the JSP include directive is that it causes your Web browser to cache complete pages. This makes sense when it comes to dealing with static components such as a footer, a copyright notice, or a set of static links. These files don't change, so there's no reason for the JSP interpreter to constantly re-poll their data. And caching should be implemented wherever possible, because it aids your application performance. JSP testing and development When you're building a Web application or Web site, you may need to update your headers, footers, and navigation links quite a bit. It can be a pain to have to constantly close out your browser or clear its cache just to see the changes you've made to your included files. On the other hand, it's just as much of a pain to come to the end of your development cycle only to have to sweep through and change hundreds of pages that employ the include directive. My advice is to disable your browser cache during testing. In most cases this will resolve the problem altogether. In the rare cases where this doesn't work, you can always restart your Web container to ensure no caching is occurring, either on the browser or on the server. In some cases, however, caching can cause more trouble than it's worth. If you're pulling in content from a program that uses dynamic data (such as a Weblog or a database-driven JSP file), or even if your included content is a highly volatile bit of HTML (such as a time stamp), then you need the latest version of these files or programs to show up whenever your Web page is loaded. And unfortunately the JSP include directive doesn't have that functionality. This problem is commonly handled in the test-and-development cycle (see the sidebar "JSP testing and development") by disabling caching on the browser. But for working applications, where performance is an important factor in any design decision, disabling caching isn't a viable long-term solution. A better solution is to use a jsp:include tag. The jsp:include tag jsp:include is simply a different directive from include. The advantage of jsp:include is that it will always check for changes in the included file. We'll get into how the new tag works in just a moment. But first we'll take a look at the code for each of the two includes, so that you can see the similarities and differences between them. Listing 2 shows a simple page that uses the original JSP include directive. Listing 2. The JSP include directive <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>JSP include element test</title> </head> <body> This content is statically in the main JSP file.<br /> <%@ include file="included.html" %> </body> </html> ]]> Listing 3 is the same page converted to use the jsp:include tag. Listing 3. Converting to jsp:include <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>JSP include element test</title> </head> <body> This content is statically in the main JSP file.<br /> <jsp:include page="included.html" flush="true" /> </body> </html> ]]> You should notice two big differences between the two code types. First, the jsp:include element doesn't use the %@ syntax that is part of the include directive. Instead, the jsp prefix lets the JSP compiler know that it should look for the element in the standard JSP set of tags. Second, the attribute that specifies the file to include has changed from file to page. If you like, you can test the results of the new tag for yourself. Simply change the content of your own included.html file from the last installment (see Resources), reload your browser page, and you'll see the new content immediately. The flush attribute You may have noted the flush attribute in the jsp:include code example. As its name suggests, flush indicates whether any existing buffer should be flushed before reading in the included content. The flush attribute is required in JSP 1.1, so you'll get an error if you leave it out of your code. In JSP 1.2, however, the flush attribute defaults to false. Because flushing is rarely a big concern, my advice is to leave flush set to true for JSP 1.1 and leave it off for JSP 1.2 and above. How jsp:include works If you're even mildly geeky, you're probably anxiously wondering why the jsp:include tag behaves differently from the include directive. It's actually pretty simple: jsp:include includes the response from the included URI, rather than the URI itself. This means that the indicated URI is interpreted and the resulting response is included. If the page is HTML, you get the HTML essentially unchanged. But if it's a Perl script, a Java servlet, or a CGI program, you'll get the interpreted result from that program. While this usually turns out to be HTML, the actual program is just a means to an end. And, because interpretation happens at every page request, the results are never cached as they were with the include directive. It's such a minor change, but it makes all the difference in the behavior you see. A mix-and-match solution The include directive has its place on some Web sites. For example, if your site includes header, footer, and navigation files that rarely (if ever) change, then the basic include directive is a fine choice for these components. Because the include directive mandates caching, you'll only need to pull in the included files once, the content will be cached, and your site will receive a nice performance boost as a result. For many of today's Web applications or sites, however, blanket caching won't work. While your headers and footers may be static, it's unlikely that your whole site will be. It's not uncommon to pull navigation links from a database, for example, and many JSP technology-based sites pull content from dynamic JSP pages on other sites or applications. If you are dealing with dynamic content, you will need to employ jsp:includes to handle that content. Of course, the best solution is often to mix and match, using each construct where it is most appropriate. Listing 4 is an example of a mix-and-match include solution. Listing 4. A mix-and-match solution <![CDATA[ <%@ page language="java" contentType="text/html" %> <html> <head> <title>newInstance.com</title> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" /> <link href="/styles/default.css" rel="stylesheet" type="text/css" /> </head> <body> <jsp:include page="header.jsp" flush="true"> <jsp:param name="pageTitle" value="newInstance.com"/> <jsp:param name="pageSlogan" value=" " /> </jsp:include> <%@ include file="/navigation.jsp" %> <jsp:include page="bookshelf.jsp" flush="true" /> <jsp:include page="/mt-blogs/index.jsp" flush="true" /> <%@ include file="/footer.jsp" %> </body> </html> ]]> The above code shows the example index page from the previous installment. The navigation links and footer are static content, changing once a year at most. For these files, I've used the include directive. The content pane includes Weblog and "bookshelf" components, which are generated dynamically. Both of these components need to be always up to date, so for them I've used the jsp:include tag. The header.jsp file is the oddball of the lot. This component is pulled from another JSP page that is essentially static. But, as you'll note, it pulls in the page "slogan" from the including page and writes that out. To deal with this shared information, I have to pass in parameters to the header file. And to deal with those parameters, I have to use the jsp:include element. If you're wondering about those parameters, rest assured that you won't wonder for long. In the next installment, I'll explain JSP parameters and how they interact with JavaBeans components. Until then, I'll see you online. | |
 |
Powered by Jute Powerful Forum® Version Jute 1.5.6 Ent Copyright © 2002-2021 Cjsdn Team. All Righits Reserved. 闽ICP备05005120号-1 客服电话 18559299278 客服信箱 714923@qq.com 客服QQ 714923 |