亚洲日韩欧洲乱码AV夜夜摸,精品亚洲成A人在线观看青青,亚洲色欲色欲www

jsp-java-mysql字符集统一

soochow_hhb 以java论成�?以架构论英雄 — Fri, 03 Nov 2006 01:03:00 GMT

1、在jsp中，字符集由<%@ page language="java" contentType="text/html;charset=GBK"%>来指定，如果处理中文�Q�需指定为gbk。所以提交的数据全部都��用gbk�~�码�Q?br />2、在java中，从jsp提交�q�来的数据是gbk�~�码的，而java默认的处理方式是使用ISO-8859-1�~�码�Q�所以若在java中能昄��jsp提交�q�来的gbk�~�码方式的数据，��p��q�行一个�{换：��gbk转换为ISO-8859-1�Q�这��h��据就可正常显�C��?br />3、如果要��jsp提交的数据经�q�java后台处理�Q�然后放入数据库中，��又多了一个数据编码方式：数据库的数据�~�码方式。以mysql��Z��Q�默认字�W�集为latin�Q�所以若��gbk的数据放入，必然出现��q��Q�所以首先应改变数据库的�~�码方式为gbk�?br />
�l�g��Q�简单的讄��Ҏ��如下�Q?br />
jsp (gbk) ----> java(ISO-8859-1) -------> mysql(gbk)
转换
�q�样讄��之后�Q�还有一�U�情况会出现��q��Q�当你在jsp端提交一个id��P��然后在java处理端通过id号得到数据库中的gbk�Q�汉字）数据�Q�这是就会出��Cؕ码，因�ؓjava处理的是ISO-8859-1�Q�你在java到mysql中有了一个ISO-8859-1到gbk的�{换，而这些数据本�w�就是ISO-8859-1,所以�{换后��出现问题了�?br />解决办法�Q�将在java中处理的gbk数据转换为ISO-8859-1卛_��?br />
��M��Q�在jsp�Q�java�Q�mysql三者之间的�~�码方式必须�l�一�Q�不能错误的转换或者不转换�?img src ="http://www.tkk7.com/soochow_hhb/aggbug/78845.html" width = "1" height = "1" />

soochow_hhb 以java论成�?以架构论英雄 2006-11-03 09:03 发表评论

soochow_hhb 以java论成�?以架构论英雄 — Sat, 30 Sep 2006 02:37:00 GMT

好久写有关java技术的文章了，�q�几天在图书馆看到《Spring in Action》就借过来随手翻��，觉得�W�一章的IoC的概念讲的很好，现在整理如下�?br />�Q?�Q�从�cȝ��耦合说�v
在实际的�pȝ��中，要多个类共同协作来完成某一��Q务，一般称为复合。在�pȝ��的设计过�E�中�Q�耦合是不可避免的�Q�是必须的，但是它会带来以下问题�Q?br />�Q�－难以��试�Q�看下面的两个类�Q?br />KnightOfTheRoundTable
                  |
                  |  new
                 V
         HolyGrailQuest
在KnightOfTheRoundTable�c�M��使用了HolyGrailQuest�c�，�q�且可能使用不同的Quest�c�R��这两个�cȝ��功能是骑士去执行不同的�Q务，HolyGrail只是代表了一�U��Q务。这�U�通过国new的方式，使得knight和quest两个�cȝ��密的耦合在一赗��设计完成后做单元测试，看下面的代码�Q?br />KnightOfTheRoundTable knight = new KnightOfTheRoundTable("Bedivere");
HolyGrail grail = knight.embarkOnQuery();
assertNotNull(grail);
assertTrue(grail.isHoly());
在测试KnightOfTheRoundTable�cȝ��时候，间接的测试了HolyGrailQuest�c�，但是对于HolyGrailQuest�cȝ��情况�q�没有很昑ּ�的测试，所以��用了最后的两行代码来测试，昑־�很笨拙。不知道是否��试了所有可能的情况�?br />�Q�－难以�l�护
如果以后修改了代码，或者增加了/改变了knight的�Q务，代码必须改动�Q�测试代码也要改动，�q�且改动一个地方，可能会媄响到其他很多地方�Q��ؓ日后的维护工作带来了�ȝ��
�Q�－紧耦合
提示�Q�这�U�模式引��L��问题�Q�主要是因�ؓ紧耦合所��_��所以要解决�q�种问题的首要方案就是解耦合�Q�但如何来解耦合呢？��L��l�看�Q?br />�Q?�Q�解耦合——通过接口interface来实�?br />因�ؓ骑士可能执行不同的�Q务，也有不同�U�类的骑士，所以我们可以将��d��和骑士抽象�ؓ接口�Q�将具体的实现隐藏在接口之下�Q�这样就不必在knight�c�M��通过昄��?HolyGrailQuest quest = new HolyGrailQuest ()�q�样的语句来创徏Quest对象�Q�而可以通过接口Quest quest �Q?new HolyGrailQuest()来实现。这样就形成了如下的�c�d��Q?br />

�q�样�Q�虽然��用了接口�Q��得层�ơ分明，通过接口Quest来实现探险，但是�q�是只能从事一�U�探险�Q务，而如何才能让骑士从事��M��一�U��Q务呢�Q�请�l�箋看！
�Q?�Q�给予与获得
骑士执行探险��d��有两�U�方式：
�W�一、让骑士��d��获得探险��d��Q�前面的实现方式属于�q�类�Q�通过new来实玎ͼ�
�W�二�Q�让骑士被动的获得�Q务，即给予骑士某��Ҏ��险�Q务，�q�样��p��决了上面的问题（让骑士可以从事�Q何一�U��Q务，只要�l�他们分配即可）�?br />实际实现的方式很��单，看如下代码：
public void setQuest(Quest quest){
this.quest = quest;
}
��单的代码实现了方式的改变�Q�这��L��KnightOfTheRoundTable�c�M��入�Q何的Quest��d��Q�都可以接受了�?br />到这里可以很明显的看出来�Q�我们将以前的new方式�Q�翻转过来，即不是让��d��获得依赖�c�，而是被动的获得依赖类�Q�这��是IoC的核心，实际Fowler说得依赖注入也很贴切�Q�即向类中“注入”它依赖的其他类�?br />
好了�Q�到�q�里�Q�应该知道了IoC的基本概念了吧，也知道了IoC的基本功能了吧�?br />要想更深入的了解�Q�请��x��我的文章�?/span>

soochow_hhb 以java论成�?以架构论英雄 2006-09-30 10:37 发表评论

软�g新技术课的作�?

soochow_hhb 以java论成�?以架构论英雄 — Mon, 13 Jun 2005 08:23:00 GMT

关于Struts和webwork的比较，�q�有单元��试Junit和容器测试Cactus的一些情况！pdf格式�Q�欢�q�探讨相��x��术，本来�q�想写数据持久化的一些内容，如JDO和Hibernate�Q�因为时间关�p�，没有来得及写入，希望以后能够补上�Q?BR>
�Ƣ迎交流�Q?A href="http://www.tkk7.com/Files/soochow_hhb/从开源项目看软�g技�?rar">下蝲

soochow_hhb 以java论成�?以架构论英雄 2005-06-13 16:23 发表评论

使用Tiles和Struts设计web界面

soochow_hhb 以java论成�?以架构论英雄 — Mon, 13 Jun 2005 08:11:00 GMT

摘要: �q�行Web开发，很重要的一点就是界面的�l�合�Q�如何能够高效的实现��面的重用，本文�l�出了��用Tiles和Struts集成的方式来实现最大程度的界面重用�Q�只要��用tiles的xml配置文�g��可以将web��面设计师的��C��l�合��h��Q��ƈ且可以��用Struts的控制器来引用，实现了页面和逻辑的分��！阅读全文

soochow_hhb 以java论成�?以架构论英雄 2005-06-13 16:11 发表评论

soochow_hhb 以java论成�?以架构论英雄 — Sat, 11 Jun 2005 03:58:00 GMT

摘要: 本篇文章包含了在用Struts开发web应用时经常碰到的一些异常和错误�Q�根据异常或错误信息本��n�Q�经常可以找到潜在的错误发生原因�?下面列出了一些Struts的常见错误和异常�Q��ƈ�l�出了一些可能发生此�c�错误或异常的原因。有的后面有相关�q�接�Q�你可以通过它找到更多的信息�? Cannot retrieve mapping for action 异常 javax.s... 阅读全文

soochow_hhb 以java论成�?以架构论英雄 2005-06-11 11:58 发表评论

Struts vs webwork 中文

soochow_hhb 以java论成�?以架构论英雄 — Sun, 29 May 2005 06:16:00 GMT

�q�是前一��文章的译文�Q�来�?SPAN lang=EN-US style="FONT-SIZE: 10.5pt; FONT-FAMILY: 'Times New Roman'; mso-bidi-font-size: 12.0pt; mso-fareast-font-family: 宋体; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA">moxie的《WebWork教程�Q?.90版�?BR>�Ƣ迎一��h��讨WEB Framework�?A href="http://www.tkk7.com/soochow_hhb/">http://www.tkk7.com/soochow_hhb/

�?/SPAN> �?/SPAN>	Struts1.1	WebWork2
Action�c?/SPAN>	�?/SPAN>Struts里面�Q�每一�?/SPAN>Action�c�d��需要��承一个抽象的�c?/SPAN>org.apache.struts.action.Action。这个在Java�~�程中会引来一些问题，��是关于多种�l�承的问题�?/SPAN>	WebWork�?/SPAN>Action�c�M��需要实现接�?/SPAN>com.opensymphony.xwork.Action�Q�也可以实现其它的接口来实现更多的功能，譬如�Q?/SPAN>validate�Q�验证）�Q?/SPAN>localware(国际�?/SPAN>)�{�。当�Ӟ��它也提供了一个类ActionSupport集成了上面的所有功能，我们在开发中可以�Ҏ��需要选择�?/SPAN>
�U�程模型	Struts �?/SPAN>Action必需�?/SPAN>thread�Q?/SPAN>safe方式�Q�它仅仅允许一个实例去处理所有的��h��。所�?/SPAN>action用到的所有的资源都必需�l�一同步�Q�这个就引�v了线�E�安全的问题�?/SPAN>	�?/SPAN>WebWork中，每个��h��对应一�?/SPAN>Action�Q�因此没有线�E�的安全问题。实际上Servlet容器�Ҏ��个请求也产生多个对象�Q�它也没有证明对性能和垃圑֛�收��生太多的影响�?/SPAN>
Servlet的依�?/SPAN>	Struts处理Action时必需要依�?/SPAN>ServletRequest �?/SPAN>ServletResponse�Q�所有它摆脱不了Servlet容器�?/SPAN>	WebWork�?/SPAN>Action不用依赖Web层和其它的容器。它可以通过ActionContext�Q�直接去讉K��Request�?/SPAN>Response�Q�但�q�个是可选的�Q�只有在必需的请求下使用�?/SPAN>
��试	Struts的每�?/SPAN>Action都同Web层耦合在一��P��q�样它的��试依赖�?/SPAN>Web容器�Q�单元测试也很难实现。不�q�有一�?/SPAN>Junit的扩展工�?/SPAN>Struts TestCase可以实现它的单元��试�?/SPAN>	Webwork�?/SPAN>action能够通过赋予一定的属性，��可以执行单元测试。同时也可以使用一�?/SPAN>mock的实例去��试�Q�而不是通过启动web容器来进行测试�?/SPAN>
FormBean	Struts要求�?/SPAN>FormBean对应每一个表单，而且FormBean必需�l�承抽象�c?/SPAN>ActionForm。而��?/SPAN>DynaBeans实际上没有太大的意义。不能够很好的处理现有的模型�?/SPAN>	Webwork 能够动态的攉��web的数据然后再赋值给bean。它也可以��?/SPAN>FormBean的�Ş式，FormBean可以是普通的DTO和域对象�Q�它不用重新�Ҏ��域对象来生成新的FormBean�Q�也不需�l�承抽象�c?/SPAN>ActionForm�?/SPAN>
前端表达式语�a�	Struts集成�?/SPAN>JSTL�Q�所以它主要使用JSTL的表辑ּ�语言来获取数据。可�?/SPAN>JSTL的表辑ּ�语言�?/SPAN>Collection和烦引属性方面处理显得很弱�?/SPAN>	WebWork的表辑ּ�语言使用了功能强大的OGNL。它使用OGNL建立一�?/SPAN>OgnlValueStack来搜索数据�?/SPAN>Webwork前端也可以��?/SPAN>JSTL�Q�但它同时支持：velocity�?/SPAN>freemaker�?/SPAN>jspparer�?/SPAN>xml�?/SPAN>
�c�d��的�{�?/SPAN>	Struts�?/SPAN>FormBean把所有的数据都作�?/SPAN>String�c�d��Q�它可以使用工具Commons-Beanutils�q�行�c�d��转化。但它的转化都是�?/SPAN>Class�U�别�Q�而且转化的类型是不可配置的。类型�{化时的错误信息返回给用户也是非常困难的�?/SPAN>	WebWork使用OGNL�q�行�c�d��转化�Q�提供了所有基本类型的转化功能。类型�{化可以直接对一�?/SPAN>Class�q�行�Q?/SPAN>Class�U�别�Q��{化，也可以对Class的字�D�进行类型�{化。它使用拦截器可以很�Ҏ��的将�c�d��转化的错误信息返回给用户�Q�而且错误信息可以对应��C��个相应的字段�?/SPAN>
�?/SPAN>Action 执行前和后的处理	Struts处理Action的时候是��Z��class�?/SPAN>hierarchies�Q�很隑֜�action处理前和后进行操作�?/SPAN>	Webwork2 允许您处�?/SPAN>Action可以通过拦截器，��是在每一�?/SPAN>Action处理前或者后�q�行其它操作。它的拦截器可以在配�\|�文件中动态添加，�q�样Action和拦截器之间完全解藕�Q�更好的实现了组件化�?/SPAN>
验证处理	Struts的验证是调用FormBean�?/SPAN>validator()�Ҏ��Q�其实就是对FormBean的验证。它一般��用框�?/SPAN>Commons Validation�q�行数据验证处理。它使用了一个全局的配�\|�文�?/SPAN>validation.xml定义�?/SPAN>FormBean的验证信息�?/SPAN>Struts�?/SPAN>FormBean属性都被认为是String�c�d��Q�所以它在验证时也需要额外的�c�d��转化�?/SPAN>	WebWork使用Xwork的验证框架进行验证处理，它可以通过配置拦截器来�Ȁ�z�R��它可以为每个需要验证的Class指定一�?/SPAN>xml验证文�g�Q�也可以��Z��?/SPAN>Class在不同的情况指定不同�?/SPAN>xml验证文�g�?/SPAN>WebWork证可以给每个Action�c�L��定对应的验证文�g�Q�也可以�l?/SPAN>Action的字�D�去指定验证文�g。通过拦截器来�l�装Action和其验证文�g�Q��它们之间完全解藕�?/SPAN>
�?/SPAN>Action执行的控�?/SPAN>	Struts创徏一�?/SPAN>Action�Q�如果想控制它的执行��序��会非常困难。甚至你要重新去�?/SPAN>Servlet来实��C��的这个功能需求�?/SPAN>	在这个方面，WebWork的拦截器栈提供了强大的功能�?/SPAN>Action的所有切面功能都有拦截器来实玎ͼ�比如�Q�取�?/SPAN>request��h��参数、验证处理等�Q�，�q�样你就可以用拦截器栈来�l�织拦截器的执行��序。例如：你需要在使用request��h��参数来设�\|?/SPAN>Action属性之前，使用IoC框架讄��Action的属性，反之已然。这�Ӟ��你就可以�?/SPAN>package�?/SPAN>Action指定一个拦截器栈来实现�?/SPAN>

soochow_hhb 以java论成�?以架构论英雄 2005-05-29 14:16 发表评论

Sturts vs WebWork

soochow_hhb 以java论成�?以架构论英雄 — Fri, 27 May 2005 06:38:00 GMT

来自官方http://www.opensymphony.com/webwork/wikidocs/Comparison%20to%20Struts.html�|�站对Struts和WebWork的比较�?BR>

Feature	Struts	WebWork 1.x	WebWork 2.x
Action classes	Struts requires Action classes to extend an Abstract base class. This shows a common problem in Struts of programming to abstract classes instead of interfaces.	Action classes must implement the webwork.Action Interface. There are other Interfaces which can be implemented for other services, such as storing error messages, getting localized texts, etc. The ActionSupport class implements many of these Interfaces and can act as a base class. WebWork is all written to Interfaces, which allows for plugging in your own implementations.	An Action must implement the com.opensymphony.xwork.Action Interface, with a series of other Interfaces for other services, like in WebWork 1.x. WebWork2 has its own ActionSupport to implement these Interfaces.
Threading Model	Struts Actions must be thread-safe because there will only be one instance to handle all requests. This places restrictions on what can be done with Struts Actions as any resources held must be thread-safe or access to them must be synchronized.	WebWork Actions are instantiated for each request, so there are no thread-safety issues. In practice, Servlet containers generate many throw-away objects per request, and one more Object does not prove to be a problem for performance or garbage collection.	ditto
Servlet Dependency	Struts Actions have dependencies on Servlets because they get the ServletRequest and ServletResponse (not HttpServletRequest and HttpServletResponse, I've been told) when they are executed. This tie to Servlets (although not Http*) is a defacto tie to a Servlet container, which is an unneeded dependency. Servlets may be used outside a Web context, but it's not a good fit for JMS, for instance.	WebWork Actions are not tied to the web or any container. WebWork actions CAN choose to access the request and response from the ActionContext, but it is not required and should be done only when ABSOLUTELY neccessary to avoid tieing code to the Web.	ditto
Testability	Many strategies have sprung up around testing Struts applications, but the major hurdle is the fact that Struts Actions are so tightly tied to the web (receiving a Request and Response object). This often leads people to test Struts Actions inside a container, which is both slow and NOT UNIT TESTING. There is a Junit extension : Struts TestCase (http://strutstestcase.sourceforge.net/)	WebWork actions can be tested by instantiating your action, setting the properties, and executing them	ditto, but the emphasis on Inversion of Control makes testing even simpler, as you can just set a Mock implementation of your services into your Action for testing, instead of having to set up service registries or static singletons
FormBeans	Struts requires the use of FormBeans for every form, necessitating either a lot of extra classes or the use of DynaBeans, which are really just a workaround for the limitation of requiring FormBeans	WebWork 1.x allows you to have all of your properties directly accessible on your Action as regular Javabeans properties, including rich Object types which can have their own properties which can be accessed from the web page. WebWork also allows the FormBean pattern, as discussed in "WW1:Populate Form Bean and access its value"	WebWork 2 allows the same features as WebWork 1, but adds ModelDriven Actions, which allow you to have a rich Object type or domain object as your form bean, with its properties directly accessible to the web page, rather than accessing them as sub-properties of a property of the Action.
Expression Language	Struts 1.1 integrates with JSTL, so it uses the JSTL EL. This EL has basic object graph traversal, but relatively weak collection and indexed property support.	WebWork 1.x has its own Expression language which is built for accessing the ValueStack. Collection and indexed property support are basic but good. WebWork can also be made to work directly with JSTL using the Filter described in WW1:Using JSTL seamlessly with WebWork	WebWork 2 uses XW:Ognl which is a VERY powerful expression language, with additions for accessing the value stack. Ognl supports very powerful collection and indexed property support. Ognl also supports powerful features like projections (calling the same method on each member of a collection and building a new collection of the results), selections (filtering a collection with a selector expression to return a subset), list construction, and lambda expressions (simple functions which can be reused). Ognl also allows access to static methods, static fields, and constructors of classes. WebWork2 may also use JSTL as mentioned in WW1:Using JSTL seamlessly with WebWork
Binding values into views	Struts uses the standard JSP mechanism for binding objects into the page context for access, which tightly couples your view to the form beans being rendered	WebWork sets up a ValueStack which the WebWork taglibs access to dynamically find values very flexibly without tightly coupling your view to the types it is rendering. This allows you to reuse views across a range of types which have the same properties.	ditto
Type Conversion	Struts FormBeans properties are usually all Strings. Struts uses Commons-Beanutils for type conversion. Converters are per-class, and not configurable per instance. Getting a meaningful type conversion error out and displaying it to the user can be difficult.	WebWork 1.x uses PropertyEditors for type conversion. PropertyEditors are per type and not settable per Action, but field error messages are added to the field error map in the Action to be automatically displayed to the user with the field.	WebWork2 uses Ognl for type conversion with added converters provided for all basic types. Type converters default to these converters, but type conversion can be specified per field per class. Type conversion errors also have a default error message but can be set per field per class using the localization mechanism in WW2 and will be set into the field error messages of the Action.
Modular Before & After Processing	Class hierarchies of base Actions must be built up to do processing before and after delegating to the Action classes, which can lead deep class hierarchies and limitations due to the inability to have multiple inheritance WW:Comparison to Struts#1	Class hierarchies	WebWork 2 allows you to modularize before and after processing in Interceptors. Interceptors can be applied dynamically via the configuration without any coupling between the Action classes and the Interceptors.
Validation	Struts calls validate() on the FormBean. Struts users often use Commons Validation for validation. I don't know a lot about this, so I'll put some questions here: Because FormBean properties are usually Strings, some types of validations must either be duplicated (checking type conversion) or cannot be done? Can Commons Validation have different validation contexts for the same class? (I've been told yes, so that's a good thing) Can Commons Validation chain to validations on sub-objects, using the validations defined for that object properties class?	WebWork1.x calls the validate() method on Actions, which can either do programmatic validations or call an outside validation framework (this is apparently the same as Struts)	WebWork2 can use the validate() method of WebWork and Struts and / or use the XW:Validation Framework, which is activated using an XWork Interceptor. The Xwork Validation Framework allows you to define validations in an XML format with default validations for a class and custom validations added for different validation contexts. The Xwork Validation Framework is enabled via an Interceptor and is therefore completely decoupled from your Action class. The Xwork Validation Framework also allows you to chain the validation process down into sub-properties using the VisitorFieldValidator which will use the validations defined for the properties class type and the validation context.
Control Of Action Execution	As far as I know Struts sets up the Action object for you, and you have very little control over the order of operations. To change them I think you need to write your own Servlet to handle dispatching as you want	The ActionFactory chain controls the order in which an Action is constructed and initialised, but this requires writing a class	The interceptor stacks in WebWork 2 are hugely powerful in this regard. All aspects of Action setup have been moved into Interceptor implementations (ie setting paramters from the web, validation etc), so you can control on a per action basis the order in which they are performed. For example you might want your IOC framework to setup the action before the parameters are set from the request or vice versa - you can thusly control this on a per package or per action basis with interceptor stacks.

soochow_hhb 以java论成�?以架构论英雄 2005-05-27 14:38 发表评论

soochow_hhb 以java论成�?以架构论英雄 — Fri, 27 May 2005 03:18:00 GMT

从国外网站上看到一��文章，转发�q�来�Q�附��单注解。原文在http://raibledesigns.com/page/rd/sunsets?catname=%2FJava
I updated my Web Framework Comparison presentation today. Rather than updating the graph that shows today's job availability, I did one that compares today to 6 months ago. Struts is still the clear winner (and growing). Spring is definitely growing. Tapestry has about the same amount of jobs (9 vs. 8). WebWork lost 10 opening (down to 4) and the demand for JSF skills has grown as well.

Struts明显是最大的赢家�Q��ƈ且持�l�增长，Spring明显的有所增长�Q�Tspestry基本持��^�Q�WebWorks下降�?�Q�JSF的需求有所增长�?/FONT>

Is WebWork a dying framework? I've heard folks complain about its small community, and there still aren't any books is only one book about it. Is that a jab at Patrick, Jason and Kris - or a jab at Manning? I'm not sure. The good news is WebWork in Action and WebWork Live should both be out this summer.

My search criteria for all of these was "framework and java" from the front page on dice.com. I did filter a bunch out for WebWork b/c there's some product called "WebWorks" that folks want to hire for.

In my own experience, these numbers are not as accurate as you might think. Since I gave my original presentation, I've been contacted a number of times to work on projects. It's about even between Struts, Spring MVC, WebWork and JSF. I haven't had a single inquiry to do Tapestry development. The bad part about Struts jobs is there's so many of them, that rates are likely pretty low (i.e. 35-45/hour), whereas the others can get you upwards of 80-90/hour.
依我个�h的经验，�q�些数字不一定就很准��?自从我发表第一��介�l�以来，我已�l�在��目中��用了好多�ơ，包括Struts, Spring MVC�Q?WebWorks 和JSF�Q?我还没有做Tapestry开发�?Struts不好的方面就是太多了�Q�其工资可能�E�微低一点（35/45/hour) ,而用其他框架能��你提高到80-90/hour�?/FONT>

So what do these numbers mean? Do they mean you should tailor your learnings and skills to the most popular frameworks? In a sense, it's important to do so. If nothing else, Struts skills are import so you can migrate all the Struts applications to your favorite framework. However, I don't think these numbers are that important when choosing a framework to start your project with. I think the most important thing in choosing a framework is passion. Which one do you want to work with the most? It's likely that your productivity will be higher if you're enthusiastic about the framework, rather than bored with all the skills you've accumulated using it. Then again, if you're motivated by productivity more than enthusiasm - using your skills to crank out applications quickly is probably a good idea.
那么�q�些数字意味着我们应该做什么呢�Q�是不是意味着应该学习那些最��行的框�Ӟ��乍眼看来�Q�这样做是很重要的�?如果没有别的�Q�Struts是很重要的，�q�且你可以将你所有的Struts应用�q�移��C��钟爱的框架中。不�q�，我认��些数字对你选择一�U�框架开始你的项目来说不是重要的。我惛_��选择一�U�框架时最重要的是passion�Q�激情）。你最想��用哪一个框架工作呢�Q�或许如果你对一个框架很热情�Q�你的生产力��获得极大的提高�Q�而不是乏味的使用已经�U�篏的技术�?另外�Q�如果你�׃��生��力推动而不是激情，那么�q�用你已�U�篏的技术去完成��目可能是一个好��L��?/FONT>

You might think that the number of skilled developers for framework X is important too. I don't think it is. I think the most important thing is to hire smart developers. A good developer can come up to speed on any framework in 2 weeks and be highly productive in 4 weeks. If not, the developer isn't that smart or the framework isn't that good.
你可能认为某个框架有�l�验的开发者的数量也很重要�Q�我不这么认为，我认为最重要的事情是雇用�_�明的开发者。一个好的开发�h员能在两周时间内使用��M��框架�Q��ƈ且在四周旉��内能熟练�q�用�?如果不是�q�样的话�Q?�q�个开发�h员就不是�_�明的开发者，或者这个框架不是优�U�的�?/FONT>

Just for kicks, I did some searching for other web frameworks as well:

Rife: 0
Wicket: 0
Echo: 3
Ruby on Rails: 1
ASP .NET: 2876

Now the question is - what kind of rates are these skills getting? I'd like to know what the average Rails and ASP .NET developers make. In Denver, Java developers seem to make between 65-85/hour when they're experienced contractors

soochow_hhb 以java论成�?以架构论英雄 2005-05-27 11:18 发表评论

soochow_hhb 以java论成�?以架构论英雄 — Thu, 19 May 2005 02:57:00 GMT

摘要: �W�二步：实现controller。在Struts中��承自Action。调用Model�Q�实现数据的深层�ơ检验（email是否存在�Q�和数据的插入，�E�序的蟩转等。代码如下：SignAction.java 1/**//** 2 * @author han 3 * soo... 阅读全文

soochow_hhb 以java论成�?以架构论英雄 2005-05-19 10:57 发表评论

soochow_hhb 以java论成�?以架构论英雄 — Thu, 19 May 2005 02:31:00 GMT

摘要: 本文通过一个实例——用��h��册——来说明Struts讄��的一般流�E�和�Ҏ��Q�中间设计到Struts的错误处理，资源文�g使用和持久化工具Hibernate的��用，�q�些信息都可以从本blog中找到�? 文章有一下几个部分组成：�Q?�Q�实例介�l�； �Q?�Q?不��用MVC模式�Q�（3�Q��用MVC模式�Q�（4�Q��用Struts实现用户注册的过�E��?nbsp; 阅读全文

soochow_hhb 以java论成�?以架构论英雄 2005-05-19 10:31 发表评论

Struts学习心得之Struts��程��（2�Q?

soochow_hhb 以java论成�?以架构论英雄 — Tue, 03 May 2005 01:35:00 GMT

上一节我们搭��Z��Struts的运行和开发环境。这一节着重讲�q�MVC的基本概念和与��用JSP/JavaBean模式开发Web应用的比较�?BR>�Q?�Q�Tomcat的启动�?BR> 从上一节的Tomcat的配�|�文件中�Q�我们可以看到关于Struts的配�|�，当Tomcat启动�Ӟ��会加载Struts的配�|�文件struts-config.xml�Q�关于配�|�文件的内容误��
�Q?�Q�JSP/JavaBean开发Web应用
在asp的时代，我们�q�没有讨��Z��么开发模式，也没有顾及MVC�Q�虽然MVC的概忉|��已出玎ͼ�。用asp开发不外乎以下两种情况�Q�一、就是直接将代码集成到asp文�g中，�l�验多的话，会将常用的函数放至一��P��引用时只需include卛_��Q�二、对于复杂的应用�Q�如文�g上传�{�，我们会用其他语音如VB、Delphi�{�写一个控�Ӟ��ADO卛_��于此�c�）�?BR> 当java��行�Q�JSP逐渐成熟�Q�也正是模式、框架流行的时代�Q�所以我们不��是看java的类库，�q�是看现在流行的框架�Q�无一没有模式的存在。当你初接触JSP的时候，��可以看��C��上说“尽量少的在jsp中插入Java代码”，�q�句话已是模式运用的端倪�?BR> 下面我们先不��MVC�Q�先从JSP/JavaBean来看Web的开发。这�U�模式很想上面提到的asp开发的�W�二�U�情况，我们��业务逻辑攑ֈ�JavaBean中，只用JSP去表�C�页面�?BR> 有过asp�l�验都知道，在接收表单数据的时候，我们不得不一个一个将数据接收�q�来�Q�以至于有很长的一�D�如�Q�CustomId �Q?request("CustomId")的代码。而在JSP中，我们不用�q�么�J�琐�Q�一个我们可以用EL直接表示卛_��Q�另外我们可以利用javaBean的introspection机制。直接将表单的数据映��到javabean的private变量中，当然javabean中必��d��含getter和setter函数�Q�有很幸�q�，很多的开发工具可以自动生成，又减��了工作量）。下面是一个示例�?BR>

<jsp:useBean id="prof" scope="page" class="register.Profile" />
<jsp:setProperty name="prof" property="*" />

       我们只要表单的action��面中，直接插入以上代码�Q�这些代码就不用解释了吧�Q�，��可以将表单中的数据映射到register.Profile�q�个javabean中，所以我们可以在javabean中实��C��务逻辑�Q�譬如数据的插入、修攏V��删除等。从而可以减��甚��x��除JSP中的java代码量�?BR> �Q?�Q�MVC模式
     上面提到的JSP/JavaBean与MVC模式很接�q�，但是�q�有很多差别。前者，我们可以看出�Q�在javabean中有很多的逻辑代码�Q�我们要控制��面的蟩转，�q�是要在JSP中读取javabean中的某些标志�Q�嵌入java代码来实玎ͼ��q�看��h��昑־�有些�W�拙。我们是否能在JSP和javabean中再加一层，专门来实现蟩转即控制��面的输出呢�Q�of course�Q�这��是MVC的思想�?BR>      Struts��是一个实现MVC模式的WEB开发框�Ӟ��在Struts中用JSP作�ؓViewer�Q�javabean�l�承自ActionForm来映��JSP表单数据�Q�用�l�承自Action的类来实现Controller�Q�而Model没有�Ҏ��定义�Q�它可以是EJB�Q�普通Bean或者Web服务�Q��M��Model中集成了业务逻辑的实玎ͼ�在Controller中去调用�Q�完成后Controller控制昄��面�Q�jsp文�g�Q��?BR>     �q�样一来，是不是业务流�E�非常的清晰�Q?BR>
     下面我们付诸实践�Q�实��C��个用��h��册的��单实例！

soochow_hhb 以java论成�?以架构论英雄 2005-05-03 09:35 发表评论

soochow_hhb 以java论成�?以架构论英雄 — Mon, 02 May 2005 12:00:00 GMT

摘要: 刚接触Struts�Q�刚接触MVC架构�Q�或者刚接触��M��C��西，我觉得首先要搞清楚流�E�，要对��C��物有一个大概的、��M��的认识，先有感性认识，先培养�v来兴��是非常重要的。下面就把我学习Struts的��M��业务��程通过一个实例来�q�行概括�Q�有什么错误和不妥之处�Q�请你发表评论指教！阅读全文

soochow_hhb 以java论成�?以架构论英雄 2005-05-02 20:00 发表评论