文章来源:http://blog.csdn.net/Wingel/archive/2007/01/25/1493585.aspx
]]>
我们要拥p划的变更Q?br /> 你跟客户Q或者负责需求的人熟吗？只有时刻掌握着需求的变化Q才能时L握好自己的计划?br /> 你跟QA熟吗QQA对你q个人开发质量的印象如何Q清楚自q开发质量，才能保证把事情做好的能力一直在q步?br /> 你跟领导熟吗Q你保证你做的事情领导都知道吗？你想做什么领g知道Q?br /> 你敢不敢_(d)所有跟你有关的情况Q都在你的掌握Q?br /> ?x)不会(x)觉得这些很像空话，很不实际Q?br /> 但是有做L有好处的Q?br /> 你做得越多，你越q程序员p快。因Z不能Q也不想只是单线E的E序员！
]]>
]]>
l对~程的好处：(x)
]]>
或者：(x) 下蝲
摘录一些东西，具体请下附g观看Q?br />
CRC卡”?span lang="EN-US">CRC是ClassQ?span lang="EN-US">Responsibility?span lang="EN-US">Collaboration的简U?span lang="EN-US">
卡的典型应用
CRC卡，而不用文档或者更先进?span lang="EN-US">UML工具Q?span lang="EN-US">
卡片上面的空间很，q样可以防止我们给q个cd多的职责。如果一个类的职责太多的话（比如Q超q?span lang="EN-US">4个）(j)Q尝试以更抽象的方式去考虑一下，职责划分?span lang="EN-US">
卡主要是用在探烦(ch)或者讨论类的设计的阶段。如果我们觉得这个设计不行的话，我们既不用修Ҏ(gu)档，也不用修改类图，只要把卡片丢?jin)就行?jin)。此外，一旦设计完成，我们可以把所有的卡丢?jin)。它们不是用来做文档的?span lang="EN-US">
如果我们觉得现在的卡片不合适，之前设计的比较好Q我们只要简单的把之前的卡片拿出来组合就行了(jin)?/span>
]]>
http://wingel.javaeye.com/topics/download/ce15b67a-1df7-4a75-8f03-1a505aca35d8
请从链接中下载，下面的内容只是摘要?br />
处理三种cd的代码都混在?jin)一P(x)
(tng) (tng) 1.UI: JDialog, JTextField, 响应用户事g的代码?br />
(tng) (tng) 2.数据库访? Connection, PreparedStatement, SQL statements, ResultSet {等?br />
(tng) (tng) 3.域逻辑: 参会(x)者的默认idQ参?x)者的名字必填Q所属地区的限制{等。域逻辑又称为“域模型”或者“业务逻辑”?br />
q三个不同类别的代码混在一P?x)造成下面的问题：(x)
1.代码很复杂?br />2.代码很难重用。如果我们想创徏一个EditParticipantDialogQ让用户更改参会(x)者的信息Q我们就想重用部分域逻辑Q比如，地区的限Ӟ(j)。但实现q部分域逻辑的代码跟AddParticipantDialog混在?jin)一PҎ(gu)不能重用。如果是在一个webpȝ中，更N用了(jin)?br />3.代码很难试。每ơ要这L(fng)一D代码，我们都要Z个数据库Q还要通过一个用h作界面来试?br /> (tng) (tng) (tng) (tng) 4.如果数据库表l构更改?jin)，AddParticipantDialogq个c，q有其他的很多地斚w要跟着更改?br />5.它导致我们一直在考虑一些低层的太细节的概念Q比如数据库字段Q表的记录之cȝQ而不是类Q对象，Ҏ(gu)和属性这一cȝ概念。或者说白了(jin)一点，一直在考虑怎么往数据库里面装数据Q而没有了(jin)面向对象的概念，没有?jin)徏立业务模型的思维?br />
因此Q我们应该将q三U类别的代码分离开QUIQ数据库讉KQ域逻辑Q。?tng) (tng) (tng) (tng) (tng) (tng) (tng)?a href="/Files/Wingel/%E7%AC%AC7%E7%AB%A0%E5%B0%86%E6%95%B0%E6%8D%AE%E5%BA%93%E8%AE%BF%E9%97%AE%EF%BC%8CUI%E5%92%8C%E5%9F%9F%E9%80%BB%E8%BE%91%E5%88%86%E7%A6%BB.rar">
]]>
or
http://wingel.javaeye.com/topics/download/afd36f87-a11b-4d18-a01b-a843092ec1bc
如果现在有一个类ParentQ里面有个属性的cd是ChildQadd的方法里面还有个参数的类型是GirlQ?br /> class Parent{
(tng) (tng) (tng) (tng) (tng) (tng) (tng) Child child;
(tng)void add(Girl girl){
(tng) (tng) (tng) ...
(tng)}
(tng) (tng) (tng) (tng) }
(tng) (tng) (tng) (tng) 因ؓ(f)上面Parent里面用到?jin)Child跟Girlq两个类Q我们就_(d)Parent引用?jin)类Child跟类Girl。现在的问题是，如果Childq个cL者Girlq个cȝ译不q的话，那么Parentq个cM~译不了(jin)?jin)。也是_(d)Parent依赖于Child跟Girl。这章讲q的Q就是因Z些类的依赖造成的无法重用的问题?br />
具体的内容在上面的下载链接里面的pdf文g里，内容较多?/font>
]]>
http://www.aygfsteel.com/Wingel/category/17919.html
]]>
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414843.aspx
]]>
<link rel="stylesheet" type="text/css" href="<ww:url value="/layout/css/style.css"/>"><link rel="stylesheet" type="text/css" href="<ww:url value="/layout/css/thickbox.css"/>" media="screen"><script type="text/javascript" src="<ww:url value="/layout/js/jquery-latest.js"/>"></script><script type="text/javascript" src="<ww:url value="/layout/js/thickbox.js"/>"></script><script type="text/javascript" src="<ww:url value="/layout/js/wingel.js"/>"></script><decorator:head/>
然后在要验证的Form里面加个属性validatable=true,如下Q?br />
<form id="registerForm" action="user-register.action" validatable="true">注意Q这边不要加onsubmitҎ(gu)
接下来，好?jin)，比如说有个输入框Q?br /><ww:textfield name="name" id="name"/>
我想验证Q让它必填，如下可以了(jin)Q?br /><label for="name" validate="required">请填写名U?lt;/label> 其中 for属性里面填的要是验证的输入框idQvalidate填的是验证方?;label里面的文本就是验证不q的时候要昄的信息?br />如果我想验证一个输入框的输入值长度怎么办，q样子就行了(jin)
<label for="password" validate="lengthRange:6;20">后面的参数用Q号隔开Q验证的Ҏ(gu)名跟参数? 隔开?br />wingel.js里面已经包括?jin)一些常用的验证Ҏ(gu)Q现在问题来?jin)，如果要自定义验证?gu)怎么办，如下办：(x)
比如你想加个验证Ҏ(gu)是hello
则label里面的validate属性写成hello,
然后加一个JavaScriptҎ(gu)Q?br />
<script type="text/javascript">
wingel.validator.rules.hello= function(value, element, parameters,utility) {
wingel.validator.rules.hello
... }</script>
里面三个参数,shit , couldn't input Chinese. now English will be used.
The first parameter is the value of the input element you want to validate,the second one is the validated element, the third one, is the parameters you add in validate label, the last one, is a utility class, you can invoke its method to make your code easier.
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414856.aspx
]]>
PROPAGATION_MANDATORY:
Indicates that the method must run within a transaction. If no
existing transaction is in progress, an exception will be thrown.
PROPAGATION_NESTED:
Indicates that the method should be run within a nested transaction
if an existing transaction is in progress. The nested transaction
can be committed and rolled back individually from the enclosing
transaction. If no enclosing transaction exists, behaves like
PROPAGATION_REQUIRED:
Beware that vendor support for this propagation behavior is spotty at best.
Consult the documentation for your resource manager to determine if nested
transactions are supported.
PROPAGATION_NEVER:
Indicates that the current method should not run within a transactional
context. If there is an existing transaction in progress, an
exception will be thrown.
PROPAGATION_NOT_SUPPORTED:
Indicates that the method should not run within a transaction. If an
existing transaction is in progress, it will be suspended for the
duration of the method. If using JTATransactionManager,
access to TransactionManager is required.
PROPAGATION_REQUIRED:
Indicates that the current method must run within a transaction. If
an existing transaction is in progress, the method will run within
that transaction. Otherwise, a new transaction will be started.
PROPAGATION_REQUIRES_NEW:
Indicates that the current method must run within its own transaction.
A new transaction is started and if an existing transaction is in
progress, it will be suspended for the duration of the method. If
using JTATransactionManager, access to Transaction-
Manager is required.
PROPAGATION_SUPPORTS:
Indicates that the current method does not require a transactional
context, but may run within a transaction if one is already in
progress.
Isolation levels:
In a typical application, multiple transactions run concurrently, often working
with the same data to get their job done. Concurrency, while necessary, can lead
to the following problems:
?Dirty read—Dirty reads occur when one transaction reads data that has
been written but not yet committed by another transaction. If the
changes are later rolled back, the data obtained by the first transaction
will be invalid.
?Nonrepeatable read—Nonrepeatable reads happen when a transaction performs
the same query two or more times and each time the data is different.
This is usually due to another concurrent transaction updating the
data between the queries.
?Phantom reads—Phantom reads are similar to nonrepeatable reads. These
occur when a transaction (T1) reads several rows, then a concurrent transaction
(T2) inserts rows. Upon subsequent queries, the first transaction
(T1) finds additional rows that were not there before.
Isolation level:
ISOLATION_DEFAULT:
Use the default isolation level of the underlying datastore.
ISOLATION_READ_UNCOMMITTED:
Allows you to read changes that have not yet been committed. May
result in dirty reads, phantom reads, and nonrepeatable reads.
ISOLATION_READ_COMMITTED:
Allows reads from concurrent transactions that have been committed.
Dirty reads are prevented, but phantom and nonrepeatable
reads may still occur.
ISOLATION_REPEATABLE_READ:
Multiple reads of the same field will yield the same results, unless
changed by the transaction itself. Dirty reads and nonrepeatable
reads are prevented by phantom reads may still occur.
ISOLATION_SERIALIZABLE:
This fully ACID-compliant isolation level ensures that dirty reads,
nonrepeatable reads, and phantom reads are all prevented. This is
the slowest of all isolation levels because it is typically accomplished
by doing full table locks on the tables involved in the transaction.
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414826.aspx
]]>
有一回在|上看到hibernate3.2的发布了(jin). 一直都觉得写hibernate的映文件是一件苦力活,于是决定用一下hibernate的annotation.
升的时?q真是出?jin)一堆的问题.
要嘛是ecache的问?要嘛又是hibernate的second level cache的问?用了(jin)spring自带的hibernate包有问题,但用?jin)hibernate|站上下载的包又有问?
最?去spring下了(jin)最新的发布版本,又用?jin)hibernate?.2GA?l于,问题解决?反正都升U了(jin)一部分?后面军_,webwork也升U到?jin)最新版?.24,
然后以前写的自定义标签再升U一?也像webwork一h用freemarker做模?
于是手头上一个在webwork+spring+hibernate的最新开发框架出炉了(jin).
军_,接下来的两个目qq个框架.
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414831.aspx
]]>
Webwork的Interceptor机制?br /> hIOCQ用Setter的注法Q不用配|文Ӟ加annotation卛_
而比较主要的是Q一ơ请求的程是，servlet自动Ҏ(gu)hL以名字匹配的Model和ControlQ将参数和值注入Model中，转到Controld理，再根据Control的结果去扄应的面。而这里有些是从MonoRails学来的特性：(x)~省的，面的文件名是Control中的Ҏ(gu)名，面所在的文g夹就是Control的类名。这样就省了(jin)很多配置和写跌{l果的代码?br /> 如果url中请求的Ҏ(gu)名在q个Control中不存在的话Q就l过interceptor后直接去N面，也就是说Q不l过Control处理?jin)?br /> q有一点，很爽的就是Layout的功能：(x)在Control里面的方法加个AnnotationQLayoutQLayout写上面框架的文件名Q则每个跌{出来的页面，自动内嵌到页面框架中。这点就有点cMsitemesh?jin)?br /> q个MVC框架做完后，大致p些功能。可惜，q个框架是一个自己写着好玩的框Ӟq没有用在实际的开发项目中Q如果有Z(x)的话Q还真希望将q个框架做完_(d)到网上开源?/p>
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414832.aspx
]]>
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414834.aspx
]]>
Q。在spring配置文g里面加：(x) (tng) (tng) (tng) <aop:spring-configured/>
Q。在要被理的类中加上annotation:@Configurable(autowire = Autowire.BY_TYPE) autowire应该不用说了(jin)。?tng)?img src="http://blog.csdn.net/Wingel/aggbug/1414836.aspx" height="1" width="1" />
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414836.aspx
]]>
/**
(tng)* (tng)Author: (tng)Wingel (tng)* (tng)Date: (tng)2006-7-29 (tng)* (tng)Time: (tng)15:17:52 (tng)
*/
public
(tng)
enum
(tng)ApplicationState (tng)
{ (tng) (tng) (tng) (tng)unsubmited(
"
未提?/span>
"
),submited(
"
提交/待审?/span>
"
),agreed(
"
通过
"
),disagreed(
"
未同?/span>
"
); (tng) (tng) (tng) (tng)
private
(tng)String (tng)text;
(tng) (tng) (tng) (tng)ApplicationState(String (tng)text) (tng)
{
(tng) (tng) (tng) (tng)ApplicationState(String (tng)text) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
this
.text (tng)
=
(tng)text;
(tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)
public
(tng)
static
(tng)String (tng)getText(String (tng)name) (tng)
{
//
l页面用的方?/span>
(tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
try
(tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)valueOf(name).text; (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)}
(tng)
catch
(tng)(Exception (tng)e) (tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)
null
; (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)
public
(tng)String (tng)getText() (tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)text; (tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)
public
(tng)
static
(tng)
boolean
(tng)isValidRegisterType(String (tng)name) (tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
try
(tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)valueOf(name); (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)
true
; (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)}
(tng)
catch
(tng)(Exception (tng)e) (tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)
false
; (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)
public
(tng)
static
(tng)ApplicationState[] (tng)allState() (tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)ApplicationState.values(); (tng) (tng) (tng) (tng)}
(tng) (tng) (tng) (tng)
public
(tng)String (tng)getName() (tng)
{ (tng) (tng) (tng) (tng) (tng) (tng) (tng) (tng)
return
(tng)
this
.name(); (tng) (tng) (tng) (tng)}
}
(tng)
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414837.aspx
]]>
Webwork的Interceptor机制?br /> hIOCQ用Setter的注法Q不用配|文Ӟ加annotation卛_
而比较主要的是Q一ơ请求的程是，servlet自动Ҏ(gu)hL以名字匹配的Model和ControlQ将参数和值注入Model中，转到Controld理，再根据Control的结果去扄应的面。而这里有些是从MonoRails学来的特性：(x)~省的，面的文件名是Control中的Ҏ(gu)名，面所在的文g夹就是Control的类名。这样就省了(jin)很多配置和写跌{l果的代码?br /> 如果url中请求的Ҏ(gu)名在q个Control中不存在的话Q就l过interceptor后直接去N面，也就是说Q不l过Control处理?jin)?br /> q有一点，很爽的就是Layout的功能：(x)在Control里面的方法加个AnnotationQLayoutQLayout写上面框架的文件名Q则每个跌{出来的页面，自动内嵌到页面框架中。这点就有点cMsitemesh?jin)?br /> q个MVC框架做完后，大致p些功能。可惜，q个框架是一个自己写着好玩的框Ӟq没有用在实际的开发项目中Q如果有Z(x)的话Q还真希望将q个框架做完_(d)到网上开源?/p>
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414832.aspx
]]>
后来才听说了(jin)QC++有个库，叫QTQ写桌面E序非常的不错，q很像swing。呜Q真TMD的知道得太迟?jin)，q被折磨?jin)很久，查?jin)暴多的资料?nbsp;
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414851.aspx
]]>
文章来源:http://blog.csdn.net/Wingel/archive/2006/05/26/756667.aspx
]]>
4月䆾在这家公司是以实?fn)的w䆾q来的，在实?fn)的q段旉里，学会(x)?jin)struts和hibernate的基本应用，也用q些框架做了(jin)几个模块，׃中间要用C(jin)一个统计图Q去?jin)解q用了(jin)jfreechartQ也是Z学了(jin)q个开源包吧。等我向目l理认我可以像其他员工那样完成dӞ才结束实?fn)，回到学校?br />7月䆾毕业出来Q回到这家公司时Q一开始做的，是l护公司自己的OApȝ。第一个周是做原有模块的修改工作，因ؓ(f)在学校玩?jin)一D|_(d)刚开始做的时候还真有一点生疏?br />W一周的d完成后，开始上手了(jin)Q接下来Q就开始增加两个新的客服模块了(jin)。这两个模块用了(jin)三周完成Q这三周里面Q也学会(x)?jin)一个不h技术（后来才知道，q叫AJAXQ，q在剩余的时间里面，比较大的收获是掌握struts-menuq个开源包的普通应用?br />8月䆾的前两周Q开始做一个薪资审批的程Q也是W一ơ接触工作流?jin)。这ơ做的这个模块，q没有学C么新的技术，是业务逻辑很复杂的~码Q但是想惌׃刚工作没几天Q这Ҏ(gu)来说Q益处还是很大的?br />接下来的一个月里，做的是OA上添加一个跟企业的运营^台整合的企业注册模块Q这个模块一开始看业务q是很简单的Q后面功能慢慢添加，竟然做了(jin)一个月才完成，期间抽空?jin)解了(jin)一个在U编辑HTML的东西，而我想主要的Q这个模块有很多H发事情和意外要考虑Q跟以前做的那些相比Q这ơ的q个模块Q才是真正煅炼程序员能力的编E吧。已l到9月䆾的中旬了(jin)Q那个注册企业模块也l于完成?jin)，现在回想hQ做那个模块用的最有技术含量的东西Q应该是Ajax吧（我ƈ没有用什么开源的框架Q如果现在有嘲笑我说Q?ldquo;你只是简单的一些不hQ也叫AjaxQ真搞笑”Q那我只能说Q果然不只文人相轻，做技术的也看不v做技术的?br />刚做完这个企业注册模块，目l理问我_(d)(x)“你以前用q?net吧?rdquo;我告诉他是的。没惛_Q一个Q务就来了(jin)Q把一个aspx+CQ做的短信业务管理网站修改一下，因ؓ(f)里面都是适合SQL Server的数据库Q而且用到很多存储q程Q现在想q到Oracle上。而一周后Q终于把q个目Ҏ(gu)适合于大部分数据库的Q去掉所有存储过E，增加?jin)日志功能。这ơ的q个dq没有煅炼到什么J2EE的能力，不过也算再熟(zhn)了(jin)一?netQ再ơ感受了(jin).net的h性化Q可惜有些功能还是没有Jbuilder好。接下来Q又是修改了(jin)一个用EJB做的工程中的一些模块，和调整了(jin)OA中的一部分功能。期间也搭徏?jin)一个Tapetry的工E，是感受?jin)他一把，不过q是觉得它没有struts好用?br />接下来的q个月，是我学得最多的一个月?jin)，公司x(chng)以前的C/S二层l构改ؓ(f)Delphi做界面层Q用Java做中间服务层Q最后再数据库服务器。他们最先考虑到的是，用Web service。于是我的Q务就下来?jin)，研究一下Web serviceQ看能不能做一个框架满_司的要求。我起先则是用axis搭徏Web serviceQ后面发玎ͼ它实在是太复杂了(jin)Q我们想要的Q是一个轻便的东西?br />于是开始研ISOAPQ因为Web service是用SOAP传输消息的。SOAP然后是javaxmlQ接着是dom4j。后面自׃生一个想法，如果客户端传递一个XMLq来Q里面描qC(jin)要调用的cdQ方法名Q以?qing)传递的参数对象。服务端解析XML出来Q利用反和序列化XML转化为本地化持久c，调用q个Ҏ(gu)Q再把方法返回的对象转化成XMLq回l客L(fng)Q那么不可以实现客L(fng)自由调用Java的类?br />有了(jin)q个x(chng)很兴奋Q开始付诸行动。很q运的是Q我把这个框架做出来?jin)，那时候真的很兴奋Q因为在我以为，那个cd的框架还?sh)来没有看见q，甚至x(chng)它放到网上开源去。后来才意识刎ͼ其实我做的这个就是轻量的没有规范的Web service。既然这P后面又给它增加了(jin)cMwsdl的功能，而后在优化过E中Q增加了(jin)cacheQ大大提高(sh)(jin)速度。测试了(jin)一下，完成同样的功能，它的速度是axis web service的十几倍，呵呵呵呵呵，那个叫兴奋啊。那也是一ơ性把Java的几关键技术一起学?jin)顺便用上?br />做完q个框架后，又做?jin)两个项目，一个交通查询的Q一个社区。这两个目倒也没有像上回研IWeb service那样令h兴奋的感觉了(jin)Q不q还是学?jin)很多开源的东西和技术。交通查询很多次用到AjaxQ也在做q个目的期_(d)学到?jin)几J2EE技术，在一ơ用JavaScript做树(wi)状控件的时候，大大?jin)提高?sh)(jin)Javascript一把。而在此，也ȝ刎ͼ其实懂不懂一些新技术ƈ不是主要的，新技术很Ҏ(gu)可以用hQ而难的，是军_用哪些技术，哪些模式Q直臛_何来设计自己的框架。很q运的，在做那个C的时候，我有?jin)这个机会(x)?br />除此以外Q做q个C的时候其他o(h)人激动的事情Q就是写?jin)几个复杂的标签Q学C(jin)velocity和全文搜索Lucene?jin)?br />之后Q就是发生一些跟技术无关的不如人意的事情了(jin)?br />
文章来源:http://blog.csdn.net/Wingel/archive/2006/05/26/756669.aspx
]]>
而这回，我想一下第二个观点Q?nbsp;
其实如果你对代码要求比较严格的话Q你׃(x)l常发现Q你的代码有很多东西可以抽取出来Q或者做在公q模块Q或者作为框架的底层Q我们就单的拿jdbc来说吧，
首先Q是connection的管理，q点一般用jdbc熟一些的话，都会(x)有管理connection的公共模块，虽然偶尔?x)碰到性能的问题，但是q点我们暂且压下不表?br /> 我们查询的时候，每次都要?br /> rs.get...("name"),
rs.get...("id"),
rs.get...("age"),
rs.get...("gender"),
rs.get...("hobby"),
然后修改数据库的时候，q要拼写update语句跟insert语句Q经常还要费很多旉来调试这些多余代码的问题Q这时候你想Q不行，我一定要写一个公共模块，省得让我每次都要定这么多代码Q于是你的第一个公共模块生了(jin)Q然后测试啊试Q改q啊改进Q叮叮响Q过?jin)几天时间的考验Q这个公共模块终于可以放?j)用?jin)Q项目进度开始快一些了(jin)Qȝ不用再拼SQL?jin)?br />
后来在做l计模块的时候，H然又发玎ͼ之前在用到的一些SQL函数Q好像在客户要求的数据库上不怎么行啊Q于是又L?jin)一下资料，又过?jin)几天（可能q次不用几天Q，然后l于攑ֿ(j)Q所有的函数都正怺(jin)?br />
接着又不可避免的到?jin)分늚问题Q你对自pQ不用怕，我上回就写了(jin)一个分늚Q没有问题！可是Ya的你H然发现Q上回的那个分页是用游标实现的，q回客户是要求用SQLServerQ唉QSQLServer的游标，不提也ŞQ想来想去，只好自己拼SQL语句来写分页?jin)，又是count又是topQ测?jin)又之后，又过了(jin)几天，啊哈Q终于分늚公共模块也做好的Q可以放?j)用?jin)Q好Q项目的q度又可以加快了(jin)?/p>
做着做着的时候，发现Q咦Q好像这表得增加一个字D|行，增加?jin)，然后所有查询的SQL语句加一下，所有insert跟update的代码修改一下，面修改一下，嗯，现在应该正常?jin)，看v来倒是没什么问题，咦，报表好像不怎么对啊Q靠Q这边还有调用这个表的代码，妈的Q改吧改吧。磨y了(jin)好几个小Ӟ当然Q熟l的话，q不用几个小Ӟ(j)Qȝ看v来都正常?jin)?/p>
q一回，q个功能中有一ơ用戯求，讉K?jin)好几次数据库，不行Q这里应该用个cacheQ否则性能上会(x)有问题啊Q算?jin)，用算法解决一下，量访问数据库好了(jin)Q我对cacheq(sh)熟呢。（写啊写啊QBatch SizeQ这样多Q那样多QFetch Size。。。，l于Q看h正常一些了(jin)Q。过?jin)一D|_(d)靠，q边又要讉K好几ơ数据库QYa的受不鸟?jin)，性能爱咋的咋的，反正一个地Ҏ(gu)又不要紧。Oh shit!!!q边也是好几ơ，q边又是好几ơ，那边又是好几ơ。不行了(jin)Q我老老实实写个cache支持吧，于是又叮叮当当了(jin)好几天，l于Q有个粗p的cache出来?jin)，l于速度可以看一些了(jin)。后来改q又改进Q测试又试Q篏M(jin)Q老子好不爽啊?/p>
好像天下有点太^?jin)，啊，你说我这个地方忘记更C增加的那个子表啊Q算?jin)，没关p，我明天看一下代码，q个Ҏ(gu)解决炏V嗯Q我改了(jin)那边的代码了(jin)Q会(x)更新子表信息?jin)。什么？你说取主表的记录跟相应的子表记录列表ȝ(ch)啊，没关p，我更C下处理resultset的公共模块，明天再说?br /> Oh shit......对这样子复杂的查询好像现在的公共模块支持不了(jin)啊，了(jin)Q这样子的查询不要用q个公共模块Q我们手动写一些代码好啊，别跟我讲q样代码l构很难看，你以为我不知道啊QTMD?/p>
TMD的，怎么q边的SQL老是q行不了(jin)啊，不会(x)是分底层模块的问题吧，靠，怎么你的SQL语句有这么多order,group byQ靠Q还有top啊，q当然过不了(jin)?jin)，不要吵?jin)Q现在时间改Q不理它Q直接用个假分页p?jin)。你又说代码l构隄Q小?j)我抽你哦?/p>
公司新来一个程序员Q看?jin)几天代码，不停的抱怨说Q这代码写得真差啊。。。。。?nbsp;
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414852.aspx
]]>
PROPAGATION_MANDATORY:
Indicates that the method must run within a transaction. If no
existing transaction is in progress, an exception will be thrown.
PROPAGATION_NESTED:
Indicates that the method should be run within a nested transaction
if an existing transaction is in progress. The nested transaction
can be committed and rolled back individually from the enclosing
transaction. If no enclosing transaction exists, behaves like
PROPAGATION_REQUIRED:
Beware that vendor support for this propagation behavior is spotty at best.
Consult the documentation for your resource manager to determine if nested
transactions are supported.
PROPAGATION_NEVER:
Indicates that the current method should not run within a transactional
context. If there is an existing transaction in progress, an
exception will be thrown.
PROPAGATION_NOT_SUPPORTED:
Indicates that the method should not run within a transaction. If an
existing transaction is in progress, it will be suspended for the
duration of the method. If using JTATransactionManager,
access to TransactionManager is required.
PROPAGATION_REQUIRED:
Indicates that the current method must run within a transaction. If
an existing transaction is in progress, the method will run within
that transaction. Otherwise, a new transaction will be started.
PROPAGATION_REQUIRES_NEW:
Indicates that the current method must run within its own transaction.
A new transaction is started and if an existing transaction is in
progress, it will be suspended for the duration of the method. If
using JTATransactionManager, access to Transaction-
Manager is required.
PROPAGATION_SUPPORTS:
Indicates that the current method does not require a transactional
context, but may run within a transaction if one is already in
progress.
Isolation levels:
In a typical application, multiple transactions run concurrently, often working
with the same data to get their job done. Concurrency, while necessary, can lead
to the following problems:
?Dirty read—Dirty reads occur when one transaction reads data that has
been written but not yet committed by another transaction. If the
changes are later rolled back, the data obtained by the first transaction
will be invalid.
?Nonrepeatable read—Nonrepeatable reads happen when a transaction performs
the same query two or more times and each time the data is different.
This is usually due to another concurrent transaction updating the
data between the queries.
?Phantom reads—Phantom reads are similar to nonrepeatable reads. These
occur when a transaction (T1) reads several rows, then a concurrent transaction
(T2) inserts rows. Upon subsequent queries, the first transaction
(T1) finds additional rows that were not there before.
Isolation level:
ISOLATION_DEFAULT:
Use the default isolation level of the underlying datastore.
ISOLATION_READ_UNCOMMITTED:
Allows you to read changes that have not yet been committed. May
result in dirty reads, phantom reads, and nonrepeatable reads.
ISOLATION_READ_COMMITTED:
Allows reads from concurrent transactions that have been committed.
Dirty reads are prevented, but phantom and nonrepeatable
reads may still occur.
ISOLATION_REPEATABLE_READ:
Multiple reads of the same field will yield the same results, unless
changed by the transaction itself. Dirty reads and nonrepeatable
reads are prevented by phantom reads may still occur.
ISOLATION_SERIALIZABLE:
This fully ACID-compliant isolation level ensures that dirty reads,
nonrepeatable reads, and phantom reads are all prevented. This is
the slowest of all isolation levels because it is typically accomplished
by doing full table locks on the tables involved in the transaction.
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414826.aspx
]]>
SQL server实现?jin)四个?br />Oracle只实C间两个别?/font>
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414835.aspx
]]>
Transaction A Transaction B
begin begin
query1 from table A
....... modify1 to table A(insert/update/delete)
commit
query2 from table A
commit
那么在事务A中，query1跟query2查询出来的结果是否一样呢Q这p事务隔离U别有关?jin)?br />SQL的标准定义里面，一共有四种U别Q?br />read uncommitedd未提交的数据 是其他事务求提交的数据Q都可以d出来
read commitedd已提交的数据 query2?x)跟query1d的数据不一?br />repeatable read可重复读取，即query1跟query2d的数据是一L(fng)
serializable 序列化，
SQL 标准用三个必dq行的事务之间避免的现象定义?jin)四个别的事务隔离?q些不希望发生的现象是：(x)
脏读Qdirty readsQ?br /> 一个事务读取了(jin)另一个未提交的ƈ行事务写的数据?
不可重复读（non-repeatable readsQ?br /> 一个事务重新读取前面读取过的数据， 发现该数据已l被另一个已提交的事务修改过?
q读Qphantom readQ?br /> 一个事务重新执行一个查询，q回一套符合查询条件的行， 发现q些行因为其他最q提交的事务而发生了(jin)改变?
隔离U别 脏读QDirty ReadQ?nbsp; 不可重复读（NonRepeatable ReadQ?nbsp; q读QPhantom ReadQ?br />L提交QRead uncommittedQ?nbsp; 可能 可能 可能
d提交QRead committedQ? 不可能 可能 可能
可重复读QRepeatable readQ?nbsp; 不可?nbsp; 不可?nbsp; 可能
可串行化QSerializable Q?nbsp; 不可?nbsp; 不可?nbsp; 不可?/p>
SQLServer
我们首先在SQLServer上做?jin)实验，SQLServer一共支持四U隔ȝ别，read uncommited跟read commited我们没有实验Q我们直接先实验
repeatable readQ如果事务A定义?jin)如下别，那么当事务B执行到modify1q条语句Ӟ如果modify1是update的话Q就被锁hQ一直等
C务A提交完以后，锁才?x)被释放Q而如果是insert的话Q刚可以利q行下去Q然后在事务A中，query2查到的数据，是已l被事务B
修改q的数据Q如果将U别定义在serializable的话Q则在modify1语句中，update,insert或者delete都会(x)被锁掉?br />也就是说QSQLServer对这些别的支持Q是通过锁来做到的，虽然可以保证事务正常q行Q但是ƈ行的性能却很差?br />Oracle
oracle只支持两个别，read commited跟serializableQ结果这里就不用详细说明Q实验的l果是，oracle的serializable是通过版本
控制来完成的Q而不是通过锁机Ӟq就保证?jin)ƈ行的性能。Oracle的默认别是read commited
MySQL
MySQLServer也支持四个别，而且MySQL也是通过版本控制而非锁机制来完成的?br />
文章来源:http://blog.csdn.net/Wingel/archive/2006/11/26/1414839.aspx
]]>