当前位置:  编程技术>软件工程/软件设计

[1]简单工厂模式--使用--扩展--缺点--适用情况--

简单工厂模式的一些基本知识

本文主要参见自《Java与模式》

一、UML类图表示如下

二、角色如下

1.Product接口：

a.含有具体产品的共有接口。

2.ProductA,ProductB

a.工厂类所创建的这个角色的实例。

3.Factroy:

a.简单工厂模式的核心。

b.工厂类在客户端的直接调用下创建产品对象。

三、基本代码实现如下：

1.Product接口：

package com.jue.dp;
public interface Fruit {

	void grow();
	
	void harvest();
	
	void plant();
}

2.ProductA,ProductB

package com.jue.dp;

public class Apple implements Fruit {
	public static final String APPLE = "apple";

	@Override
	public void grow() {
		LogUtil.logs(APPLE, "grow");
	}

	@Override
	public void harvest() {
		LogUtil.logs(APPLE, "harvest");
	}

	@Override
	public void plant() {
		LogUtil.logs(APPLE, "plant");
	}

}

其他雷同！

3.Factroy:

package com.jue.dp;

public class FruitFactory {

	public static Fruit newInstance(String name) {
		if (name.equals("Apple")) {
			return new Apple();
		}

		if (name.equals("Grape")) {
			return new Grape();
		}

		if (name.equals("Strawberry")) {
			return new Strawberry();
		}
		throw new BadFruitException(name);
	}

	/**
	 * @param args
	 */
	public static void main(String[] args) {
		// TODO Auto-generated method stub
		Fruit f1 = FruitFactory.newInstance("Apple");
		f1.plant();
		f1.grow();
		f1.harvest();
	}
}

四、优点：

1.责任分割：客户端可以免除直接创建产品对象的责任，客户端无需考虑如何创建，只需要知道创建什么以及负责消费产品。

五、缺点

1.开闭原则：

a.客户端符合开比原则

b.工厂类不符合：需要修改核心类，增加新的产品将导致工厂类的修改。

2.Factory变得臃肿：当产品比较多的时候，factory会无比臃肿。

3.无法继承：简单静态工厂的创建方法通常是static的，导致工厂角色无法形成继承结构。

六、扩展 : 使用反射可以创建任意产品对象。

public class FruitFactory2 {

	public static Fruit newInstance(String name) {

		Fruit f = null;
		try {
			f = (Fruit) Class.forName(name).newInstance();
		} catch (Exception e) {
			e.printStackTrace();
		}
		return f;
	}
	
	public static void main(String args[]){
		Fruit f1 = FruitFactory2.newInstance("com.jue.dp.Apple");
		f1.plant();
		f1.grow();
		f1.harvest();
	}
}

七、使用场景

a.简单工厂模式所创建的对象往往属于一个产品等级。

b.创建的对象比较少。

c.客户端关心于创建什么以及如何消费，而不关心如何创建。

八、简单工厂模式与单例模式的关系

1.抽象产品省略，工厂角色与具体产品合并,这种退化的简单工厂模式与单例有很多相似之处，但是与单例是不同的。

2.不同之处：

工厂模式总是创建新的实例。

单例模式要求单例类的构造器是私有的，禁止外部客户端对其实例化。

3. 联系

单例模式是建立在简单静态工厂模式基础之上的。 

[2]【便利贴】Struts默认标配

今天看了下Struts2 的struts-default.properties 配置，发现路还很长，拷了下官网的 Struts2默认配置，下还包括最新的struts-default.xml

### Struts default properties
###(can be overridden by a struts.properties file in the root of the classpath)
###

### Specifies the Configuration used to configure Struts
### one could extend org.apache.struts2.config.Configuration
### to build one's customize way of getting the configurations parameters into Struts
# struts.configuration=org.apache.struts2.config.DefaultConfiguration

### This can be used to set your default locale and encoding scheme
# struts.locale=en_US
struts.i18n.encoding=UTF-8

### if specified, the default object factory can be overridden here
### Note: short-hand notation is supported in some cases, such as "spring"
###       Alternatively, you can provide a com.opensymphony.xwork2.ObjectFactory subclass name here
# struts.objectFactory = spring

### specifies the autoWiring logic when using the SpringObjectFactory.
### valid values are: name, type, auto, and constructor (name is the default)
struts.objectFactory.spring.autoWire = name

### indicates to the struts-spring integration if Class instances should be cached
### this should, until a future Spring release makes it possible, be left as true
### unless you know exactly what you are doing!
### valid values are: true, false (true is the default)
struts.objectFactory.spring.useClassCache = true

### ensures the autowire strategy is always respected.
### valid values are: true, false (false is the default)
struts.objectFactory.spring.autoWire.alwaysRespect = false

### if specified, the default object type determiner can be overridden here
### Note: short-hand notation is supported in some cases, such as "tiger" or "notiger"
###       Alternatively, you can provide a com.opensymphony.xwork2.util.ObjectTypeDeterminer implementation name here
### Note: By default, com.opensymphony.xwork2.util.DefaultObjectTypeDeterminer is used which handles type detection
###       using generics. com.opensymphony.xwork2.util.GenericsObjectTypeDeterminer was deprecated since XWork 2, it's
###       functions are integrated in DefaultObjectTypeDeterminer now.
###       To disable tiger support use the "notiger" property value here.
#struts.objectTypeDeterminer = tiger
#struts.objectTypeDeterminer = notiger

### Parser to handle HTTP POST requests, encoded using the MIME-type multipart/form-data
# struts.multipart.parser=cos
# struts.multipart.parser=pell
struts.multipart.parser=jakarta
# uses javax.servlet.context.tempdir by default
struts.multipart.saveDir=
struts.multipart.maxSize=2097152

### Load custom property files (does not override struts.properties!)
# struts.custom.properties=application,org/apache/struts2/extension/custom

### How request URLs are mapped to and from actions
#struts.mapper.class=org.apache.struts2.dispatcher.mapper.DefaultActionMapper

### Used by the DefaultActionMapper
### You may provide a comma separated list, e.g. struts.action.extension=action,jnlp,do
### The blank extension allows you to match directory listings as well as pure action names
### without interfering with static resources, which can be specified as an empty string
### prior to a comma e.g. struts.action.extension=, or struts.action.extension=x,y,z,,
struts.action.extension=action,,

### Used by FilterDispatcher
### If true then Struts serves static content from inside its jar.
### If false then the static content must be available at <context_path>/struts
struts.serve.static=true

### Used by FilterDispatcher
### This is good for development where one wants changes to the static content be
### fetch on each request.
### NOTE: This will only have effect if struts.serve.static=true
### If true -> Struts will write out header for static contents such that they will
###             be cached by web browsers (using Date, Cache-Content, Pragma, Expires)
###             headers).
### If false -> Struts will write out header for static contents such that they are
###            NOT to be cached by web browser (using Cache-Content, Pragma, Expires
###            headers)
struts.serve.static.browserCache=true

### Set this to false if you wish to disable implicit dynamic method invocation
### via the URL request. This includes URLs like foo!bar.action, as well as params
### like method:bar (but not action:foo).
### An alternative to implicit dynamic method invocation is to use wildcard
### mappings, such as <action name="*/*" method="{2}" class="actions.{1}">
struts.enable.DynamicMethodInvocation = true

### Set this to true if you wish to allow slashes in your action names.  If false,
### Actions names cannot have slashes, and will be accessible via any directory
### prefix.  This is the traditional behavior expected of WebWork applications.
### Setting to true is useful when you want to use wildcards and store values
### in the URL, to be extracted by wildcard patterns, such as
### <action name="*/*" method="{2}" class="actions.{1}"> to match "/foo/edit" or
### "/foo/save".
struts.enable.SlashesInActionNames = false

### use alternative syntax that requires %{} in most places
### to evaluate expressions for String attributes for tags
struts.tag.altSyntax=true

### when set to true, Struts will act much more friendly for developers. This
### includes:
### - struts.i18n.reload = true
### - struts.configuration.xml.reload = true
### - raising various debug or ignorable problems to errors
###   For example: normally a request to foo.action?someUnknownField=true should
###                be ignored (given that any value can come from the web and it
###                should not be trusted). However, during development, it may be
###                useful to know when these errors are happening and be told of
###                them right away.
struts.devMode = false

### when set to true, resource bundles will be reloaded on _every_ request.
### this is good during development, but should never be used in production
struts.i18n.reload=false

### Standard UI theme
### Change this to reflect which path should be used for JSP control tag templates by default
struts.ui.theme=xhtml
struts.ui.templateDir=template
#sets the default template type. Either ftl, vm, or jsp
struts.ui.templateSuffix=ftl

### Configuration reloading
### This will cause the configuration to reload struts.xml when it is changed
struts.configuration.xml.reload=false

### Location of velocity.properties file.  defaults to velocity.properties
struts.velocity.configfile = velocity.properties

### Comma separated list of VelocityContext classnames to chain to the StrutsVelocityContext
struts.velocity.contexts =

### Location of the velocity toolbox
struts.velocity.toolboxlocation=

### used to build URLs, such as the UrlTag
struts.url.http.port = 80
struts.url.https.port = 443
### possible values are: none, get or all
struts.url.includeParams = none

### Load custom default resource bundles
# struts.custom.i18n.resources=testmessages,testmessages2

### workaround for some app servers that don't handle HttpServletRequest.getParameterMap()
### often used for WebLogic, Orion, and OC4J
struts.dispatcher.parametersWorkaround = false

### configure the Freemarker Manager class to be used
### Allows user to plug-in customised Freemarker Manager if necessary
### MUST extends off org.apache.struts2.views.freemarker.FreemarkerManager
#struts.freemarker.manager.classname=org.apache.struts2.views.freemarker.FreemarkerManager

### Enables caching of FreeMarker templates
### Has the same effect as copying the templates under WEB_APP/templates
struts.freemarker.templatesCache=false

### Enables caching of models on the BeanWrapper
struts.freemarker.beanwrapperCache=false

### See the StrutsBeanWrapper javadocs for more information
struts.freemarker.wrapper.altMap=true

### maxStrongSize for MruCacheStorage for freemarker, when set to 0 SoftCacheStorage which performs better in heavy loaded application
### check WW-3766 for more details
struts.freemarker.mru.max.strong.size=0

### configure the XSLTResult class to use stylesheet caching.
### Set to true for developers and false for production.
struts.xslt.nocache=false

### Whether to always select the namespace to be everything before the last slash or not
struts.mapper.alwaysSelectFullNamespace=false

### Whether to allow static method access in OGNL expressions or not
struts.ognl.allowStaticMethodAccess=false

### Whether to throw a RuntimeException when a property is not found
### in an expression, or when the expression evaluation fails
struts.el.throwExceptionOnFailure=false

### Logs as Warnings properties that are not found (very verbose)
struts.ognl.logMissingProperties=false

### Caches parsed OGNL expressions, but can lead to memory leaks
### if the application generates a lot of different expressions
struts.ognl.enableExpressionCache=true

### Indicates if Dispatcher should handle unexpected exceptions by calling sendError()
### or simply rethrow it as a ServletException to allow future processing by other frameworks like Spring Security
struts.handle.exception=true

最新struts-default.xml  时间20130508

<?xml version="1.0" encoding="UTF-8" ?>
<!--
/*
 * $Id$
 *
 * Licensed to the Apache Software Fou      

      
      
    

      

[3]双边滤波Matlab实现<The Bilateral Filter>

双边滤波与一般的高斯滤波的不同就是：双边滤波既利用了位置信息<or 几何信息——高斯滤波只用了位置信息>又利用了像素信息来定义滤波窗口的权重。

像素值越接近，权重越大。双边滤波会去除图像的细节信息，又能保持边界。

对于彩色图像，像素值的接近与否不能使用RGB空间值，双边滤波的原始文献建议使用CIE颜色空间。

代码如下：

function resultI = BilateralFilt2(I,d,sigma)
%%%
%Author：LiFeiteng
%Version：1.0——灰色图像  Time：2013/05/01
%Version：1.1——灰色/彩色图像  Time：2013/05/02  2013/05/05
%d 半窗口宽度
I = double(I);
if size(I,3)==1
    resultI = BilateralFiltGray(I,d,sigma);
elseif size(I,3)==3
    resultI = BilateralFiltColor(I,d,sigma);
else 
    error('Incorrect image size')    
end
end

function resultI = BilateralFiltGray(I,d,sigma)

[m n] = size(I);
newI = ReflectEdge(I,d);
resultI = zeros(m,n);
width = 2*d+1;
%Distance
D = fspecial('gaussian',[width,width],sigma(1));
S = zeros(width,width);%pix Similarity
h = waitbar(0,'Applying bilateral filter...');
set(h,'Name','Bilateral Filter Progress');
for i=1+d:m+d
    for j=1+d:n+d
        pixValue = newI(i-d:i+d,j-d:j+d);
        subValue = pixValue-newI(i,j);
        S = exp(-subValue.^2/(2*sigma(2)^2));
        H = S.*D;
        resultI(i-d,j-d) = sum(pixValue(:).*H(:))/sum(H(:)); 
    end
    waitbar(i/m);
end
close(h);
end

function resultI = BilateralFiltColor(I,d,sigma)
I = applycform(I,makecform('srgb2lab'));
[m n ~] = size(I);
newI = ReflectEdge(I,d);
resultI = zeros(m,n,3);
width = 2*d+1;
%Distance
D = fspecial('gaussian',[width,width],sigma(1));
% [X,Y] = meshgrid(-d:d,-d:d);
% D = exp(-(X.^2+Y.^2)/(2*sigma(1)^2));
S = zeros(width,width);%pix Similarity
h = waitbar(0,'Applying bilateral filter...');
set(h,'Name','Bilateral Filter Progress');
sigma_r = 100*sigma(2);
for i=1+d:m+d
    for j=1+d:n+d
        pixValue = newI(i-d:i+d,j-d:j+d,1:3);
        %subValue = pixValue-repmat(newI(i,j,1:3),width,width);
        dL = pixValue(:,:,1)-newI(i,j,1);
        da = pixValue(:,:,2)-newI(i,j,2);
        db = pixValue(:,:,3)-newI(i,j,3);
        S = exp(-(dL.^2+da.^2+db.^2)/(2*sigma_r^2));
        H = S.*D;
        H = H./sum(H(:));
        resultI(i-d,j-d,1) = sum(sum(pixValue(:,:,1).*H)); 
        resultI(i-d,j-d,2) = sum(sum(pixValue(:,:,2).*H));    
        resultI(i-d,j-d,3) = sum(sum(pixValue(:,:,3).*H));    
    end
    waitbar(i/m);
end
close(h);
resultI = applycform(resultI,makecform('lab2srgb'));
end

其中newI = ReflectEdge(I,d); %对称地扩展边界，在原始图像I的边界处镜像映射像素值 

function newI = ReflectEdge(I,d)
%Version：1.0——灰色图像  Time：2013/05/01
%Version：1.1——灰色/彩色图像  Time：2013/05/02
%考虑到实用性，决定不添加更多的边界处理选择，统一使用：reflect across edge

if size(I,3)==1
    newI = ReflectEdgeGray(I,d);
elseif size(I,3)==3
    newI = ReflectEdgeColor(I,d);
else 
    error('Incorrect image size')    
end
end

function newI = ReflectEdgeGray(I,d)
[m n] = size(I);
newI = zeros(m+2*d,n+2*d);
%中间部分
newI(d+1:d+m,d+1:d+n) = I;
%上
newI(1:d,d+1:d+n) = I(d:-1:1,:);
%下
newI(end-d:end,d+1:d+n) = I(end:-1:end-d,:);
%左
newI(:,1:d) = newI(:,2*d:-1:d+1);
%右
newI(:,m+d+1:m+2*d) = newI(:,m+d:-1:m+1);
end

function newI = ReflectEdgeColor(I,d)
%扩展图像边界
[m n ~] = size(I);
newI = zeros(m+2*d,n+2*d,3);
%中间部分
newI(d+1:d+m,d+1:d+n,1:3) = I;
%上
newI(1:d,d+1:d+n,1:3) = I(d:-1:1,:,1:3);
%下
newI(end-d:end,d+1:d+n,1:3) = I(end:-1:end-d,:,1:3);
%左
newI(:,1:d,1:3) = newI(:,2*d:-1:d+1,1:3);
%右
newI(:,m+d+1:m+2*d,1:3) = newI(:,m+d:-1:m+1,1:3);
end

测试用例：

img = imread('.\lena.tif');
%%img = imread('.\images\lena_gray.tif');
img = double(img)/255;
img = img+0.05*randn(size(img));
img(img<0) = 0; img(img>1) = 1;
%img = imnoise(img,'gaussian');
figure, imshow(img,[])
title('原始图像')
d = 6;
sigma = [3 0.1];
resultI = BilateralFilt2(double(img), d, sigma);

figure, imshow(resultI,[])
title('双边滤波后的图像')

结果：

Reference：

1.C Tomasi, R Manduchi.Bilateral Filtering for Gray and Color Images, - Computer Vision, 1998.