Interfaces...slightly confused =S

Hey all
Basically Iv got a board game that mostly uses just ordinary classes but now im trying to introduce interfaces so that testing can be done easily...However I'm not sure how to go about naming these or even how to set them up properly. For example:
Currently I have a CardStack class and this is used for objects whenever there is a need for a representation of a bunch of card....if I was to introduce an interface one options would be like so:
- Have a CardStack interface
- Have a AvailableCards implement CardStack
- Have a DiscardedCards implement CardStack
- Have a PlayerCards implement CardStack
But now those three implementation use a lot of duplicated code since they have almost the same methods (pop, push, add, remove, get, etc.) so it seems pointless to have this CardStack interface =S might have as well used inheritance instead of an interface no? Or even just stick to how I had it before...I guess another way would be to have a Stack interface and a CardStack implement this Stack interface? But now what will I have as the return types/parameter types that the interface's methods use? They can use
Card for example but then if I want to implement a PigStack rather than a CardStack I'm doomed.
So yeh im not too sure on how to go about incorporating interfaces like this...any help?
Also sorry for the long post!

kman91 wrote:
hmm...but I don't see how having getters will make it any better? I mean now your just passing the reference to the private field, essentially giving it full access (just cant overwrite it) now.First of all, you need not provide a setter. For example, it's easily possible that only the Board itself needs to set the variable. In that case you can provide a getter but no setter.
Also, you can do stuff other than just returning the variable in the getter: you can log access, count access, deny access (by throwing an exception) or returning a wrapper object.
The thing is there will be alot of methods (and apparently thats a design smell)"Design smell" does not necessarily mean that it's a bad thing, it just means that it's something to investigate (as you're doing it right now!)
since each card has an action that can be activated that can do a whole variety of things.Really? If each card has an action, then i'd expect a single invokeAction() method on the Card class with multiple different implementatons. How do you have multiple methods on a single class because of this?

Similar Messages

Interface/Impl confusion

I was just restructuring some of my code and I decided to extract interfaces out of some concrete
classes. I'm not really expecting to have multiple impls for them, but I did it to reinforce thinking in interfaces and to have a good view over the class without having to look at implemntation details.
Here's a Channel-interface I extracted (an abstraction to IRC-channel):
public interface Channel {
     public void addChannelListener(ChannelListener listener);
     public void removeChannelListener(ChannelListener listener);
     public void send(String message);
     public void rejoin();
     public void leave();
     public void say(String text);
     public void ban(String nickOrAddress);
     public void unban(String nickOrAddress);
     public void kick(String nick);
     public void kick(String nick, String reason);
     public void invite(String nick);
     public void setInviteOnly(boolean enabled);
     public void setTopic(String topic);
     public Server getServer();
     public String getName();
     public String getKey();
     public String getTopic();
     public int getUserCount();
     public User[] getUsers();
     public boolean isConnected();
     public boolean containsNick(String nick);
}Note that this is still an incomplete implementation and there'll be some more of these methods.
Now, earlier I had a concrete Channel-class and it also had methods like this:
public final void topicChanged(Message msg) {...}
public final void messageReceived(Message msg) {...}
public final void modeChanged(Message msg) {...}
public final void nickChanged(Message msg) {...}
public final void userJoined(Message msg) {...}I decided to leave them out of the public interface as they don't really belong there,
since they are called only by the Server-class. The Server-class contains a list of Channel-objects.
It is also the only class that creates new Channel-objects.
So, only the Server-class calls the above methods of it's Channel-objects, when the server receives new messages (and determines they're channel-messages).
I also extracted an interface for the Server, and the ServerImpl now casts the Channel-interfaces to ChannelImpls in some places, to be able to call the above leftover methods.
Is this a bad idea? Does this break the 'Liskov substitution principle'? Do you have any better ideas? Maybe the Channel-class should do 'more work' itself?
I personally don't feel it's that bad but i still get kind of dirty feeling of doing it this way.

So th eChannel interface is the public interface for all others present
in the system but your server needs a bit more functionality. Why not
define another interface:interface ServerChannel extends Channel {
   // server specific methods here
}and your ChannelImpl class:public class ChannelImpl implements ServerChannel { ... }Now the server treats these ChannelImpl objects as ServerChannels,
while the rest of the world simply treats these objects as Channels.
btw, you can build a Factory that builds the appropriate objects for the
appropriate callers, e.g. a package scoped method for the server
(which needs to be in the same package) and a public method for the
rest of the world. The first object would return ServerChannels while
the second method (the public one) would return Channels:[code
public class ChannelFactory {
ServerChannel makeServerChannel() { ... }
public Channel makeChannel() { return makeServerChannel(); }
kind regards,
Jos

Slightly confused about notify method

Hi,
I'm currently studying for the SCJP exam and have 2 questions about the thread notify call.
(1) Is it possible when a thread owns an object lock, i.e. say in synchronized code, that it can call the notify method on a particular thread. For example, say there are 2 threads, names thread1 and thread2, and thead1 obtains object lock, can it call Thread2.notify()? The reason I'm read conflicting web-pages, saying it isn't possible for 1 thread to notify another thread and it is up to Object to decide which thread to invoke, say must use notifyAll() or notify() call. Is this true?
(2) Which thread in following code is Thread.sleep() referring to? Is it the main thread that created 2 threads?
class running extends Thread
     public void run(){
public class fredFlintStone{
     public static void main(String[] args) {
          running thread1 = new running();
          running thread2 = new running();
          thread1.start();
          thread2.start();
          try
          //confused here
        Thread.sleep(12L);
          catch(InterruptedException e)
}Cheers!

its best not to confuse terminology here.
marktheman wrote:
I ... have 2 questions about the thread notify call.Thread is an object which has a notify() method. However you should never use it. You should only use notify() on regular objects.
(1) Is it possible when a thread owns an object lock, A Thread holds a lock, see [http://java.sun.com/javase/6/docs/api/java/lang/Thread.html#holdsLock(java.lang.Object)]
i.e. say in synchronized code, that it can call the notify method on a particular thread. A thread calls notify() on an object not a given thread. (Unless that object is a thread, but don't do that!)
For example, say there are 2 threads, names thread1 and thread2, and thead1 obtains object lock, can it call Thread2.notify()?It can, but it would fail as you can only call notify() on an object you hold the lock to.
The reason I'm read conflicting web-pages, saying it isn't possible for 1 thread to notify another thread It is possible but not in any useful way.
and it is up to Object to decide which thread to invoke,The Object has no say in the matter, its up to the scheduler in the OS.
say must use notifyAll() or notify() call. Is this true?no.
(2) Which thread in following code is Thread.sleep() referring to? The javadoc says "Causes the currently executing thread to sleep". You should always check the documentation for asking that sort of question.

Hi I am slightly confused with using CSS to control links.

I have set the Hover,active and visited colour properties.
the issue is that once a link has been visited the hover colour no longer shows up.
So I tried removing the visited colour and now even when i hover it is staying the default purpleish.
please advise on how to keep the links one colour except when hovering

To get best results in all browsers, a:active MUST follow a:hover in the CSS definition order. A simple memory device for link order is " LiVHA" for link, visited, hover, active.
CSS:
a:link {color: #FF0000}     /* unvisited link */
a:visited {color: #00FF00} /* visited link */
a:hover {color: #FF00FF}    /* mouse over link */
a:active {color: #0000FF}   /* selected link */
Nancy O.
Alt-Web Design & Publishing
Web | Graphics | Print | Media Specialists
www.alt-web.com/
www.twitter.com/altweb
www.alt-web.blogspot.com

Importing photos to Aperture - slight confusion

Dear all,
I am fairly new to Aperture and I have a question to ask. When I used iPhoto to store and manage all of my photos I used to create folders and insert albums inside. For instance I had three major folders assigned to three different years e.g. 2009,2010 and 2011, inside of which I had many different albums. Now, in iPhoto when I wanted to add a photo to already existing album I simply drag and droped it there. As far as I can tell, it is the same with the Aperture, however, for some reason everytime I add a new photo to already existing album in Aperture it automatically created a new Unnamed Project at the top of the folder in which the album was found. Attached is a printscreen describing the question I want to find an answer for.
I am sory for the poor explanation, but I hope my question makes sense.
Thank you very much in advance.

What are your settings in the Import Settings panel - in the File Types brick and Raw and Jpeg pairs?
Did you exclude any file types?
Regards
Léonie

Aggregator preload slightly confused

I am having a problem trying to figure out the preloader,when
you use the aggregator in Captivate 4. It seems like there are two
preloaders,one for the aggregator and one for the swf file.
If I do not place a preloader on in the project preference
for the swf film when published,a default preloader is shown, when
published in the aggregator. There is no loading percent action
given ,just an image of the default preloader. If I place a
preloader on the swf file when published,when I load it and publish
it in the aggregator,two preloaders are shown,the video one works
showing a percentage up to 30% then shuts off and the still image
of the aggregators preloader is shown.(Even though I placed 100%
for the value of the preload for the swf film) Is this normal? Does
the preloader for the aggregator supposed to show no percentage,
and just an image of the preloader. I am using AS2.

Same here on the preloader issues. You're right about the
preloader having a different appearance in aggregator content
playback.
Also, I'm finding that the SWFs are not preloading at all
before they start playing.
I'm using AS3.

Remote interface distribution

Hi,
I am just learning to use RMI, have a typical setup, Server and Client and everything works fine, BUT.......
I have compiled my remote classes, ie RmtServer, RmtServerImpl, then run the rmiregistry
and then started the service, so far so good.
I then compile etc the Client class and run
My question is this, I am currently working from the same directory for all classes, I would like to truly
distribute the client and server. Do I need to put a copy of the RmtServer interface on the client?
Currently in my client I have the line:
RmtServer server = (RmtServer)Naming.lookup("rmi://" + serverHost + "/ProjectServer");
I have been reading about dynamic class loading but I am slightly confused!!!
It says that the stubs created by running "rmic" on the server classes can be loaded dynamically, thats great, but the stubs created and the RmtServer interface are not the same thing. If I have to have a "local" copy of every interface I want to use then it seems a bit limited.
My impression was that I could specify a location for a class repositry on a remote machine and then
instantiate classes from the server on the local machine! So then I would only need to have one interface on the client to enable me to connect to the server for the first time!!!
Im really confused as people can probably tell. Any help would be really appreciated

The purpose of dynamic loading is really so that you don't have to distribute the stub, or the implementation classes of any interfaces.
You can certainly load the remote interface dynamically from the codebase in the client, e.g. with URLClassLoader, but you will also have to have already loaded all the classes that use the remote interface the same way, rather than via the system class loader. Otherwise they won't load (NoClassDefFoundError). You can download the entire client actually, and this is not a bad way to go: see the the RMI 1.1 Specification, 'Bootstrapping the client', for details (if you can find it: try http://java.sun.com/products/archive/jdk/1.1/index.html), and the RMI-USERS archives of about five-six years ago for discussions (see http://archives.java.sun.com/archives/rmi-users.html). The description of this was removed from the 1.2 specification for some reason, and I did encounter problems migrating a 1.1 bootstrap client to 1.2, so beware.

OTV Internal Interface Configuration

Hi
I am trying to implement OTV between 2 sites. I am slightly confused about the config for the join interface and site VLAN. Attached is the basic setup on one of the sites, I am using ASR 1002X routers to perform the OTV functions.
There seem to be limited sources for the ASR OTV configs, but on source states the internal interface on the ASR router should be configured as follows for each vlan
no ip address
service instance 10 ethernet
  encapsulation dot1q 10
  bridge-domain 10
service instance 20 ethernet
  encapsulation dot1q 20
  bridge-domain 20
service instance 30 ethernet
  encapsulation dot1q 30
  bridge-domain 30
I guess on the internal switch it will just be a trunk port allowing the above VLANs?
Thanks

Thanks Minh,
So it is possible to have switchports configured as routed, fabricpath and trunk/access in a fabricpath configuration? Do i need to add any spanning-tree pseudo or priority configuration?
Sample configs:
#ASR
interface GigabitEthernet0/0/1
no ip address
service instance 1 ethernet
  encapsulation dot1q 1
  bridge-domain 1
service instance 2 ethernet
  encapsulation dot1q 2
  bridge-domain 2
service instance 3 ethernet
  encapsulation dot1q 3
  bridge-domain 3
#Nexus 56xx
interface e1/5
switchport mode trunk
switchport trunk allow vlan 1,2,3

Internet via interface 1, LAN via interface 2? How can this be done?

Hi,
I have a Mac Pro with 2 Ethernet interfaces, I will call them E01 & E02 from here on.
I have recently gotten my own ADSL line which runs over my own phone number so I have the ADSL Modem come into the E01.
I am however still required to be in the 'old' network that is setup in my house, so I figured I could just have E02 plug into our Airport Extreme to be able to local LAN connect that way.
Seeing as both E02 as well as E01 have internet access it seems that there is no way to tell my Mac to use E01 exclusively for internet access while using E02 only for LAN abilities but no internet activities.
I am hoping that someone here might know how I can configure this as it is quite annoying constantly having to disable E01 in the network settings, enabling E02, share a file over LAN then revert back to internet access on E01 by disabling E02.
I hope this makes sense, if not please ask me for more info.
Thanks for reading,
Jannis

Thanks for your help, however I am still slightly confused as to how this is supposed to work.
Just so we're on the same page of what I am trying to achieve I made a little sketch:
I have 2 internet lines that need to be used separately however I also need the File Sharing ability between my Mac Pro (bottom left in sketch) and the computers on the other side of the network (bottom right in sketch).
My computer will be using the internet from Modem #1, whereas all other computer should be using the internet provided by Modem #2.
I don't mean to dismiss your comments, it's just that I am still somewhat unclear on how I can have 2 internet lines running together into 1 network and then mapping 1 line to 1 computer and the other line to all other computer on the network.
Thanks again, hopefully the image will explain most of this better than my words.
Regards,
Jannis

WLAN Controller DHCP confusion.

Hello all.
I am slightly confused as the DHCP configuration on the WLAN controllers. I have created WLAN SSID 'test' assigned to dynamic interface 'test' using vlan 410. I want my WLAN clients to use DHCP to get their IP addresses when associated/authenticated to SSID 'test'.
My confusion is where to do this. I see that I can do this on BOTH the SSID (WLANs>Edit>Advanced) AND the interface (CONTROLLER>Interfaces). Why is there two places to do this and what is the difference between the two?
I am using software version 4.1.171.0.
Many thanks for any help
Darren

Ankur,
I'm setting up a school system with a WLC located at the central school board site and 1142APs setup at each remote school campus. Each of the eight remote school campus has it's own DHCP server for local 10.X.X.X IP addresses. As a result of using Class A 10.1.0.0/16, 10.2.0.0/16, 10.3.0.0/16 etc network addresses at each remote school campus there should be more than enough addresses available for both remote campus school hardware and student clients.
The issue I have is everything in currently configured for a single VLAN, VLAN 1. I would like to configure a separate WLAN for teachers/staff, WLAN for students, and WLAN for TechSupport and put each of those WLANs in it's own VLAN for security and performance reasons. Since each remote campus has a single DHCP server should I create separate scope's in the existing MS DHCP server, use a combination of internal and external DHCP servers (ie internal for wireless clients, external for everything else), or standup an additional MS server at each remote site?
Also, ifor each WLAN interface created and associated interface into that WLAN there will need to be a separate VLAN created on the L3 device so it can direct traffic to the centrally located WLC, right?
So many DHCP servers, so little time.... Any direction you can send my way will be greatly appreciated.
Warm Regards,
Michael

AP Supplier Open Interface

I have a question about the Suppliers Open Interface.
I am doing an R12 Conversion and Interface
I designed the Conversion fine...but the interface logic confuses me a bit. I am aware most of the code will be the same.
Maybe someone with working knowledge of the open interface can advise.
How do I do updates? For Sites and Contacts...do I query for the Vendor Number (segment 1), populate in the interface tables, and the concurrent program knows what to update? Or do I populate the Vendor ID?
If someone can provide a detailed description on how this works, or provide a place to read up on it, that would be fantastic!

the same situation i am also facing everything almost same .... but just i got diffirent error in request log that is
==========================
==========================
P_WHAT_TO_IMPORT='ALL'
P_COMMIT_SIZE='1000'
P_PRINT_EXCEPTIONS='Y'
P_DEBUG_SWITCH='Y'
P_TRACE_SWITCH='Y'
Current NLS_LANG and NLS_NUMERIC_CHARACTERS Environment Variables are :
American_America.US7ASCII
LOG :
Report: d:\oracle\prodappl\ap\11.5.0\reports\US\APXSSIMP.rdf
Logged onto server:
Username:
LOG :
Logged onto server:
Username: APPS
MSG MSG-00001: After SRWINIT
MSG MSG-00002: After Get_Company_Name
MSG MSG-00003: After Get_NLS_Strings
ERR REP-1419: 'beforereport': PL/SQL program aborted.
is anyone got solution and can anybody get me out of this situation
thanks in advance
regards
anwer

Confused! Airport or Airport Extreme for M9145

Hi there folks.
I'm slightly confused over this one.
I've read on a number of forums and resellers that my PowerMac G4 MDD 1.25, or more specifically the M9145 is able to take an Airport Extreme card - but then in the next breadth I read that it takes the standard Airport card!?
Hence me being confused!
Could anybody out there please confirm to me as to which one it actually requires? Also, any chance of a part number?
Many thanks in advance.
Regards, Mark.

The M9145 comes up as the dual 1.25GHz MDD from 2003. That is, it takes the older, original AirPort card.
Now the painful part, for the wireless speed the original AirPort card offers, you may as well put a chocolate biscuit in the AirPort card slot. In todays wireless standards, it's really painfully slow.
You could look at a PCI wireless option, or an Ethernet attached WDS router option. Both would come in at about 50x faster, and probably cost less too. Or live with an Ethernet connection to your router.
If you're only going to use the MDD for web browsing, rather than connecting to other local computers, and you're not on superfast broadband, then the original AirPort card may be OK.

Mutations to update entity store by several versions - confused

I am slightly confused about how mutations work. We have been working in our development environments making changes to a particular entity as follows:
Version 1 - deleted a field. It was not marked as a database key. Added a deleter like this:
new Deleter(XXXXXXX.class.getName(), 0, "xxxxx")
Version 2 - added a field. It is not marked as a database key, hence incremented version number, but no mutation code.
Version 3 - deleted a filed. It is not marked as a database key. Added a deleter like this:
new Deleter(XXXXXXX.class.getName(), 2, "yyyyy")
This worked fine for multiple deployments in local development environments, but when we tried to deploy to a test environment, we got this error message when trying to initialise the database:
Caused by: com.sleepycat.persist.evolve.IncompatibleClassException: (JE 4.0.103) Mutation is missing to evolve class: com.chello.booking.model.XXXXXXX version: 0 to class: com.chello.booking.model.XXXXXXX version: 3 Error: Field is not present or not persistent: yyyyy
Mutation is missing to evolve class: com.chello.booking.model.XXXXXXX version: 1 to class: com.chello.booking.model.XXXXXXX version: 3 Error: Field is not present or not persistent: yyyyy
(Note that when upgrading an application in a replicated environment, this exception may indicate that the Master was mistakenly upgraded before this Replica could be upgraded, and the solution is to upgrade this Replica.)
     at com.sleepycat.persist.impl.PersistCatalog.init(PersistCatalog.java:440)
     at com.sleepycat.persist.impl.PersistCatalog.<init>(PersistCatalog.java:221)
     at com.sleepycat.persist.impl.Store.<init>(Store.java:186)
This is not a replicated environment. I had a read through the documentation and don't see anything that prevents us from working like this. Can anyone explain what is going on please?
Incidentally, in case it matters, this is how the mutations are handled during entity store initialisation:
storeConfig.setMutations(mutations);
storeConfig.setAllowCreate(true);
storeConfig.setTransactional(true);
EntityModel model = new AnnotationModel();
model.registerClass(XXXXXXX.class);
storeConfig.setModel(model);
entityStore = new EntityStore(......
Kind regards
James

Hi James,
Mutations must be configured to transform all old versions to the current version. So you also need:
new Deleter(XXXXXXX.class.getName(), 0, "yyyyy")
new Deleter(XXXXXXX.class.getName(), 1, "yyyyy")
I figured this out from the version numbers in the error messages. Note that there are two error messages listed in the exception. The doc explains this in a little more detail, although in very abstract way:
http://www.oracle.com/technology/documentation/berkeley-db/je/java/com/sleepycat/persist/evolve/package-summary.html
Mutations are therefore responsible for converting each existing incompatible class version to the current version as defined by a current class definition. For example, consider that class-version A-1 is initially changed to A-2 and a mutation is added for converting A-1 to A-2. If later changes in version A-3 occur before converting all A-1 instances to version A-2, the converter for A-1 will have to be changed. Instead of converting from A-1 to A-2 it will need to convert from A-1 to A-3. In addition, a mutation converting A-2 to A-3 will be needed.When you say this:
This worked fine for multiple deployments in local development environments, but when we tried to deploy to a test environment, we got this error message when trying to initialise the database:I worry that you're evolving your environment separately on different machines. This won't work. I'm not sure exactly what happened, but if you add mutations and create multiple new versions in a development environment, that sequence of steps won't work for the deployed environment that hasn't gone through the same sequence of steps. Perhaps something like this is what happened.
--mark

JComboBox event confusion: not enough generated? :-(

Hi,
I am slightly confused about the events generated by the combobox.
I have a combobox and and actionlistener registered on it. If I programmatically use setSelectedIndex on the combobox, and event is generated and the actionlistener is triggered. If I do a setSelectedIndex on the same combobox within the code of the action listener, no event is generated, though the index does get updated. Though I am happy it doesn't generate that event, I don't understand why it doesn't. Anyone?
Rene'

Instead of using ActionListener you should use ItemListener for combo box. This is an example
comboBox.addItemListener(new java.awt.event.ItemListener() {
     public void itemStateChanged(ItemEvent e) {
     try {
     JComboBox box = (JComboBox) e.getSource();
     System.out.println(box.getSelectedItem());
     } catch (ArrayIndexOutOfBoundsException exp) {}
});Note: Item Listener only listens when you change something in Combo Box. If you click on a combo box, and selects the same value, item listeners doesn't call.
Hope this helps.

Widget confusion

These widgets seem to be very useful, but I'm slightly confused about what they are, exactly and what standard they adhere to, if any.
Do they conform to a generic widget standard that can be used with any widget framework in a platform-independent way? Or are iWeb's Widgets specific to iWeb?
Also, if I wanted to make my own iWeb widget, where should I look?
Thanks!

what standard they adhere to, if any....
Do they conform to a generic widget standard that can be used with any widget framework in a platform-independent way?
none and no.
Or are iWeb's Widgets specific to iWeb?
yes, they are very specific to iweb... iweb widgets are special kind of folder/bundle, they have instruction/directive for iweb to recognize them (while in iweb) and add code to the page when publish.
Also, if I wanted to make my own iWeb widget, where should I look?
the first place you should look is 'HTML Snippet' widget.
as with iweb themes and templates, every iweb widget is controlled by an XML file, XML schema is the way to decode XML, and only iweb sw engineers know this schema.
as of now, there is no instructions, tutorials, tips and tricks or tools to build iweb widgets.
but you can use Dashcode (OSX Dashboard widget tool) to create iweb widgets.
give them feedback and ask for tools to make widgets:
http://www.apple.com/feedback/iweb.html
if you like to make your own iweb3 widgets, you should look into:
XML and inside iweb.app
and to build anything meaningful, you should look into: HTML, javascript, css and DOM.
I'd built many iweb3 widgets with the above method:
http://www.cyclosaurus.com/WidgetAdditions/iWeb3Widgets.html

Interfaces...slightly confused =S

Similar Messages

Maybe you are looking for