Search Results

Search found 79791 results on 3192 pages for 'design problem'.

Page 15/3192 | < Previous Page | 11 12 13 14 15 16 17 18 19 20 21 22  | Next Page >

  • Open closed prinicple, problem

    - by Marcus
    Hi, I'm trying to apply OCP to a code snippet I have that in it's current state is really smelly, but I feel I'm not getting all the way to the end. Current code: public abstract class SomeObject {} public class SpecificObject1 : SomeObject {} public class SpecificObject2 : SomeObject {} // Smelly code public class Model { public void Store(SomeObject someObject) { if (someObject is SpecificObject1) {} else if (someObject is SpecificObject2) {} } } That is really ugly, my new approach looks like this: // No so smelly code public class Model { public void Store(SomeObject someObject) { throw new Expception("Not allowed!"); } public void Store(SpecificObject1 someObject) {} public void Store(SpecificObject2 someObject) {} } When a new SomeObject type comes along I must implement how that specific object is stored, this will break OCP cause I need to alter the Model-class. To move the store logic to SomeObject also feels wrong cause then I will violate SRP (?), becuase in this case the SomeObject is almost like a DTO, it's resposibility it not how to know to store itself. If a new implementation to SomeObject comes along who's store implementation is missing I will get a runtime error due to exception in Store method in Model class, it also feels like a code smell. This is because calling code will in the form of IEnumerable<SomeObject> sequence; I will not know the specific types of the sequence objects. I can't seem to grasp the OCP-concept. Anyone has any concrete examples or links that is a bit more than just some Car/Fruit example?

    Read the article

  • Improvements to Joshua Bloch's Builder Design Pattern?

    - by Jason Fotinatos
    Back in 2007, I read an article about Joshua Blochs take on the "builder pattern" and how it could be modified to improve the overuse of constructors and setters, especially when an object has a large number of properties, most of which are optional. A brief summary of this design pattern is articled here [http://rwhansen.blogspot.com/2007/07/theres-builder-pattern-that-joshua.html]. I liked the idea, and have been using it since. The problem with it, while it is very clean and nice to use from the client perspective, implementing it can be a pain in the bum! There are so many different places in the object where a single property is reference, and thus creating the object, and adding a new property takes a lot of time. So...I had an idea. First, an example object in Joshua Bloch's style: Josh Bloch Style: public class OptionsJoshBlochStyle { private final String option1; private final int option2; // ...other options here <<<< public String getOption1() { return option1; } public int getOption2() { return option2; } public static class Builder { private String option1; private int option2; // other options here <<<<< public Builder option1(String option1) { this.option1 = option1; return this; } public Builder option2(int option2) { this.option2 = option2; return this; } public OptionsJoshBlochStyle build() { return new OptionsJoshBlochStyle(this); } } private OptionsJoshBlochStyle(Builder builder) { this.option1 = builder.option1; this.option2 = builder.option2; // other options here <<<<<< } public static void main(String[] args) { OptionsJoshBlochStyle optionsVariation1 = new OptionsJoshBlochStyle.Builder().option1("firefox").option2(1).build(); OptionsJoshBlochStyle optionsVariation2 = new OptionsJoshBlochStyle.Builder().option1("chrome").option2(2).build(); } } Now my "improved" version: public class Options { // note that these are not final private String option1; private int option2; // ...other options here public String getOption1() { return option1; } public int getOption2() { return option2; } public static class Builder { private final Options options = new Options(); public Builder option1(String option1) { this.options.option1 = option1; return this; } public Builder option2(int option2) { this.options.option2 = option2; return this; } public Options build() { return options; } } private Options() { } public static void main(String[] args) { Options optionsVariation1 = new Options.Builder().option1("firefox").option2(1).build(); Options optionsVariation2 = new Options.Builder().option1("chrome").option2(2).build(); } } As you can see in my "improved version", there are 2 less places in which we need to add code about any addition properties (or options, in this case)! The only negative that I can see is that the instance variables of the outer class are not able to be final. But, the class is still immutable without this. Is there really any downside to this improvement in maintainability? There has to be a reason which he repeated the properties within the nested class that I'm not seeing?

    Read the article

  • Design pattern for logging changes in parent/child objects saved to database

    - by andrew
    I’ve got a 2 database tables in parent/child relationship as one-many. I’ve got three classes representing the data in these two tables: Parent Class { Public int ID {get; set;} .. other properties } Child Class { Public int ID {get;set;} Public int ParentID {get; set;} .. other properties } TogetherClass { Public Parent Parent; Public List<Child> ChildList; } Lastly I’ve got a client and server application – I’m in control of both ends so can make changes to both programs as I need to. Client makes a request for ParentID and receives a Together Class for the matching parent, and all of the child records. The client app may make changes to the children – add new children, remove or modify existing ones. Client app then sends the Together Class back to the server app. Server app needs to update the parent and child records in the database. In addition I would like to be able to log the changes – I’m doing this by having 2 separate tables one for Parent, one for child; each containing the same columns as the original plus date time modified, by whom and a list of the changes. I’m unsure as to the best approach to detect the changes in records – new records, records to be deleted, records with no fields changed, records with some fields changed. I figure I need to read the parent & children records and compare those to the ones in the Together Class. Strategy A: If Together class’s child record has an ID of say 0, that indicates a new record; insert. Any deleted child records are no longer in the Together Class; see if any of the comparison child records are not found in the Together class and delete if not found (Compare using ID). Check each child record for changes and if changed log. Strategy B: Make a new Updated TogetherClass UpdatedClass { Public Parent Parent {get; set} Public List<Child> ListNewChild {get;set;} Public List<Child> DeletedChild {get;set;} Public List<Child> ExistingChild {get;set;} // used for no changes and modified rows } And then process as per the list. The reason why I’m asking for ideas is that both of these solutions don’t seem optimal to me and I suspect this problem has been solved already – some kind of design pattern ? I am aware of one potential problem in this general approach – that where Client App A requests a record; App B requests same record; A then saves changes; B then saves changes which may overwrite changes A made. This is a separate locking issue which I’ll raise a separate question for if I’ve got trouble implementing. The actual implementation is c#, SQL Server and WCF between client and server - sharing a library containing the class implementations. Apologies if this is a duplicate post – I tried searching various terms without finding a match though.

    Read the article

  • C# Design How to Elegantly wrap a DAL class

    - by guazz
    I have an application which uses MyGeneration's dOODads ORM to generate it's Data Access Layer. dOODad works by generating a persistance class for each table in the database. It works like so: // Load and Save Employees emps = new Employees(); if(emps.LoadByPrimaryKey(42)) { emps.LastName = "Just Got Married"; emps.Save(); } // Add a new record Employees emps = new Employees(); emps.AddNew(); emps.FirstName = "Mr."; emps.LastName = "dOOdad"; emps.Save(); // After save the identity column is already here for me. int i = emps.EmployeeID; // Dynamic Query - All Employees with 'A' in thier last name Employees emps = new Employees(); emps.Where.LastName.Value = "%A%"; emps.Where.LastName.Operator = WhereParameter.Operand.Like; emps.Query.Load(); For the above example(i.e. Employees DAL object) I would like to know what is the best method/technique to abstract some of the implementation details on my classes. I don't believe that an Employee class should have Employees(the DAL) specifics in its methods - or perhaps this is acceptable? Is it possible to implement some form of repository pattern? Bear in mind that this is a high volume, perfomacne critical application. Thanks, j

    Read the article

  • Domain driven design: Manager and service

    - by ryudice
    I'm creating some business logic in the application but I'm not sure how or where to encapsulate it, I've used the repository pattern for data access, I've seen some projects that use DDD that have some classes with the "Service" suffix and the "manager" suffix, what are each of this clases suppose to take care of in DDD?

    Read the article

  • Repeated properties design pattern

    - by Mark
    I have a DownloadManager class that manages multiple DownloadItem objects. Each DownloadItem has events like ProgressChanged and DownloadCompleted. Usually you want to use the same event handler for all download items, so it's a bit annoying to have to set the event handlers over and over again for each DownloadItem. Thus, I need to decide which pattern to use: Use one DownloadItem as a template and clone it as necessary var dm = DownloadManager(); var di = DownloadItem(); di.ProgressChanged += new DownloadProgressChangedEventHandler(di_ProgressChanged); di.DownloadCompleted += new DownloadProgressChangedEventHandler(di_DownloadCompleted); DownloadItem newDi; newDi = di.Clone(); newDi.Uri = "http://google.com"; dm.Enqueue(newDi); newDi = di.Clone(); newDi.Uri = "http://yahoo.com"; dm.Enqueue(newDi); Set the event handlers on the DownloadManager instead and have it copy the events over to each DownloadItem that is enqeued. var dm = DownloadManager(); dm.ProgressChanged += new DownloadProgressChangedEventHandler(di_ProgressChanged); dm.DownloadCompleted += new DownloadProgressChangedEventHandler(di_DownloadCompleted); dm.Enqueue(new DownloadItem("http://google.com")); dm.Enqueue(new DownloadItem("http://yahoo.com")); Or use some kind of factory var dm = DownloadManager(); var dif = DownloadItemFactory(); dif.ProgressChanged += new DownloadProgressChangedEventHandler(di_ProgressChanged); dif.DownloadCompleted += new DownloadProgressChangedEventHandler(di_DownloadCompleted); dm.Enqueue(dif.Create("http://google.com")); dm.Enqueue(dif.Create("http://yahoo.com")); What would you recommend?

    Read the article

  • RESTful Question/Answer design?

    - by Kirschstein
    This is a toy project I'm working on at the moment. My app contains questions with multiple choice answers. The question url is in the following format, with GET & POST mapping to different actions on the questions controller. GET: url.com/questions/:category/:difficulty => 'ask' POST: url.com/questions/:category/:difficulty => 'answer' I'm wondering if it's worth redesigning this into a RESTful style. I know I'd need to introduce answers as a resource, but I'm struggling to think of a url that would look natural for answering that question. Would a redesign be worthwhile? How would you go about structuring the urls?

    Read the article

  • Design Question on when to save

    - by Ben
    Hi, I was just after peoples opinion on when the best time to save an object (or collection of objects) is. I appreciate that it can be completely dependent on the situation that you are in but here is my situation. I have a collection of objects "MyCollection" in a grid. You can open each object "MyObject" in an editor dialogue by double clicking on the grid. Selecting "Cancel" on the dialogue will back out any changes you have made, but should selecting "ok" commit those changes back to the database, or should they commit the changes on that object back to the collection and have a save method that iterates through the collection and saves all changed objects? If i have an object "MyParentObject", that contains a collection of childen "MyChildObjectCollection", none of the changes made to each "MyChildObject" would be commited to the database until the "MyParentObject" was saved - this makes sense. However in my current situation, none of the objects in the collection are linked, therefore should the "Ok" on the dialogue commit the changes to the database? Appreciate any opinions on this. Thanks

    Read the article

  • User account design and security...

    - by espinet
    Before I begin, I am using Ruby on Rails and the Devise gem for user authentication. Hi, I was doing some research about account security and I found a blog post about the topic awhile ago but I can no longer find it again. I read something about when making a login system you should have 1 model for User, this contains a user's username, encrypted password, and email. You should also have a model for a user's Account. This contains everything else. A User has an Account. I don't know if I'm explaining this correctly since I haven't seen the blog post for several months and I lost my bookmark. Could someone explain how and why I should or shouldn't do this. My application deals with money so I need to cover my bases with security. Thanks.

    Read the article

  • Schema design: many to many plus additional one to many

    - by chrisj
    Hi, I have this scenario and I'm not sure exactly how it should be modeled in the database. The objects I'm trying to model are: teams, players, the team-player membership, and a list of fees due for each player on a given team. So, the fees depend on both the team and the player. So, my current approach is the following: **teams** id name **players** id name **team_players** id player_id team_id **team_player_fees** id team_players_id amount send_reminder_on Schema layout ERD In this schema, team_players is the junction table for teams and players. And the table team_player_fees has records that belong to records to the junction table. For example, playerA is on teamA and has the fees of $10 and $20 due in Aug and Feb. PlayerA is also on teamB and has the fees of $25 and $25 due in May and June. Each player/team combination can have a different set of fees. Questions: Are there better ways to handle such a scenario? Is there a term for this type of relationship? (so I can google it) Or know of any references with similar structures?

    Read the article

  • Project design / FS layout for large django projects

    - by rcreswick
    What is the best way to layout a large django project? The tutuorials provide simple instructions for setting up apps, models, and views, but there is less information about how apps and projects should be broken down, how much sharing is allowable/necessary between apps in a typical project (obviously that is largely dependent on the project) and how/where general templates should be kept. Does anyone have examples, suggestions, and explanations as to why a certain project layout is better than another? I am particularly interested in the incorporation of large numbers of unit tests (2-5x the size of the actual code base) and string externalization / templates.

    Read the article

  • General ORM design question

    - by Calvin
    Suppose you have 2 classes, Person and Rabbit. A person can do a number of things to a rabbit, s/he can either feed it, buy it and become its owner, or give it away. A rabbit can have none or at most 1 owner at a time. And if it is not fed for a while, it may die. Class Person { Void Feed(Rabbit r); Void Buy(Rabbit r); Void Giveaway(Person p, Rabbit r); Rabbit[] rabbits; } Class Rabbit { Bool IsAlive(); Person pwner; } There are a couple of observations from the domain model: Person and Rabbit can have references to each other Any actions on 1 object can also change the state of the other object Even if no explicit actions are invoked, there can still be a change of state in the objects (e.g. Rabbit can be starved to death, and that causes it to be removed from the Person.rabbits array) As DDD is concerned, I think the correct approach is to synchronize all calls that may change the states in the domain model. For instance, if a Person buys a Rabbit, s/he would need to acquire a lock in Person to make a change to the rabbits array AND also another lock in Rabbit to change its owner before releasing the first one. This would prevent a race condition where 2 Persons claim to be the owner of the little Rabbit. The other approach is to let the database to handle all these synchronizations. Who makes the first call wins, but then the DB needs to have some kind of business logics to figure out if it is a valid transaction (e.g. if a Rabbit already has an owner, it cannot change its owner unless the Person gives it away). There are both pros/cons in either approach, and I’d expect the “best” solution would be somewhere in-between. How would you do it in real life? What’s your take and experience? Also, is it a valid concern that there can be another race condition the domain model has committed its change but before it is fully committed in the database? And for the 3rd observation (i.e. state change due to time factor). How will you do it?

    Read the article

  • How to better design it ???

    - by Deepak
    public interface IBasePresenter { } public interface IJobViewPresenter : IBasePresenter { } public interface IActivityViewPresenter : IBasePresenter { } public class BaseView { public IBasePresenter Presenter { get; set; } } public class JobView : BaseView { public IJobViewPresenter JobViewPresenter { get { this.Presenter as IJobViewPresenter;} } } public class ActivityView : BaseView { public IActivityViewPresenter ActivityViewPresenter { get { this.Presenter as IActivityViewPresenter;} } } Lets assume that I need a IBasePresenter property on BaseView. Now this property is inherited by JobView and ActivityView but if I need reference to IJobViewPresenter object in these derived classes then I need to type cast IBasePresenter property to IJobViewPresenter or IActivityPresenter (which I want to avoid) or create JobViewPresenter and ActivityViewPresenter on derived classes (as shown above). I want to avoid type casting in derived classes and still have reference to IJobViewPresenter or IActivityViewPresenter and still have IBasePresenter in BaseView. Is there a way I can achieve it ?

    Read the article

  • Design issue with ATG CommercePipelineManager

    - by user1339772
    The definition of runProcess() method in PipelineManager is public PipelineResult runProcess(String pChainId, Object pParam) throws RunProcessException This gives me an impression that ANY object can be passed as the second param. However, ATG OOTB has PipelineManager component referring to CommercePipelineManager class which overrides the runProcess() method and downcast pParam to map and adds siteId to it. Basically, this enforces the client code to send only Map. Thus, if one needs to create a new pipeline chain, has to use map as data structure to pass on the data. Offcourse, one can always get around this by creating a new PipelineManager component, but I was just wondering the thought behind explicitly using map in CommercePipelineManager

    Read the article

  • Exploring the Factory Design Pattern

    - by asksuperuser
    There was an article here: http://msdn.microsoft.com/en-us/library/Ee817667%28pandp.10%29.aspx The first part of tut implemented this pattern with abstract classes. The second part shows an example with Interface class. But nothing in this article discusses why this pattern would rather use abstract or interface. So what explanation (advantages of one over the other) would you give ? Not in general but for this precise pattern.

    Read the article

  • Repository Design Pattern Guidance

    - by thefactor
    Let's say you have an MVVM CRM application. You have a number of customer objects in memory, through a repository. What would be the appropriate place to handle tasks that aren't associated with traditional MVVM tasks from a GUI? For example, let's say every few minutes you want to check to see if their address is valid and pop up a notification if it is not. Or you want to send out an hourly e-mail update. Or you want a window to pop up to remind you to call a customer at a specific time. Where does this logic go? It's not GUI/action-oriented, and it's not logic that would be appropriate for a repository, I think.

    Read the article

  • Domain-Driven-Design question

    - by Michael
    Hello everyone, I have a question about DDD. I'm building a application to learn DDD and I have a question about layering. I have an application that works like this: UI layer calls = Application Layer - Domain Layer - Database Here is a small example of how the code looks: //****************UI LAYER************************ //Uses Ioc to get the service from the factory. //This factory would be in the MyApp.Infrastructure.dll IImplementationFactory factory = new ImplementationFactory(); //Interface and implementation for Shopping Cart service would be in MyApp.ApplicationLayer.dll IShoppingCartService service = factory.GetImplementationFactory<IShoppingCartService>(); //This is the UI layer, //Calling into Application Layer //to get the shopping cart for a user. //Interface for IShoppingCart would be in MyApp.ApplicationLayer.dll //and implementation for IShoppingCart would be in MyApp.Model. IShoppingCart shoppingCart = service.GetShoppingCartByUserName(userName); //Show shopping cart information. //For example, items bought, price, taxes..etc ... //Pressed Purchase button, so even for when //button is pressed. //Uses Ioc to get the service from the factory again. IImplementationFactory factory = new ImplementationFactory(); IShoppingCartService service = factory.GetImplementationFactory<IShoppingCartService>(); service.Purchase(shoppingCart); //**********************Application Layer********************** public class ShoppingCartService : IShoppingCartService { public IShoppingCart GetShoppingCartByUserName(string userName) { //Uses Ioc to get the service from the factory. //This factory would be in the MyApp.Infrastructure.dll IImplementationFactory factory = new ImplementationFactory(); //Interface for repository would be in MyApp.Infrastructure.dll //but implementation would by in MyApp.Model.dll IShoppingCartRepository repository = factory.GetImplementationFactory<IShoppingCartRepository>(); IShoppingCart shoppingCart = repository.GetShoppingCartByUserName(username); //Do shopping cart logic like calculating taxes and stuff //I would put these in services but not sure? ... return shoppingCart; } public void Purchase(IShoppingCart shoppingCart) { //Do Purchase logic and calling out to repository ... } } I've seem to put most of my business rules in services rather than the models and I'm not sure if this is correct? Also, i'm not completely sure if I have the laying correct? Do I have the right pieces in the correct place? Also should my models leave my domain model? In general I'm I doing this correct according DDD? Thanks!

    Read the article

  • Is there a design pattern for this ?

    - by ytrewq
    I have a component that needs to call a specific service depending on the input it receives. So my component has to look at the input and based on a configuration that says "for this input call this service with this data" needs to call the proper service. The services have a common signature method and a specific one (each). I thought about an abstract class that includes the signatures for all three methods. The implementation for the two services will override all three methods (throwing NotImplementedException for the methods that are not supported by current service). A component that could be initialized with a map (that for each input type will have the type of the service to be called) will also be defined. Do you have a better approach to cope this scenario ?

    Read the article

  • Object model design choice

    - by spinon
    I am currently working on a ASP.NET MVC reporting application using C#. This is a redesign from a PHP application that was just initially thrown together and is now starting to gain some more traction. SowWe are in the process of reworking the backend to have a more OO approach. One of the descisions I am currently wrestling with is how to structure the domain objects. Since 95% of the site is readonly I am not sure if the typical approaches are practical. Should I create domain objects for the primary pieces of the application (ticket, assignment, assignee) and then create static methods off of these areas to pull the reporting data? Or should I just skip that part and create the chart data classes and have some get method off of these classes? It's not a real big application and currenlty I am the only one developing on it. But I feel torn as to which approach. I feel that the first one is the better choice but maybe overkill given that the majority of uses is for aggregate reporting. Anybody have some good insight on why I should go one way or another?

    Read the article

  • Design For Asynchronous User Interface

    - by Sohnee
    I have been working on a integration that has posed an interesting user interface conundrum that I would like suggestions for. The user interface is displayed within a third party product. The state of the interface is supplied by calls to a service I have written. There can be small delays between the actual state changing the the user interface changing due to the polling for state by the third party. When a user interacts with the user interface, requests are sent back to my application. This then affects the state and the next state poll request will update the user interface. The problem is that the delay between pressing a button and seeing the user interface update is perhaps 1 or 2 seconds and in usability testing I can see that people are clicking again before the user interface updates, thinking that they haven't properly clicked the first time. Given the constraints (we can only update the user interface via the polling mechanism - if we updated it when they clicked, the polling might return and overwrite the change causing unpredictable / undesirable results)... what can we do to make the user experience better. My current idea is to show a message for a couple of seconds so people know their click was accepted, the message would not be affected by the state polling, so wouldn't be prematurely removed / overwritten etc. I'm sure there are other ideas out there and I'm also confident someone has a better idea that I have!

    Read the article

  • Design of input files reading when it comes to defaults/transformations

    - by Stefano Borini
    Suppose you have an application that reads an input file, on a language that does not support the concept of None. The input is read, parsed, and the contents are stored on a structure for later use. Now, in general you want to keep into account transformation of the data from the input, such as adding default values when not specified, or adding full path information to relative path specified in the input. There are two different strategies to achieve this. The first strategy is to perform these transformations at input file reading time. In practice, you put all the intelligence into the input parser, and your application has no logic to deal with unexpected circumstances, such as an unspecified value. You lose the information of what was specified and what wasn't, but you gain in black-boxing the details. Your "running code" needs that information in any case and in a proper form, and is not concerned if it's the default or a user-specified information. The second strategy is to have the file reader a real one-to-one mapper from the file to a memory-stored object, with no intelligent behavior. unspecified values are not filled (which may however be a problem in languages not supporting None) and data is stored verbatim from the file. The intelligence for recovery must now go into the "running code", which must check what was specified in the file, eventually fall back to a default, or modify the input properly before using it. I would like to know your opinion on these two approaches, and in particular which one you found the most frequently implemented.

    Read the article

  • Need advice on framework design: how to make extending easy

    - by beginner_
    I'm creating a framework/library for a rather specific use-case (data type). It uses diverse spring components, including spring-data. The library has a set of entity classes properly set up and according service and dao layers. The main work or main benefit of the framework lies in the dao and service layer. Developers using the framework should be able to extend my entity classes to add additional fields they require. Therefore I made dao and service layer generic so it can be used by such extended entity classes. I now face an issue in the IO part of the framework. It must be able to import the according "special data type" into the database. In this part I need to create a new entity instance and hence need the actual class used. My current solution is to configure in spring a bean of the actual class used. The problem with this is that an application using the framework could only use 1 implementation of the entity (the original one from me or exactly 1 subclass but not 2 different classes of the same hierarchy. I'm looking for suggestions / desgins for solving this issue. Any ideas?

    Read the article

  • How to program for constraints/rules

    - by Gaurav
    First the background, during interviews in the past, many times I have been asked to design some or other variation of card game as programming puzzle, and I have tried to design it in OO way, but I have never been satisfied with my solutions. However it was not until recently that I realized that I had been approaching the problem from the wrong direction. Specifically I was trying to solve the problem by modeling individual card as an object. Problem with this is individual cards don't have any non-trivial intrinsic behavior and therefore are not suitable (or primary) candidate as objects. What is interesting and important about cards are rules and constraints, such as there could be only four suits, or only thirteen cards in each suit. Of course, then there are any number of rules for games. So my questions are Are there any idioms/constructs/patterns to program for rules & constraints. How many in 1 can be applied in conjunction with OO paradigm.

    Read the article

  • Advice on designing a robust program to handle a large library of meta-information & programs

    - by Sam Bryant
    So this might be overly vague, but here it is anyway I'm not really looking for a specific answer, but rather general design principles or direction towards resources that deal with problems like this. It's one of my first large-scale applications, and I would like to do it right. Brief Explanation My basic problem is that I have to write an application that handles a large library of meta-data, can easily modify the meta-data on-the-fly, is robust with respect to crashing, and is very efficient. (Sorta like the design parameters of iTunes, although sometimes iTunes performs more poorly than I would like). If you don't want to read the details, you can skip the rest Long Explanation Specifically I am writing a program that creates a library of image files and meta-data about these files. There is a list of tags that may or may not apply to each image. The program needs to be able to add new images, new tags, assign tags to images, and detect duplicate images, all while operating. The program contains an image Viewer which has tagging operations. The idea is that if a given image A is viewed while the library has tags T1, T2, and T3, then that image will have boolean flags for each of those tags (depending on whether the user tagged that image while it was open in the Viewer). However, prior to being viewed in the Viewer, image A would have no value for tags T1, T2, and T3. Instead it would have a "dirty" flag indicating that it is unknown whether or not A has these tags or not. The program can introduce new tags at any time (which would automatically set all images to "dirty" with respect to this new tag) This program must be fast. It must be easily able to pull up a list of images with or without a certain tag as well as images which are "dirty" with respect to a tag. It has to be crash-safe, in that if it suddenly crashes, all of the tagging information done in that session is not lost (though perhaps it's okay to loose some of it) Finally, it has to work with a lot of images (10,000) I am a fairly experienced programmer, but I have never tried to write a program with such demanding needs and I have never worked with databases. With respect to the meta-data storage, there seem to be a few design choices: Choice 1: Invidual meta-data vs centralized meta-data Individual Meta-Data: have a separate meta-data file for each image. This way, as soon as you change the meta-data for an image, it can be written to the hard disk, without having to rewrite the information for all of the other images. Centralized Meta-Data: Have a single file to hold the meta-data for every file. This would probably require meta-data writes in intervals as opposed to after every change. The benefit here is that you could keep a centralized list of all images with a given tag, ect, making the task of pulling up all images with a given tag very efficient

    Read the article

  • Why use link classes in oql instead of classes that contain links

    - by Isaac
    itop abstracts its very complex database design with an object query language (oql). For this there are classes definded, like 'Ticket' and 'Server'. Now a Ticket usually is linked to a Server. In my naive way I would give the Ticket class an attribute 'affected_server_list', where I could reference the affected servers. itop does it different: neither Servers nor Tickets know of each other. Instead there is a class 'linkTicketToServer', which provides the link between the two. The first thing I noticed is that it makes oql queries more complex. So I wondered why they designed it this way. One thing that occured to me is that it allows for more flexiblity, in that I can add links without modifying the original classes. Is this allready why one would implement it this way, or are there other reasons for this kind of design?

    Read the article

< Previous Page | 11 12 13 14 15 16 17 18 19 20 21 22  | Next Page >