Search Results

Search found 49320 results on 1973 pages for 'system architecture'.

Page 86/1973 | < Previous Page | 82 83 84 85 86 87 88 89 90 91 92 93  | Next Page >

  • Futures/Monads vs Events

    - by c69
    So, the question is quite simple: in an application framework, when performance impact can be ignored (10-20 events per second at max), what is more maintainable and flexible to use as a preferred medium for communication between modules - Events or Futures/Promices/Monads ? Its often being said, that Events (pub/sub, mediator) allow loose-coupling and thus - more maintainable app... My experience deny this: once you have more that 20+ events - debugging becomes hard, and so is refactoring - because it is very hard to see: who, when and why uses what. Promices (i'm coding in javascript) are much uglier and dumber, than Events. But: you can clearly see connections between function calls, so application logic becomes more straight-forward. What i'm afraid. though, is that Promices will bring more hard-coupling with them... p.s: the answer does not have to be based on JS, experience from other functional languages is much welcome.

    Read the article

  • Should package structure closely resemble class hierarchy?

    - by Panzercrisis
    Pretty simple question. Should package structure closely resemble class hierarchy? If so, how closely? Why or why not? For instance, let's say you've got class A and class B, plus class AFactory and class BFactory. You put class A and class B in the package com.something.elements, and you put AFactory and BFactory in com.something.elements.factories. AFactory and BFactory would be further down the hierarchy package-wise, but they'd be further up class-wise. Is this sort of thing a good idea or a bad idea?

    Read the article

  • Tester/Doer pattern: Assume the caller conforms to the pattern or be defensive and repeat the check?

    - by Daniel Hilgarth
    Assume a simple class that implements the Tester/Doer pattern: public class FooCommandHandler : ICommandHandler { public bool CanHandle(object command) { return command is FooCommand; } public void Handle(object command) { var fooCommand = (FooCommand)command; // Do something with fooCommand } } Now, if someone doesn't conform to the pattern and calls Handle without verifying the command via CanHandle, the code in Handle throws an exception. However, depending on the actual implementation of Handle this can be a whole range of different exceptions. The following implementation would check CanHandle again in Handle and throw a descriptive exception: public void Handle(object command) { if(!CanHandle(command)) throw new TesterDoerPatternUsageViolationException("Please call CanHandle first"); // actual implementation of handling the command. } This has the advantage that the exception is very descriptive. It has the disadvantage that CanHandle is called twice for "good" clients. Is there a consensus on which variation should be used?

    Read the article

  • Looking for menu-driven coding platforms

    - by user2634047
    Can anyone point me to an application development environment that uses menu-driven coding? This would mean where commands, variable names, etc. are not keyed in, but rather are selected from a menu of context-specific options. For example, the user selects an If...then command from a menu of commands, and is then presented with a menu of variables to choose from for the the 'if' conditions(s) (or creates new variable(s) on the fly via the menu), and is then presented with a menu of applicable functions that are applicable to the selected variable (e.g., val()), and so on until the If...then statement has been fully coded. The idea is that the user never types any portion of the code, but selects all code elements from a menu, or defines them on the fly via the menu. Thanks.

    Read the article

  • Could I be going crazy with Event Handlers? Am I going the "wrong way" with my design?

    - by sensae
    I guess I've decided that I really like event handlers. I may be suffering a bit from analysis paralysis, but I'm concerned about making my design unwieldy or running into some other unforeseen consequence to my design decisions. My game engine currently does basic sprite-based rendering with a panning overhead camera. My design looks a bit like this: SceneHandler Contains a list of classes that implement the SceneListener interface (currently only Sprites). Calls render() once per tick, and sends onCameraUpdate(); messages to SceneListeners. InputHandler Polls the input once per tick, and sends a simple "onKeyPressed" message to InputListeners. I have a Camera InputListener which holds a SceneHandler instance and triggers updateCamera(); events based on what the input is. AgentHandler Calls default actions on any Agents (AI) once per tick, and will check a stack for any new events that are registered, dispatching them to specific Agents as needed. So I have basic sprite objects that can move around a scene and use rudimentary steering behaviors to travel. I've gotten onto collision detection, and this is where I'm not sure the direction my design is going is good. Is it a good practice to have many, small event handlers? I imagine going the way I am that I'd have to implement some kind of CollisionHandler. Would I be better off with a more consolidated EntityHandler which handles AI, collision updates, and other entity interactions in one class? Or will I be fine just implementing many different event handling subsystems which pass messages to each other based on what kind of event it is? Should I write an EntityHandler which is simply responsible for coordinating all these sub event handlers? I realize in some cases, such as my InputHandler and SceneHandler, those are very specific types of events. A large portion of my game code won't care about input, and a large portion won't care about updates that happen purely in the rendering of the scene. Thus I feel my isolation of those systems is justified. However, I'm asking this question specifically approaching game logic type events.

    Read the article

  • Stop myself from over-complicating applications

    - by stuartmclark
    Recently I worked on a fairly large project involving C# and MVVM. This application had around 160 projects in the solutions each seprarated into their own layers. As I have been working on this application for almost a year, building it from scratch as part of a team, I am now coming off that project and onto smaller more trivial projects. As I was beginning to develop a small in-house tool I found myself trying to mimic the larger applications structure and layering but in the end I just had a simple application with several DLLs which I know I wouldn't have done if I had not worked on that larger application before. I am just wondering if there are any techniques I can utilise to stop myself from turning a "code-behind" style trivial application into a full blown MVVM application? Or should I continue developing as I am and try to keep the unnecessary fluff out of the project?

    Read the article

  • Restructuring a large Chrome Extension/WebApp

    - by A.M.K
    I have a very complex Chrome Extension that has gotten too large to maintain in its current format. I'd like to restructure it, but I'm 15 and this is the first webapp or extension of it's type I've built so I have no idea how to do it. TL;DR: I have a large/complex webapp I'd like to restructure and I don't know how to do it. Should I follow my current restructure plan (below)? Does that sound like a good starting point, or is there a different approach that I'm missing? Should I not do any of the things I listed? While it isn't relevant to the question, the actual code is on Github and the extension is on the webstore. The basic structure is as follows: index.html <html> <head> <link href="css/style.css" rel="stylesheet" /> <!-- This holds the main app styles --> <link href="css/widgets.css" rel="stylesheet" /> <!-- And this one holds widget styles --> </head> <body class="unloaded"> <!-- Low-level base elements are "hardcoded" here, the unloaded class is used for transitions and is removed on load. i.e: --> <div class="tab-container" tabindex="-1"> <!-- Tab nav --> </div> <!-- Templates for all parts of the application and widgets are stored as elements here. I plan on changing these to <script> elements during the restructure since <template>'s need valid HTML. --> <template id="template.toolbar"> <!-- Template content --> </template> <!-- Templates end --> <!-- Plugins --> <script type="text/javascript" src="js/plugins.js"></script> <!-- This contains the code for all widgets, I plan on moving this online and downloading as necessary soon. --> <script type="text/javascript" src="js/widgets.js"></script> <!-- This contains the main application JS. --> <script type="text/javascript" src="js/script.js"></script> </body> </html> widgets.js (initLog || (window.initLog = [])).push([new Date().getTime(), "A log is kept during page load so performance can be analyzed and errors pinpointed"]); // Widgets are stored in an object and extended (with jQuery, but I'll probably switch to underscore if using Backbone) as necessary var Widgets = { 1: { // Widget ID, this is set here so widgets can be retreived by ID id: 1, // Widget ID again, this is used after the widget object is duplicated and detached size: 3, // Default size, medium in this case order: 1, // Order shown in "store" name: "Weather", // Widget name interval: 300000, // Refresh interval nicename: "weather", // HTML and JS safe widget name sizes: ["tiny", "small", "medium"], // Available widget sizes desc: "Short widget description", settings: [ { // Widget setting specifications stored as an array of objects. These are used to dynamically generate widget setting popups. type: "list", nicename: "location", label: "Location(s)", placeholder: "Enter a location and press Enter" } ], config: { // Widget settings as stored in the tabs object (see script.js for storage information) size: "medium", location: ["San Francisco, CA"] }, data: {}, // Cached widget data stored locally, this lets it work offline customFunc: function(cb) {}, // Widgets can optionally define custom functions in any part of their object refresh: function() {}, // This fetches data from the web and caches it locally in data, then calls render. It gets called after the page is loaded for faster loads render: function() {} // This renders the widget only using information from data, it's called on page load. } }; script.js (initLog || (window.initLog = [])).push([new Date().getTime(), "These are also at the end of every file"]); // Plugins, extends and globals go here. i.e. Number.prototype.pad = .... var iChrome = function(refresh) { // The main iChrome init, called with refresh when refreshing to not re-run libs iChrome.Status.log("Starting page generation"); // From now on iChrome.Status.log is defined, it's used in place of the initLog iChrome.CSS(); // Dynamically generate CSS based on settings iChrome.Tabs(); // This takes the tabs stored in the storage (see fetching below) and renders all columns and widgets as necessary iChrome.Status.log("Tabs rendered"); // These will be omitted further along in this excerpt, but they're used everywhere // Checks for justInstalled => show getting started are run here /* The main init runs the bare minimum required to display the page, this sets all non-visible or instantly need things (such as widget dragging) on a timeout */ iChrome.deferredTimeout = setTimeout(function() { iChrome.deferred(refresh); // Pass refresh along, see above }, 200); }; iChrome.deferred = function(refresh) {}; // This calls modules one after the next in the appropriate order to finish rendering the page iChrome.Search = function() {}; // Modules have a base init function and are camel-cased and capitalized iChrome.Search.submit = function(val) {}; // Methods within modules are camel-cased and not capitalized /* Extension storage is async and fetched at the beginning of plugins.js, it's then stored in a variable that iChrome.Storage processes. The fetcher checks to see if processStorage is defined, if it is it gets called, otherwise settings are left in iChromeConfig */ var processStorage = function() { iChrome.Storage(function() { iChrome.Templates(); // Templates are read from their elements and held in a cache iChrome(); // Init is called }); }; if (typeof iChromeConfig == "object") { processStorage(); } Objectives of the restructure Memory usage: Chrome apparently has a memory leak in extensions, they're trying to fix it but memory still keeps on getting increased every time the page is loaded. The app also uses a lot on its own. Code readability: At this point I can't follow what's being called in the code. While rewriting the code I plan on properly commenting everything. Module interdependence: Right now modules call each other a lot, AFAIK that's not good at all since any change you make to one module could affect countless others. Fault tolerance: There's very little fault tolerance or error handling right now. If a widget is causing the rest of the page to stop rendering the user should at least be able to remove it. Speed is currently not an issue and I'd like to keep it that way. How I think I should do it The restructure should be done using Backbone.js and events that call modules (i.e. on storage.loaded = init). Modules should each go in their own file, I'm thinking there should be a set of core files that all modules can rely on and call directly and everything else should be event based. Widget structure should be kept largely the same, but maybe they should also be split into their own files. AFAIK you can't load all templates in a folder, therefore they need to stay inline. Grunt should be used to merge all modules, plugins and widgets into one file. Templates should also all be precompiled. Question: Should I follow my current restructure plan? Does that sound like a good starting point, or is there a different approach that I'm missing? Should I not do any of the things I listed? Do applications written with Backbone tend to be more intensive (memory and speed) than ones written in Vanilla JS? Also, can I expect to improve this with a proper restructure or is my current code about as good as can be expected?

    Read the article

  • Game State / Screen Management

    - by Ashylnn Mac
    What's the best way to handle game states / screens? My problem is this: PlayGameScreen adds a new InventoryGameScreen to the game during it's update. This immediately adds InventoryGameScreen to the array of GameScreens. That's throwing an exception when iterating over the array that the contents of the array have changed. Should I have two more arrays, like screensToBeAdded and screensToBeRemoved and do all the processing for them at the end of the game loop after drawing all the other screens?

    Read the article

  • Java game design question (graphical objects)

    - by vemalsar
    Hello Guys, I'm beginner in game development, in Java and here on this site too and I have a game design question. Please comment my idea: I have a main loop which call update and draw method. I want to use an ArrayList which store graphical objects, they have coordinate and image or text to draw and my game objects extends this class. In update, I can choose which objects should be put in the array and in draw method I'll display the elements of array on the screen. I'm using a buffer and draw first there, but it is not important now I guess...Here is a simple (not full) code, only the logic: public class GamePanel extends JPanel implements KeyListener { ArrayList<graphicalObjects> graphArray = new ArrayList<graphicalObjects>(); public void update() { //change the game scene, update the graphArray, process input etc. } public void draw() { //draws every element of graphArray to a JPanel } public static main(String[] args) { while(true) { update(); draw(); } } } My questions: Should have I use interface or abstract class for graphicalObjects? graphicalObjects class and the ArrayList really needs or there is some better solution? How to draw objects? They draw themself with their own method or in the draw method I have to draw manually based on graphicalObjects variables (x,y coordinates, image etc.)? If this conception is wrong, please suggest another one! All comments are welcome and sorry if this is dumb question, thanks!

    Read the article

  • Pure functional programming and game state

    - by Fu86
    Is there a common technique to handle state (in general) in a functional programming language? There are solutions in every (functional) programming language to handle global state, but I want to avoid this as far as I could. All state in a pure functional manner are function parameters. So I need to put the whole game state (a gigantic hashmap with the world, players, positions, score, assets, enemies, ...)) as a parameter to all functions which wants to manipulate the world on a given input or trigger. The function itself picks the relevant information from the gamestate blob, do something with it, manipulate the gamestate and return the gamestate. But this looks like a poor mans solution for the problem. If I put the whole gamestate into all functions, there is no benefit for me in contrast to global variables or the imperative approach. I could put just the relevant information into the functions and return the actions which will be taken for the given input. And one single function apply all the actions to the gamestate. But most functions need a lot of "relevant" information. move() need the object position, the velocity, the map for collision, position of all enemys, current health, ... So this approach does not seem to work either. So my question is how do I handle the massive amount of state in a functional programming language -- especially for game development?

    Read the article

  • Learning how to design knowledge and data flow [closed]

    - by max
    In designing software, I spend a lot of time deciding how the knowledge (algorithms / business logic) and data should be allocated between different entities; that is, which object should know what. I am asking for advice about books, articles, presentations, classes, or other resources that would help me learn how to do it better. I code primarily in Python, but my question is not really language-specific; even if some of the insights I learn don't work in Python, that's fine. I'll give a couple examples to clarify what I mean. Example 1 I want to perform some computation. As a user, I will need to provide parameters to do the computation. I can have all those parameters sent to the "main" object, which then uses them to create other objects as needed. Or I can create one "main" object, as well as several additional objects; the additional objects would then be sent to the "main" object as parameters. What factors should I consider to make this choice? Example 2 Let's say I have a few objects of type A that can perform a certain computation. The main computation often involves using an object of type B that performs some interim computation. I can either "teach" A instances what exact parameters to pass to B instances (i.e., make B "dumb"); or I can "teach" B instances to figure out what needs to be done when looking at an A instance (i.e., make B "smart"). What should I think about when I'm making this choice?

    Read the article

  • How far do I take Composition?

    - by whiterook6
    (Although I'm sure this is a common problem I really don't know what to search for. Composition is the only thing I could come up with.) I've read over and over that multiple inheritance and subclassing is really, really bad, especially for game entities. If I have three types of motions, five types of guns, and three types of armoring, I don't want to have to make 45 different classes to get all the possible combinations; I'm going to add a motion behavior, gun behavior, and armor behavior to a single generic object. That makes sense. But how far do I take this? I can have as many different types of behaviors as I can imagine: DamageBehavior, MotionBehavior, TargetableBehavior, etc. If I add a new class of behaviors then I need to update all the other classes that use them. But what happens when I have functionality that doesn't really fit into one class of behaviors? For example, my armor needs to be damageable but also updateable. And should I be able to have use more than one type of behavior on an entity at a time, such as two motion behaviors? Can anyone offer any wisdom or point me in the direction of some useful articles? Thanks!

    Read the article

  • Inventory Consignment Flow

    - by ipohfly
    Not sure whether this is the right place to ask this question, but here goes.. Currently I have requirement to add support for consignment transaction in our inventory module. I have a very limited understanding of what consignment means in inventory, i.e. Customer get stocks/products from Seller without actually buying them, the product just resides in the Customer's inventory and it's still owned by the Seller. Only when the Customer actually buy the stocks then only will the ownership of the stock is transferred. The issue is i can't imagine how the data will be presented to both the Customer and the Seller. What i know is that i would need to deduct the stock from the Seller's inventory when the Customer raise a request to get the stock through consignment, but what about the 'ownership' of the stocks/products? Does that mean i would need to create another column in my table to state that for each inventory it is owned by who? Anywhere i can get information on how i should work out an inventory module like this? Thanks.

    Read the article

  • Include in service layer all the application's functions or only the reusable ones?

    - by BornToCode
    Background: I need to build a main application with some operations (CRUD and more) (-in winforms), I need to make another application which will re-use some of the functions of the main application (-in webforms). I understood that using service layer is the best approach here. If I understood correctly the service should be calling the function on the BL layer (correct me if I'm wrong) The dilemma: In my main winform UI - should I call the functions from the BL, or from the service? (please explain why) Should I create a service for every single function on the BL even if I need some of the functions only in one UI? for example - should I create services for all the CRUD operations, even though I need to re-use only update operation in the webform? YOUR HELP IS MUCH APPRECIATED

    Read the article

  • How to automatically restart Tomcat7 on system reboots?

    - by coding crow
    I have installed Tomcat 7 on Ubuntu 12.04 LTS which run on an Amzon EC2 instance. Now I wish tomcat should restart automatically on system reboot. I read this blog which suggest adding below script to /etc/init.d/tomcat7 # Tomcat auto-start # # description: Auto-starts tomcat # processname: tomcat # pidfile: /var/run/tomcat.pid case $1 in start) sh /usr/share/tomcat7/bin/startup.sh ;; stop) sh /usr/share/tomcat7/bin/shutdown.sh ;; restart) sh /usr/share/tomcat7/bin/shutdown.sh sh /usr/share/tomcat7/bin/startup.sh ;; esac exit 0 and issue the following commands sudo chmod 755 /etc/init.d/tomcat7 sudo ln -s /etc/init.d/tomcat7 /etc/rc1.d/K99tomcat sudo ln -s /etc/init.d/tomcat7 /etc/rc2.d/S99tomcat sudo /etc/init.d/tomcat7 restart My Questions The tomcat7 already has script in it, where do we have to paste the suggested script? Is the suggested procedure correct?

    Read the article

  • Where can I find good (well organized) examples of game code?

    - by smasher
    Where can I find good (well organized) examples of game code? I'm hoping that I can pick up some organizational tips. Most examples in books are too short and leave out lots of detail for the sake of brevity. I'm particularly interested on how to group your variables and methods so that another programmer would know where to look in the code. For example initializers at the top, then methods that take input, then methods that update views. I don't care about a particular language, as long as its OOP. I looked at the Quake 2 and 3 sources, but they're straight C and not much help for getting tips on organizing your objects. So, have you seen some good source? Any pointers to code that makes you say "wow, that's well organized" would be great.

    Read the article

  • Two components offering the same functionality, required by different dependencies

    - by kander
    I'm building an application in PHP, using Zend Framework 1 and Doctrine2 as the ORM layer. All is going well. Now, I happened to notice that both ZF1 and Doctrine2 come with, and rely on, their own caching implementation. I've evaluated both, and while each has its own pro's and cons, neither of them stand out as superior to the other for my simple needs. Both libraries also seem to be written against their respective interfaces, not their implementations. Reasons why I feel this is an issue is that during the bootstrapping of my application, I have to configure two caching drivers - each with its own syntax. A mismatch is easily created this way, and it feels inefficient to set up two connections to the caching backend because of this. I'm trying to determine what the best way forward is, and would welcome any insights you may be able to offer. What I've thought up so far are four options: Do nothing, accept that two classes offering caching functionality are present. Create a Facade class to stick Zend's interface onto Doctrine's caching implementation. Option 2, the other way around - create a Facade to map Doctrine's interface on a Zend Framework backend. Use multiple-interface-inheritance to create one interface to rule them all, and pray that there aren't any overlaps (ie: if both have a "save" method, they'll need to accept params in the same order due to PHP's lack of proper polymorphism). What option is best, or is there a "None of the above" variant that I'm not aware of?

    Read the article

  • Layers - Logical seperation vs physical

    - by P.Brian.Mackey
    Some programmers recommend logical seperation of layers over physical. For example, given a DL, this means we create a DL namespace not a DL assembly. Benefits include: faster compilation time simpler deployment Faster startup time for your program Less assemblies to reference Im on a small team of 5 devs. We have over 50 assemblies to maintain. IMO this ratio is far from ideal. I prefer an extreme programming approach. Where if 100 assemblies are easier to maintain than 10,000...then 1 assembly must be easier than 100. Given technical limits, we should strive for < 5 assemblies. New assemblies are created out of technical need not layer requirements. Developers are worried for a few reasons. A. People like to work in their own environment so they dont step on eachothers toes. B. Microsoft tends to create new assemblies. E.G. Asp.net has its own DLL, so does winforms. Etc. C. Devs view this drive for a common assembly as a threat. Some team members Have a tendency to change the common layer without regard for how it will impact dependencies. My personal view: I view A. as silos, aka cowboy programming and suggest we implement branching to create isolation. C. First, that is a human problem and we shouldnt create technical work arounds for human behavior. Second, my goal is not to put everything in common. Rather, I want partitions to be made in namespaces not assemblies. Having a shared assembly doesnt make everything common. I want the community to chime in and tell me if Ive gone off my rocker. Is a drive for a single assembly or my viewpoint illogical or otherwise a bad idea?

    Read the article

  • What's the right/standard way of achieving separation of concerns?

    - by Ghanima
    Some background: I want to start developing games, and taking some of the advice given in this site, I've started with something simple and familiar, such as pong, tetris, etc. I want to take as much time as needed to make sure that I have the basics right before moving on to something bigger. I have medium programming experience but I realize making games is a different thing. I find myself wondering many things like should this be in a separate class? Should this module handle this stuff or is it better to let other modules have that kind of functionality? For example, the bouncing of a ball in pong, right now is handled in the ball module, but maybe it's better that some other module did it. Right now I have different modules: one for the graphics, one for the game logic, and others for the objects (depending on the kind of movement required, not all the objects are alike). I know I am asking a lot, any tips you have will be very much appreciated. Short question: What's the right or standard way of separating the modules? What have you found most effective? Is it enough to just keep the drawing (graphics) and the logic separate? Is it necessary to have a lot of classes? (for example for the objects in the game, to handle the movement, etc)

    Read the article

  • SDL - Getting a single keypress event instead of a keystate?

    - by MrKatSwordfish
    Right now I'm working on a simple SDL project, but I've hit an issue when trying to get a single keypress event to skip past a splash screen. Right now, there are 4 start-up splash screens that I would like to be able to skip with a single keypress (of any key). My issue is that, as of now, if I hold down a key, it skips through each splash screen to the very last one immediately. The splash screens are stored as an array of SDL surfaces which are all loaded at the initialization of the state. I have an variable called currentSplashImage that controls which element of the array is being rendered on the screen. I've set it up so that whenever there's a SDL_KEYDOWN event, it triggers a single incrementation of the currentSplashImage variable. So, I'm really not sure why my code isn't working correctly. For some reason, when I hold down a button, it seems to be treating the held button as a new key press event every time it ticks through the code. Does anyone know how I can go about fixing this issue? [Here's a snippet of code that I've been using...] void SplashScreenState::handleEvents() { SDL_PollEvent( &localEvent ); if ( localEvent.type == SDL_KEYDOWN ) { if ( currentSplashImage < 3 && currentSplashImage >= 0) { currentSplashImage++; } } else if ( localEvent.type == SDL_QUIT ) { smgaEngine.setRunning(false); } } I should also mention that the SDL_Event 'localEvent' is part of the GameState parent class, while this event handling code is part of a SplashScreenState subclass. If anyone knows why this is happening, or if there is any way to improve my code, It'd be helpful to me! :D I'm still a very new programmer, trying to learn. UPDATE: I added a std::cout line to that the code runs multiple times with a single KEYDOWN event. I also tried disabling SDL_EnableKeyRepeat, but it didn't fix the issue. void SplashScreenState::handleEvents() { SDL_PollEvent( &localEvent ); if ( localEvent.type == SDL_KEYDOWN ) { if ( currentSplashImage < 3 && currentSplashImage >= 0) { currentSplashImage++; std::cout << "KEYDOWN.."; //<---- test cout line } } else if ( localEvent.type == SDL_QUIT ) { smgaEngine.setRunning(false); } } This prints out "KEYDOWN..KEYDOWN..KEYDOWN.." in the cout stream when a button is held.

    Read the article

  • PHP - Internal APIs/Libraries - What makes sense?

    - by Mark Locker
    I've been having a discussion lately with some colleagues about the best way to approach a new project, and thought it'd be interesting to get some external thoughts thrown into the mix. Basically, we're redeveloping a fairly large site (written in PHP) and have differing opinions on how the platform should be setup. Requirements: The platform will need to support multiple internal websites, as well as external (non-PHP) projects which at the moment consist of a mobile app and a toolbar. We have no plans/need in the foreseeable future to open up an API externally (for use in products other than our own). My opinion: We should have a library of well documented native model classes which can be shared between projects. These models will represent everything in our database and can take advantage of object orientated features such as inheritance, traits, magic methods, etc. etc. As well as employing ORM. We can then add an API layer on top of these models which can basically accept requests and route them to the appropriate methods, translating the response so that it can be used platform independently. This routing for each method can be setup as and when it's required. Their opinion: We should have a single HTTP API which is used by all projects (internal PHP ones or otherwise). My thoughts: To me, there are a number of issues with using the sole HTTP API approach: It will be very expensive performance wise. One page request will result in several additional http requests (which although local, are still ones that Apache will need to handle). You'll lose all of the best features PHP has for OO development. From simple inheritance, to employing the likes of ORM which can save you writing a lot of code. For internal projects, the actual process makes me cringe. To get a users name, for example, a request would go out of our box, over the LAN, back in, then run through a script which calls a method, JSON encodes the output and feeds that back. That would then need to be JSON decoded, and be presented as an array ready to use. Working with arrays, as appose to objects, makes me sad in a modern PHP framework. Their thoughts (and my responses): Having one method of doing thing keeps things simple. - You'd only do things differently if you were using a different language anyway. It will become robust. - Seeing as the API will run off the library of models, I think my option would be just as robust. What do you think? I'd be really interested to hear the thoughts of others on this, especially as opinions on both sides are not founded on any past experience.

    Read the article

  • non-volatile virtual memory for C++ containers

    - by arieberman
    Is there a virtual memory management process that would allow a program to use the standard container structures and classes, but retain these structures and their data when the program is not running (or being used), for use by the program at a later time? This should be possible, but can it be done without changing the source code and its (container) declarations? Is there a standard way of doing this?

    Read the article

  • How do I implement input and movement with characters that get into vehicles?

    - by Xkynar
    I'm making a game similar to GTA2. When the player enters the vehicle, what happens in terms of logic? Does the player becomes the vehicle? Does the vehicle override the player movement? The main question is how should it look at a vehicle? I want to understand if the player becomes the car or if the player has a "motion state" like "driving, walking, flying" depending on what he is doing in a moment, I know there are tons of ways to implement vehicles in a game.

    Read the article

  • What do you do when one think the code isn't complicated enough?

    - by Chris
    After six months of development on a project, our stakeholders have had a "gut check" and have decided that the path that we've been walking (a custom designed application framework and data access layer) is holding us (the developers) back from quickly developing the features they would like to see. After several days of debate management and the development team have decided to scrap the current incarnation and start over using ASP.net MVC, with Entity Framework as the bases of the a 'quick and dirty', lets just get it done project. In days following, our senior developer who has never worked with MVC or Entity Framework has finally gotten into a sample project and done some work. His take on ASP.net MVC, "this is not software engineering". So my question is this; what do you do, when one doesn't think the code is complicated enough?

    Read the article

  • Service Layer - how broad should it be, and should it also be used from the local application?

    - by BornToCode
    The background: I need to build a desktop application with some operations (CRUD and more) (=winforms), I need to make another application which will re-use some of the functions of the main application (=webforms). I'm using service layer for reusing my functions. The service is calling the functions on the BL layer (correct me if I'm doing this wrong). so my desktop has 4 projects - DAL, BL, UI, WEBSERVICES. The dilemma (simple but I still need some more experienced opinions): In my main winform UI - should I call the functions from the BL - bl.getcustomers(), or do it similar to how I call it in the webform, and call the functions from the service - webservices.getcustomers? Should I create a service for every single function on the BL even if I need some of the functions only in one UI? for example - should I create services for all the CRUD operations, even though I need to re-use only update operation in the webform? YOUR HELP IS MUCH APPRECIATED

    Read the article

< Previous Page | 82 83 84 85 86 87 88 89 90 91 92 93  | Next Page >