Search Results

Search found 20566 results on 823 pages for 'folder structure'.

Page 19/823 | < Previous Page | 15 16 17 18 19 20 21 22 23 24 25 26  | Next Page >

  • Is my class structure good enough?

    - by Rivten
    So I wanted to try out this challenge on reddit which is mostly about how you structure your data the best you can. I decided to challenge my C++ skills. Here's how I planned this. First, there's the Game class. It deals with time and is the only class main has access to. A game has a Forest. For now, this class does not have a lot of things, only a size and a Factory. Will be put in better use when it will come to SDL-stuff I guess A Factory is the thing that deals with the Game Objects (a.k.a. Trees, Lumberjack and Bears). It has a vector of all GameObjects and a queue of Events which will be managed at the end of one month. A GameObject is an abstract class which can be updated and which can notify the Event Listener The EventListener is a class which handles all the Events of a simulation. It can recieve events from a Game Object and notify the Factory if needed, the latter will manage correctly the event. So, the Tree, Lumberjack and Bear classes all inherits from GameObject. And Sapling and Elder Tree inherits from Tree. Finally, an Event is defined by an event_type enumeration (LUMBERJACK_MAWED, SAPPLING_EVOLUTION, ...) and an event_protagonists union (a GameObject or a pair of GameObject (who killed who ?)). I was quite happy at first with this because it seems quite logic and flexible. But I ended up questionning this structure. Here's why : I dislike the fact that a GameObject need to know about the Factory. Indeed, when a Bear moves somewhere, it needs to know if there's a Lumberjack ! Or it is the Factory which handles places and objects. It would be great if a GameObject could only interact with the EventListener... or maybe it's not that much of a big deal. Wouldn't it be better if I separate the Factory in three vectors ? One for each kind of GameObject. The idea would be to optimize research. If I'm looking do delete a dead lumberjack, I would only have to look in one shorter vector rather than a very long vector. Another problem arises when I want to know if there is any particular object in a given case because I have to look for all the gameObjects and see if they are at the given case. I would tend to think that the other idea would be to use a matrix but then the issue would be that I would have empty cases (and therefore unused space). I don't really know if Sapling and Elder Tree should inherit from Tree. Indeed, a Sapling is a Tree but what about its evolution ? Should I just delete the sapling and say to the factory to create a new Tree at the exact same place ? It doesn't seem natural to me to do so. How could I improve this ? Is the design of an Event quite good ? I've never used unions before in C++ but I didn't have any other ideas about what to use. Well, I hope I have been clear enough. Thank you for taking the time to help me !

    Read the article

  • Structure of a Git repository

    - by Luke Puplett
    Sorry if this is a duplicate, I looked. We're moving to Git. In Subversion, I'm used to having \trunk, \branches and \tags folders. With Git, switching between branches will replace the contents of the working directory, so am I right to assume that the way we used to work just doesn't apply with Git? My guess is that I'd have a repo folder with maybe a gitignore and readme.txt, then the folders for the projects that make up the repo, and that's it.

    Read the article

  • Why are folders disappearing in Windows XP?

    - by XenoFoxx
    I am researching a problem for a friend, and unfortunatly do not have direct access to his computer. I've tried to gather as much information as possible and I have researched it on various websites. I've not found anyone having the same problem my friend is having. So here goes: He has a media server in his home running Microsoft Windows XP. It has 3 drives, 1 for the OS and 2 for mass storage. Not long ago he went to access one of the mass storage media drives and it was empty, except for a single folder. His first assumption was that his roommate had deleted everything on the drive (excluding the remaining folder). He then checked the properties of the drive and it was still saying that the hard drive was nearly full. I told him to check the recycling bin, thinking that whoever deleted them didn't clear them from recycling and that they were still taking up space on the drive. My friend said the recycling bin was empty. So we have a drive that the Windows file management system says is empty (again except for the remaining folder), but the properties of the drive say it's mostly full. Now it gets weirder My friend tried to create a new folder on this drive and it auto-named itself "New Folder(1)" which means that it recognizes there is already a "New Folder" in that directory. He tried to rename it to a name that he KNEW was there previsouly, and Windows wouldn't allow it because it was a duplicate folder name. SO now it seems the folders are there, but not displaying in Windows Explorer. Both of us have no idea why this is occuring, why the folders vanished, why the one remaining folder didn't vanish, or how to make them visable again. Anyone else ever experience this? I can get more details if needed.

    Read the article

  • Redefine folder structure of document library with metadata

    - by Sachin
    Hi all, I have a problem in my sharepoint document library structure. Currently the document library consiste of folder sub-folder structure to store a document categorywise. Now our client want to redefine this folder structure with a metadata structure. Can any one tell me how can I use metadata instade of folder sub folder structure..? any related articles or links will be appriciated. Thanks Sachin

    Read the article

  • How to configure custom error page in Plesk 9.3 for non existing folder?

    - by Junior Mayhé
    I'm trying to configure Plesk in order to show website visitors a custom error html. The current hosted site is an ASP.NET site. This site shows its custom errors on error403.aspx and error404.aspx files. Now to comply with plesk, I've created error_docs with required files like forbidden.html, etc... When user try to navigate http://mysite.com/a_missing_page.aspx, the visitor is redirected to error404.aspx correctly. But when user try to navigate to a non existent directory http://mysite.com/a_missing_folder/ the site takes me to IIS 404 regular page. Plesk has Custom error documents activated on Web hosting settings. ASP.NET Error pages defined in web.config are showing fine. But it seems plesk wont show its custom html error documents. The bottom line here is about setting up a custom error page to a directory. Is it possible to do this using Plesk or do I have to change it manually on IIS?

    Read the article

  • Windows XP Disappearing Folders

    - by XenoFoxx
    I am researching a problem for a friend, and unfortunatly do not have direct access to his computer. I've tried to gather as much information as possible and I have researched it on various websites. I've not found anyone having the same problem my friend is having. So here goes: He has a media server in his home running Microsoft Windows XP. It has 3 drives, 1 for the OS and 2 for mass storage. Not long ago he went to access one of the mass storage media drives and it was empty, except for a single folder. His first assumption was that his roommate had deleted everything on the drive (excluding the remaining folder). He then checked the properties of the drive and it was still saying that the hard drive was nearly full. I told him to check the recycling bin, thinking that whoever deleted them didn't clear them from recycling and that they were still taking up space on the drive. My friend said the recycling bin was empty. So we have a drive that the windows file management system says is empty (again except for the remaining folder), but the properties of the drive say it's mostly full. Now it gets weirder My friend tried to create a new folder on this drive and it auto-named itself "New Folder(1)" which means that it recognizes there is already a "New Folder" in that directory. He tried to rename it to a name that he KNEW was there previsouly, and windows wouldn't allow it because it was a duplicate folder name. SO now it seems the folders are there, but not displaying in Windows Explorer. Both of us have no idea why this is occuring, why the folders vanished, why the one remaining folder didn't vanish, or how to make them visable again. Anyone else ever experience this? I can get more details if needed.

    Read the article

  • How to copy a directory structure but only include certain files (using windows batch files)

    - by Martin
    As the title says, how can I recursively copy a directory structure but only include some files. E.g given the following directory structure: folder1 folder2 folder3 data.zip info.txt abc.xyz folder4 folder5 data.zip somefile.exe someotherfile.dll The files data.zip and info.txt can appear everywhere in the directory structure. How can I copy the full directory structure, but only include files named data.zip and info.txt (all other files should be ignored)? The resulting directory structure should look like this: copy_of_folder1 folder2 folder3 data.zip info.txt folder4 folder5 data.zip

    Read the article

  • How to Open Any Folder as a Project in the NetBeans Platform

    - by Geertjan
    Typically, as described in the NetBeans Project Type Tutorial, you'll define a project type based on the presence of a file (e.g., "project.xml" or "customer.txt" or something like that) in a folder. I.e., if the file is there, then its parent, i.e., the folder that contains the file, is a project and should be opened in your application. However, in some scenarios (as with the HTML5 project type introduced in NetBeans IDE 7.3), the user should be able to open absolutely any folder at all into the application. How to create a project type that is that liberal? Here you go, the only condition that needs to be true is that the selected item in the "Open Project" dialog is a folder, as defined in the "isProject" method below. Nothing else. That's it. If you select a folder, it will be opened in your application, displaying absolutely everything as-is (since below there's no ProjectLogicalView defined): import java.beans.PropertyChangeListener; import java.io.IOException; import javax.swing.Icon; import org.netbeans.api.project.Project; import org.netbeans.api.project.ProjectInformation; import org.netbeans.spi.project.ProjectFactory; import org.netbeans.spi.project.ProjectState; import org.openide.filesystems.FileObject; import org.openide.loaders.DataFolder; import org.openide.loaders.DataObjectNotFoundException; import org.openide.nodes.FilterNode; import org.openide.util.Exceptions; import org.openide.util.ImageUtilities; import org.openide.util.Lookup; import org.openide.util.lookup.Lookups; import org.openide.util.lookup.ServiceProvider; @ServiceProvider(service = ProjectFactory.class) public class FolderProjectFactory implements ProjectFactory { @Override public boolean isProject(FileObject projectDirectory) { return DataFolder.findFolder(projectDirectory) != null; } @Override public Project loadProject(FileObject dir, ProjectState state) throws IOException { return isProject(dir) ? new FolderProject(dir) : null; } @Override public void saveProject(Project prjct) throws IOException, ClassCastException { // leave unimplemented for the moment } private class FolderProject implements Project { private final FileObject projectDir; private Lookup lkp; private FolderProject(FileObject dir) { this.projectDir = dir; } @Override public FileObject getProjectDirectory() { return projectDir; } @Override public Lookup getLookup() { if (lkp == null) { lkp = Lookups.fixed(new Object[]{ new Info(), }); } return lkp; } private final class Info implements ProjectInformation { @Override public Icon getIcon() { Icon icon = null; try { icon = ImageUtilities.image2Icon( new FilterNode(DataFolder.find( getProjectDirectory()).getNodeDelegate()).getIcon(1)); } catch (DataObjectNotFoundException ex) { Exceptions.printStackTrace(ex); } return icon; } @Override public String getName() { return getProjectDirectory().getName(); } @Override public String getDisplayName() { return getName(); } @Override public void addPropertyChangeListener(PropertyChangeListener pcl) { //do nothing, won't change } @Override public void removePropertyChangeListener(PropertyChangeListener pcl) { //do nothing, won't change } @Override public Project getProject() { return FolderProject.this; } } } } Even the ProjectInformation implementation really isn't needed at all, since it provides nothing more than the icon in the "Open Project" dialog, the rest (i.e., the display name in the "Open Project" dialog) is provided by default regardless of whether you have a ProjectInformation implementation or not.

    Read the article

  • Problem with executing dssp (secondary structure assignment)

    - by Mana
    I followed a previous post to install dssp on ubuntu How to install dssp (secondary structure assignments) under 12.04? After the installation I tried to execute dssp which was in /usr/local/bin/dssp But it gave me the following error bash: /usr/local/bin/dssp: cannot execute binary file Also I tried to analyse some trajectory files from a simulation using the code do_dssp -s md.tpr -f traj.xtc But it also failed giving me the error below Reading file md.tpr, VERSION 4.5.5 (single precision) Reading file md.tpr, VERSION 4.5.5 (single precision) Segmentation fault (core dumped) Please post me a solution for this problem. Thank you!

    Read the article

  • Most efficient way to rebuild a tree structure from data

    - by Ahsan
    Have a question on recursively populating JsTree using the .NET wrapper available via NuGet. Any help would be greatly appreciated. the .NET class JsTree3Node has a property named Children which holds a list of JsTree3Nodes, and the pre-existing table which contains the node structure looks like this NodeId ParentNodeId Data AbsolutePath 1 NULL News /News 2 1 Financial /News/Financial 3 2 StockMarket /News/Financial/StockMarket I have a EF data context from the the database, so my code currently looks like this. var parentNode = new JsTree3Node(Guid.NewGuid().ToString()); foreach(var nodeItem in context.Nodes) { parentNode.Children.Add(nodeItem.Data); // What is the most efficient logic to do this recursively? } as the inline comment says in the above code, what would be the most efficient way to load the JStree data on to the parentNode object. I can change the existing node table to suite the logic so feel free to suggest any changes to improve performance.

    Read the article

  • Recovery from URL structure change?

    - by Dejan Pelzel
    in July this year, we have changed the URL structure of the website from: Post: domain.com/blog/post/986/dance/heart-beats-dance-video-by-chinatsu/ Category: domain.com/blog/index/cosplay/ to Post: domain.com/dance/heart-beats-dance-video-by-chinatsu-986/ Category: domain.com/cosplay/ Everything was (supposedly) properly redirected with 301 redirects and it first seemed that the traffic returned after a couple of days, but it has now been close to 2 months and things keep going worse although Google is slowly indexing the changes. What is worrying me even more is that the Pages crawled per day from Webmaster Tools started drastically dropping a few days ago and has just reached a new low in months (from over 2000 to 700). Should I be worried or will things sort out eventually?

    Read the article

  • Organizing code for iOS app development

    - by KronoS
    I've been developing an app for the iOS platform, and as I've been going along, I've noticed that I've done a terrible job of keeping my files (.h, .m, .mm) organized. Is there any industry standards or best practices when it comes to organizing files for an iOS project? My files include custom classes (beside the view controllers), customized View Controllers, third-party content, code that works only on iOS 5.0+ and code that works on previous versions. What I'm looking for is a solution to keep things organized in a manner that others (or myself in years to come) can look at this and understand the basic structure of the application and not get lost in the multiple files found therein.

    Read the article

  • Read only array, deep copy or retrieve copies one by one? (Performance and Memory)

    - by Arthur Wulf White
    In a garbage collection based system, what is the most effective way to handle a read only array if such a structure does not exist natively in the language. Is it better to return a copy of an array or allow other classes to retrieve copies of the objects stored in the array one by one? @JustinSkiles: It is not very broad. It is performance related and can actually be answered specifically for two general cases. You only need very few items: in this situation it's more effective to retrieve copies of the objects one by one. You wish to iterate over 30% or more objects. In this cases it is superior to retrieve all the array at once. This saves on functions calls. Function calls are very expansive when compared to reading directly from an array. A good specific answer could include performance, reading from an array and reading indirectly through a function. It is a simple performance related question.

    Read the article

  • Structure gameobjects and call events

    - by waco001
    I'm working on a 2D tile based game in which the player interacts with other game objects (chests, AI, Doors, Houses etc...). The entire map will be stored in a file which I can read. When loading the tilemap, it will find any tile with the ID that represents a gameobject and store it in a hashmap (right data structure I think?). private static HashMap<Integer, Class<GameObject>> gameObjects = new HashMap<Integer, Class<GameObject>>(); How exactly would I go about calling, and checking for events? I figure that I would just call the update, render and input methods of each gameobject using the hashmap. Should I got towards a Minecraft/Bukkit approach (sorry only example I can think of), where the user registers an event, and it gets called whenever that event happens, and where should I go as in resources to learn about that type of programming, (Java, LWJGL). Or should I just loop through the entire hashmap looking for an event that fits? Thanks waco

    Read the article

  • SEO URL structure for tag search on site

    - by Theo G
    I am looking to add tags to each product on my site e.g. brown, x products under £x, second hand x, refurbished x etc. Once you click these tags it will then search for other tags that are similar. I was thinking of using a url structure of www.site.com/tags/this%is%the%tag%name and then simply have a page that shows the results of all the products with that tag. I heard a while back that google generally ignores or downgrades anything with ‘search’ in the url and was wondering if anyone had any experience with this? Also, would you say /tags/ is a pretty valid destination or is it best to break it down and add more levels e.g. /product-type/product%variation Thanks in advance!

    Read the article

  • How to setup an encrypted voip structure?

    - by strapakowsky
    What is the simplest way to set up a voip structure in a Linux machine with the following features: 1) Using free software 2) For computer-to-computer calls: end-to-end encryption set by the users, unpaid, no central authority (so skype is out) 3) For computer-to-phone calls: paid or unpaid, desirable encryption on the computer side if that is even possible 4) Ability to have a number to receive calls from regular phones My research concluded that the sip protocol is the most popular. However most discussions I've read on sip are too technical and I felt it discourages the regular user who wants to just click and talk. So I put the question above and created some separate questions about privacy with sip registrars, privacy with voip suppliers, what to look for in a sip registrar, what to look for in a voip provider. As for the software, I noticed most software either don't provide encryption (eg Ekiga) or the software doesn't work nicely and the project is abandoned (eg Twinkle), so no option seemed satisfying.

    Read the article

  • SOLID Principles and code structure

    - by S-Unit
    My question is on SOLID principles, because I couldn't answer a question (beyond the obvious meaning of the various principles of SOLID) at an interview and it really bugs me. I have done a couple of days worth of digging around and have yet to come up with a satisfactory summary. The question was: If you were to look at a .Net project that I told you strictly followed SOLID principles, what would expect to see in terms of the project and code structure ? I sort of floundered around a bit, didn't really answer the question and then bombed out. Any help or thoughts will really be appreciated.

    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

  • How are larger games organized?

    - by Matthew G.
    I'm using Java, but the language I'm using here is probably irrelevant. I'd like to create an economy based on an ancient civilization. I'm not sure how to design it. If I were working on a smaller game, like a copy of "Space Invaders", I'd have no problem structuring it like this. Game -Main Control Class --Graphics Class --Player Class --Enemy class I'd pass the graphics class to both the player and enemy class so they could call graphics functions. I don't understand how I'd do this for larger projects. Do I create a country class that contains a bunch of towns? Do the towns contain a lot building class, most contain classes of people? Do I make a path finding class that the player can access to get around? How exactly do I structure this and pass all these references around? Thanks.

    Read the article

  • Using NPM to share resources between UI projects [on hold]

    - by guy mograbi
    I am a UI team leader. My team has a lot of projects using different languages/technologies. In some parts we will rewrite (gradually - @Ampt this is for you) the application in order to enable new fresh technologies in and get old dinosaurs out. I am going to use Node Package Manager to set up an "all powerful" build/dependency manager. Can I use NPM to depend on a private github repository? Can I use NPM to depend on SVN? Will NPM play nice with quickbuild? Since each project might have a slightly different structure (think jetty/maven or play!framework) can I configure NPM to install some dependencies in different folders while still running it from the project's root? How can I, using NPM, get development resources out and build a packaged product? (like a war) Yes/No - is there a reason to use grunt? No discussion, just one liners.

    Read the article

  • How is this "interface"-like structure/pattern called?

    - by Sebastian Negraszus
    Let's assume we have an XmlDoc class that contains basic functionality for dealing with an XML data structure and saving/loading data to/from a file. Now we have several subclasses, A, B and C. They all inherit from XmlDoc and add component-specific methods for setting and getting lots of data. They are like "interfaces" but also add an implementation for the signatures. Finally, we have an ABCDoc class that joins all the "interfaces" via virtual multiple inheritence and adds some ABCDoc-specific stuff, such as using XMLDoc-methods to set an appropriate doc type. We may also have an ADoc class for only saving A data. How is this pattern called? "Interface" is not really the right word since interfaces usually do not contain an implementation. Bonus points for C++ code conventions.

    Read the article

  • Data structure supporting the following operations

    - by 500865
    I'm looking for a data structure for working with a set of data which is most efficient to do the following : Check whether an item has been categorized or not. (The categorized and uncategorized set are disjoint sets). Get the category of an item. Get all the items in a particular category. Get all the uncategorized items. Remove a particular item from the data set. I was thinking of having a Dictionary<String, Set<String>> to hold all the items in a given category, but that doesn't solve 2.

    Read the article

  • Data structure for file search

    - by poly
    I've asked this question before and I got a few answers/idea, but I'm not sure how to implement them. I'm building a telecom messaging solution. Currently, I'm using a database to save my transaction/messages for the network stack I've built, and as you know it's slower than using a data structure (hash, linkedlist, etc...). My problem is that the data can be really huge, and it won't fit in the memory. I was thinking of saving the records in a file and the a key and line number in a hash, then if I want to access some record then I can get the line number from the hash, and get it from the file. I don't know how efficient is this; I think the database is doing a way better job than this on my behalf. Please share whatever you have in mind.

    Read the article

  • Settlers-like terrain representation

    - by Olle
    Remember this beauty? I'm playing it now on my old Amiga 1200. My question is: How do you think they represented the terrain, data structure wise? Obviously it's some kind of points, with a height. Or hexagons. And how did they decide which dots were buildable? EDIT: I could rephrase the question to say "how do I achieve this kind of terrain", but I would still only be interested in how to do it on a machine with 1 MB of RAM and a 7 Mhz processor, because this is the machine i currently developing games for. If that seems like a vague or meaningless question to you, that's alright, but I'm still curious if someone has any knowledge about this.

    Read the article

< Previous Page | 15 16 17 18 19 20 21 22 23 24 25 26  | Next Page >