Search Results

Search found 10131 results on 406 pages for 'natural sort'.

Page 337/406 | < Previous Page | 333 334 335 336 337 338 339 340 341 342 343 344  | Next Page >

  • Bespoke Development or Leverage SharePoint With Web Parts etc?

    - by Asim
    Hi all, We are currently in the process of drawing up a solution for an existing client, creating a number of eServices. The client currently have MOSS 2007. The proposed solution is to use MOSS as the launching pad for the eServices… The requirement involves drawing up several online forms which provide registration facilities as well as facilitating a workflow of some sort. I have been told that the proposed solution requires complex web forms. Most are complex forms with parent child details that have multiple windows. The proposed solution is to do some bespoke development, developing ASP .NET forms. These forms would be deployed under the _layouts folder of the current MOSS portal, inheriting the master page design on the current site. I have been told that this approach make development and deployment more simple, as well has having ‘complete integration’ with MOSS. My questions are: Is this the best way to leverage SharePoint – it seems like the proposed solution is not leveraging MOSS at all..! I thought perhaps utilizing Web Parts would be better, but I have been told that this is more complex and developing more smarter intuitive UI is more difficult. Is this really the case? If not, what should be the recommended approach? We will be utilizing Ultimus as the workflow engine. However, I have been recommended K2 Workflows. Anyone used both/have any opinions on either? Many thanks in advance! Kind Regards,

    Read the article

  • Good guidelines for developing an ecommerce application

    - by kaciula
    I'm making an ecommerce application on Android and, as it is my first serious project, I'm trying to figure out beforehand the best approach to do this. The application talks to a web service (magento api, meaning soap or xml rpc unfortunately) and gets all the content on the phone (product categories, product details, user credentials etc.). I think it should do lazy loading or something like that. So, I was thinking to keep the user credentials in a custom Object which will be kept in a SharedPreferences so that every Activity cand easily access it. I'll use a couple of ListViews to show the content and AsyncTask to fetch the data needed. Should I keep all the data in memory in objects or should I use some sort of cache or a local database? Also, I'm planning to use a HashMap with SoftReferences to hold the bitmaps I am downloading. But wouldn't this eat a lot of memory? How cand all the activities have access to all these objects (ecommerce basket etc.)? I'm thinking of passing them using Intents but this doesn't seem right to me. Can SharedPreferences be used for a lot of objects and are there any concurrency issues? Any pointers would be really appreciated. What are some good guidelines? What classes should I look into? Do you know of any resources on the Internet for me to check out?

    Read the article

  • unique items from an observableArray of object properties

    - by Joe
    I'm trying to extract unique properties from a knockout.js observableArray of objects, to populate a drop menu. Being new to knockout, I'm really struggling with this! I want to iterate over a contacts list, and populate a drop menu with a unique value from each person object within the observableArray. So in my code example below, I wish to populate my drop menu with a list of people 'type' - family, friend etc. Looking on Google, I found a similar function, but it does not return any values, even if I console.log the results? //dummy data more rows in actual code... var people = [ { name: "Contact 1", address: "1, a street, a town, a city, AB12 3CD", tel: "0123456789", email: "[email protected]", type: "family" }, { name: "Contact 2", address: "1, a street, a town, a city, AB12 3CD", tel: "0123456789", email: "[email protected]", type: "friend" } ]; function ContactsViewModel(people) { var self = this; self.contacts = ko.observableArray(people); self.uniqueSelect = ko.dependentObservable(function() { return( ko.utils.arrayGetDistinctValues(self.contacts().type).sort()); }, self); }; ko.applyBindings(new ContactsViewModel()); And HTML template <p>Show me: <select data-bind="options: ContactsViewModel.uniqueSelect"></select></p> Any help appreciated, as a noob I'm lost! Thanks

    Read the article

  • What arguments to use to explain why a SQL DB is far better then a flat file

    - by jamone
    The higher ups in my company were told by good friends that flat files are the way to go, and we should switch from MS SQL server to them for everything we do. We have over 300 servers and hundreds of different databases. From just the few I'm involved with we have 10 billion records in quite a few of them with upwards of 100k new records a day and who knows how many updates... Me and a couple others need to come up with a response saying why we shouldn't do this. Most of our stuff is ASP.NET with some legacy ASP. We thought that making a simple console app that tests/times the same interactions between a flat file (stored on the network) and SQL over the network doing large inserts, searches, updates etc along with things like network disconnects randomly. This would show them how bad flat files can be espically when you are dealing with millions of records. What things should I use in my response? What should I do with my demo code to illustrate this? My sort list so far: Security Concurent access Performance with large ammounts of data Ammount of time to do such a massive rewrite/switch Lack of transactions PITA to map relational data to flat files I fear that this will be a great post on the Daily WTF someday if I can't stop it now.

    Read the article

  • Spring 3.0 vs J2EE 6.0

    - by StudiousJoseph
    Hi everybody, I'm confronted with a situation... I've been asked to give an advise regarding which approach to take, in terms of J2EE development between Spring 3.0 and J2EE 6.0. I was, and still am, a promoter of Spring 2.5 over classic J2EE 5 development, specially with JBoss, I even migrated old apps to Spring and influenced the re-definition of the development policy here to include Spring specific APIs, and helped the development of a strategic plan to foster more lightweight solutions like Spring + Tomcat, instead of the heavier ones of JBoss, right now, we're using JBoss merely as a Web container, having what i call the "container inside the container paradox", that is, having Spring apps, with most of its APIs, running inside JBoss, So we're in the process of migrating to tomcat. However, with the coming of J2EE 6.0 many features, that made Spring attractive at that time, easy deployment, less-coupling, even some sort of D.I, etc, seems to have been mimicked, in one way or the other. JSF 2.0, JPA 2.0, WebBeans, WebProfiles, etc. So, the question goes... From your point of view, how save, and logical, it is to continue to invest in a non-standard J2EE development framework like Spring given the new perspectives offered by J2EE 6.0? Can we talk about maybe 3 or 4 more years of Spring development, or do you recommend early adoption of J2EE 6.0 APIs and it's practices? I'll appreciate any insights with this...

    Read the article

  • [WEB] Local/Dev/Live deployment - best workflow

    - by Adam Kiss
    Hello, situation We our little company with 3 people, each has a localhost webserver and most projects (previous and current) are on one PC network shared disk. We have virtual server, where some of our clients' sites and our site. Our standard workflow is: Coder PC ? Programmer localhost ? dev domain (client.company.com) ? live version (client.com) It often happens, that there are two or three guys working on same projects at the same time - one is on dev version, two are on localhost. When finished, we try to synchronize the files on dev version and ideally not to mess (thanks ILMV:]) up any files, which *knock knock * doesn't happen often. And then one of us deploys dev version on live webserver. question we are looking for a way to simplify this workflow while updating websites - ideally some sort of diff uploader or VCS probably (Git/SVN/VCS/...), but we are not completely sure where to begin or what way would be ideal, therefore I ask you, fellow stackoverflowers for your experience with website / application deployment and recommended workflow. We probably will also need to use Mac in process, so if it won't be a problem, that would be even better. Thank you

    Read the article

  • How do I know which include path will be used in PHP?

    - by Joe Majewski
    When I run phpinfo() and look by the Configuration category under PHP Core, I see a directive titled include_path, with a local value and a master value. In this case, my local value is set to .: ./include: ../include: /usr/share/php: /usr/share/php/smarty: /usr/share/pear and my master value is set to .: /usr/share/php: /usr/share/pear: /usr/share/php/pear: /usr/share/php/smarty The reason I am trying to learn how this works is because there is a file in the system I am working on titled Smarty.class.php, which I'm sure sounds very familiar to anyone who uses Smarty Templating Engine. One of the PHP files has the following includes: require_once("Smarty.class.php"); require_once("user_info_class.inc"); The file user_info_class.inc is in the same directory as the file making the include, which makes perfect sense to me, and is the way that I've always referenced files. I decided that I wanted to open up the Smarty.class.php file and had assumed it would be in the same directory, but it was not. After doing a bit of digging, I discovered those php_ini variables, and was finally able to locate the file in the directory usr/share/php/smarty/. So it would seem that when making an include, it follows some sort of order between the Local and Master values for the include_path. Assuming that my deductions were correct thus far, can someone explain the order in which PHP searches for the files to be included?

    Read the article

  • Recommendations to handle development and deployment of php web apps using shared project code

    - by Exception e
    I am wondering what the best way (for a lone developer) is to develop a project that depends on code of other projects deploy the resulting project to the server I am planning to put my code in svn, and have shared code as a separate project. There are problems with svn:externals which I cannot fully estimate. I've read subversion:externals considered to be an anti-pattern, and How do you organize your version control repository, but there is one special thing with php-projects (and other interpreted source code): there is no final executable resulting from your libraries. External dependencies are thus always on raw source code. Ideally I really want to be able to develop simultaneously on one project and the projects it dependends on. Possible way: Check out a projects' dependency in a sub folder as a working copy of the trunk. Problems I foresee: When you want to deploy a project, you might want to freeze its dependencies, right? The dependency code should not end up as a duplicate in the projects repository, I think. *(update1: I additionally assume svn:ignore will pose problems if I cannot fall back on symlinks, see my comment) I am still looking for suggestions that do not require the use junction points. They are a sort of unsupported hack in winxp, which may break some programs* This leads me to the last part of the question (as one has influence on the other): how do you deploy apps whith such dependencies? I've looked into BuildOut for Python, but it seems to be tightly related to the python ecosystem (resolving and fetching python modules from the web etc). I am very eager to learn about your best practices.

    Read the article

  • What is wrong with the JavaScript event handling in this example? (Using click() and hover() jQuery

    - by Bungle
    I'm working on a sort of proof-of-concept for a project that approximates Firebug's inspector tool. For more details, please see this related question. Here is the example page. I've only tested it in Firefox: http://troy.onespot.com/static/highlight.html The idea is that, when you're mousing over any element that can contain text, it should "highlight" with a light gray background to indicate the boundaries of that element. When you then click on the element, it should alert() a CSS selector that matches it. This is somewhat working in the example linked above. However, there's one fundamental problem. When mousing over from the top of the page to the bottom, it will pick up the paragraphs, <h1> element, etc. But, it doesn't get the <div>s that encompass those paragraphs. However, for example, if you "sneak up" on the <div> that contains the two paragraphs "The area was settled..." and "Austin was selected..." from the left - tracing down the left edge of the page and entering the <div> just between the two paragraphs (see this screenshot) - then it is picked up. I assume this has something to do with the fact that I haven't attached an event handler to the <body> element (where you're entering the <div> from if you enter from the left), but I have attached handlers to the <p>s (where you're entering from if you come from the top or bottom). There are also other issues with mousing in and out elements - background colors that "stick" and the like - that I think are also related. As indicated in the related question posted above, I suspect there is something about event bubbling that I don't understand that is causing unexpected behavior. Can anyone spot what's wrong with my code? Thanks in advance for any help!

    Read the article

  • Sorting a list in PHP

    - by Andy
    Alright, so I've just (almost) finished my first little php script. Basically all it does is handling forms. It writes whatever the user put in to the fields of the form to a text file, then I include that text file in the index of the little page I have set up. So, currently it writes to the beginning of the text file (but doesn't overwrite previous data). But, several users wants this list to be alphabetically sorted instead. So what I want to do is make whatever they submit fall into the list in alphabetical order. The thing here is also that all I use in the text file are divs. The list is basically 'divided' into 3 parts. 'Title', 'Link', and 'Poster'. All I have done is positioned them with css. So, can I sort this list (the titles, in this case) alphabetically and still have the 'link' and 'poster' information assigned the way they already are, but just with the titles sorted? It don't use databases at all on my site, so there is no database handling at all used in this script (mainly because I'm not experienced at all in this).

    Read the article

  • How to create a "Shell IDList Array" to support drag-and-drop of virtual files from C# to Windows Ex

    - by JustABill
    I started trying to implement drag-and-drop of virtual files (from a C# 4/WPF app) with this codeplex tutorial. After spending some time trying to figure out a DV_E_FORMATETC error, I figured out I need to support the "Shell IDList Array" data format. But there seems to be next to zero documentation on what this format actually does. After some searching, I found this page on advanced data transfer which said that a Shell IDList Array was a pointer to a CIDA structure. This CIDA structure then contains the number of PIDLs, and a list of offsets to them. So what the hell is a PIDL? After some more searching, this page sort of implies it's a pointer to an ITEMIDLIST, which itself contains a single member that is a list of SHITEMIDs. My next idea was to try dragging a file from another application with virtual files. I just got a MemoryStream back for this format. At least I know what class to provide for the thing, but that doesn't help at all for explaining what to put in it. So now that that's explained, I still have no idea how to create one of these things so that it's valid. There's two real questions here: What is a valid "abID" member for a SHITEMID? These virtual files only exist with my program; will the thing they are dragged to pass the "abID" back later when it executes GetData? Do they have to be system-unique? Why are there two levels of lists; a list of PIDLs and each PIDL has a list of SHITEMIDs? I'm assuming one of them is one for each file, but what's the other one for? Multiple IDs for the same file? Any help or even links that explain what I should be doing would be greatly appreciated.

    Read the article

  • Use continue or Checked Exceptions when checking and processing objects

    - by Johan Pelgrim
    I'm processing, let's say a list of "Document" objects. Before I record the processing of the document successful I first want to check a couple of things. Let's say, the file referring to the document should be present and something in the document should be present. Just two simple checks for the example but think about 8 more checks before I have successfully processed my document. What would have your preference? for (Document document : List<Document> documents) { if (!fileIsPresent(document)) { doSomethingWithThisResult("File is not present"); continue; } if (!isSomethingInTheDocumentPresent(document)) { doSomethingWithThisResult("Something is not in the document"); continue; } doSomethingWithTheSucces(); } Or for (Document document : List<Document> documents) { try { fileIsPresent(document); isSomethingInTheDocumentPresent(document); doSomethingWithTheSucces(); } catch (ProcessingException e) { doSomethingWithTheExceptionalCase(e.getMessage()); } } public boolean fileIsPresent(Document document) throws ProcessingException { ... throw new ProcessingException("File is not present"); } public boolean isSomethingInTheDocumentPresent(Document document) throws ProcessingException { ... throw new ProcessingException("Something is not in the document"); } What is more readable. What is best? Is there even a better approach of doing this (maybe using a design pattern of some sort)? As far as readability goes my preference currently is the Exception variant... What is yours?

    Read the article

  • Getting my head around object oriented programing

    - by nLL
    I am entry level .Net developer and using it to develop web sites. I started with classic asp and last year jumped on the ship with a short C# book. As I developed I learned more and started to see that coming from classic asp I always used C# like scripting language. For example in my last project I needed to encode video on the webserver and wrote a code like public class Encoder { Public static bool Encode(string videopath) { ...snip... return true; } } While searching samples related to my project I’ve seen people doing this public class Encoder { Public static Encode(string videopath) { EncodedVideo encoded = new EncodedVideo(); ...snip... encoded.EncodedVideoPath = outputFile; encoded.Success = true; ...snip... } } public class EncodedVideo { public string EncodedVideoPath { get; set; } public bool Success { get; set; } } As I understand second example is more object oriented but I don’t see the point of using EncodedVideo object. Am I doing something wrong? Does it really necessary to use this sort of code in a web app?

    Read the article

  • Switch gettext translated language with original language

    - by Ruben
    Hi everyone, I started my PHP application with all text in German, then used gettext to extract all strings and translate them to English. So, now I have a .po file with all msgids in German and msgstrs in English. I want to switch them, so that my source code contains the English as msgids. There are numerous reasons for this: More translators will know English, so it is only appropriate to serve them up a file with msgids in English. I could always switch the file before I give it out and after I receive it It would help me to write English object & function names and comments if the content text was also English. I'd like to do that, so the project is more open to other Open Source collaborators (more likely to know English than German). I could do this manually and this is the sort of task where I anticipate it will take me more time to write an automated routine for it (because I'm very bad with shell scripts) than do it by hand. But I also anticipate despising every minute of manual computer labour (feels like a oxymoron, right?) like I always do. Has someone done this before? I figured this would be a common problem, but couldn't find anything. Many thanks ahead. Sample Problem: <title><?=_('Routinen')?></title> #: /users/ruben/sites/v/routinen.php:43 msgid "Routinen" msgstr "Routines"

    Read the article

  • Need help with Layouts in Swing

    - by Malki
    Hi, I'm using Java Swing and I have the following problem: I have a class TnaiPanel that extends JPanel. In this class I am creating 3 components and then lay them out in a horizontal line using a BoxLayout. Also, I have a class TnaimDinamimPanel that also extends JPanel. This class contains multiple occurances of TnaiPanel, layed out vertically using a BoxLayout. Also, I have a class MainFrame that extends JFrame. This frame contains a menu-bar and one main panel. The main panel can change (when choosing a certain menu-item, I create a new panel and set it to show as the main panel of the frame). Now, for some reason I get "BoxLayout can't be shared" when I add the newly created TnaimDinamimPanel to the components of the frame. I don't mind using different layout objects. The result I want to get is a sort of "table" of components, where each TnaiPanel will have fixed component sizes and spacing, essentially serving the role os a "row" in the "table". Thanks, Malki.

    Read the article

  • Why does Tex/Latex not speed up in subsequent runs?

    - by Debilski
    I really wonder, why even recent systems of Tex/Latex do not use any caching to speed up later runs. Every time that I fix a single comma*, calling Latex costs me about the same amount of time, because it needs to load and convert every single picture file. (* I know that even changing a tiny comma could affect the whole structure but of course, a well-written cache format could see the impact of that. Also, there might be situations where 100% correctness is not needed as long as it’s fast.) Is there something in the language of Tex which makes this complicated or impossible to accomplish or is it just that in the original implementation of Tex, there was no need for this (because it would have been slow anyway on those large computers)? But then on the other hand, why doesn’t this annoy other people so much that they’ve started a fork which has some sort of caching (or transparent conversion of Tex files to a format which is faster to parse)? Is there anything I can do to speed up subsequent runs of Latex? Except from putting all the stuff into chapterXX.tex files and then commenting them out?

    Read the article

  • Trying to find a PHP5 API-based embeddable CMS

    - by StrangeElement
    I've been making the rounds for a CMS that I can use as an API, in a sort of "embedded" mode. I mean by this that I don't want the CMS to do any logic or presentation. I want it to be used as an API, which I can then use within an existing site. I don't want to be tied to the architecture of the CMS. A good example of this is NC-CMS (http://www.nconsulting.ca/nc-cms/). All it needs is an include at the top, then wherever editable content is desired it's only a function call with a unique label. It's also perfect in the sense that it allows to differentiate between small strings (like titles, labels) and texts (which require a rich-text editor). It's the only CMS I found that fits this description, but it is a little too light as it does not handle site structure. I need to be able to allow my client to add pages, choosing an existing template for the layout. A minimal back-end is required. Wordpress also fits some requirements in that it handles only content editing and allows freedom for the themes by letting them call the content where and how they want it. But it is article-based and backwards, in that it embeds sites (as themes) within its structure, rather than being embeddable in sites like NC. It's funny how checking out all the CMS out there, almost all of them claim that most CMS are not self-sufficient, that they do not handle application logic, while (almost) every single on I found with only one exception do so. Many are mostly article-based blog engines, which does not fit my need. I would appreciate any CMS that fits the general description.

    Read the article

  • Predicate crashing iPhone App!

    - by DVG
    To preface, this is a follow up to an inquiry made a few days ago: http://stackoverflow.com/questions/2981803/iphone-app-crashes-when-merging-managed-object-contexts Short Version: EXC_BAD_ACCESS is crashing my app, and zombie-mode revealed the culprit to be my predicate embedded within the fetch request embedded in my Fetched Results Controller. How does an object within an object get released without an explicit command to do so? Long Version: Application Structure Platforms View Controller - Games View Controller (Predicated upon platform selection) - Add Game View Controller When a row gets clicked on the Platforms view, it sets an instance variable in Games View for that platform, then the Games Fetched Results Controller builds a fetch request in the normal way: - (NSFetchedResultsController *)fetchedResultsController{ if (fetchedResultsController != nil) { return fetchedResultsController; } //build the fetch request for Games NSFetchRequest *request = [[NSFetchRequest alloc] init]; NSEntityDescription *entity = [NSEntityDescription entityForName:@"Game" inManagedObjectContext:context]; [request setEntity:entity]; //predicate NSPredicate *predicate = [NSPredicate predicateWithFormat:@"platform == %@", selectedPlatform]; [request setPredicate:predicate]; //sort based on name NSSortDescriptor *sortDescriptor = [[NSSortDescriptor alloc] initWithKey:@"name" ascending:YES]; NSArray *sortDescriptors = [[NSArray alloc] initWithObjects:sortDescriptor, nil]; [request setSortDescriptors:sortDescriptors]; //fetch and build fetched results controller NSFetchedResultsController *aFetchedResultsController = [[NSFetchedResultsController alloc] initWithFetchRequest:request managedObjectContext:context sectionNameKeyPath:nil cacheName:@"Root"]; aFetchedResultsController.delegate = self; self.fetchedResultsController = aFetchedResultsController; [sortDescriptor release]; [sortDescriptors release]; [predicate release]; [request release]; [aFetchedResultsController release]; return fetchedResultsController; } At the end of this method, the fetchedResultsController's _fetch_request - _predicate member is set to an NSComparisonPredicate object. All is well in the world. By the time - (NSInteger)tableView:(UITableView *)tableView numberOfRowsInSection:(NSInteger)section gets called, the _predicate is now a Zombie, which will eventually crash the application when the table attempts to update itself. I'm more or less flummoxed. I'm not releasing the fetched results controller or any of it's parts, and the only part getting dealloc'd is the predicate. Any ideas?

    Read the article

  • C89, Mixing Variable Declarations and Code

    - by rutski
    I'm very curious to know why exactly C89 compilers will dump on you when you try to mix variable declarations and code, like this for example: rutski@imac:~$ cat test.c #include <stdio.h> int main(void) { printf("Hello World!\n"); int x = 7; printf("%d!\n", x); return 0; } rutski@imac:~$ gcc -std=c89 -pedantic test.c test.c: In function ‘main’: test.c:7: warning: ISO C90 forbids mixed declarations and code rutski@imac:~$ Yes, you can avoid this sort of thing by staying away from -pedantic. But then your code is no longer standards compliant. And as anybody capable of answering this post probably already knows, this is not just a theoretical concern. Platforms like Microsoft's C compiler enforce this quick in the standard under any and all circumstances. Given how ancient C is, I would imagine that this feature is due to some historical issue dating back to the extraordinary hardware limitations of the 70's, but I don't know the details. Or am I totally wrong there?

    Read the article

  • What arguments to use to explain why SQL Server is far better then a flat file

    - by jamone
    The higher ups in my company were told by good friends that flat files are the way to go, and we should switch from SQL Server to them for everything we do. We have over 300 servers and hundreds of different databases. From just the few I'm involved with we have 10 billion records in quite a few of them with upwards of 100k new records a day and who knows how many updates... Me and a couple others need to come up with a response saying why we shouldn't do this. Most of our stuff is ASP.NET with some legacy ASP. We thought that making a simple console app that tests/times the same interactions between a flat file (stored on the network) and SQL over the network doing large inserts, searches, updates etc along with things like network disconnects randomly. This would show them how bad flat files can be especially when you are dealing with millions of records. What things should I use in my response? What should I do with my demo code to illustrate this? My sort list so far: Security Concurrent access Performance with large amounts of data Amount of time to do such a massive rewrite/switch Lack of transactions PITA to map relational data to flat files NTFS doesn't support tons of files in a directory well I fear that this will be a great post on the Daily WTF someday if I can't stop it now.

    Read the article

  • Best practice way to handle variable content margin?

    - by Aithne
    Ok, so a quick site that I am throwing together for a friend has about 20 static pages. Each one with a small amount of content. the div Container contains, well, the div Content, and the div Content contains, obviously, the content that changes on each page. Now, depending on the length of the content, I want a different margin at the top. The less content, then the larger the margin. Simply an aesthetic choice. For example, if the content almost fills the static sized container, there is less padding, but a 1 line page of content might be 1/3 of the way down the static container. Centering the content in the div wont do, as that creates too large of a margin. Whats the best way to handle this? A new class for each content with a different margin? A new Id, so that it is in its own special div positioned or margined differently? Inline css on each page to override the standard css for div Content? A differnt spacer div inside Container before Content on each page? Some sort of scripting along the lines of margin of Content = (ContainerHeight - ContentHeight) / 3? Whats the acceptable way of doing this? I don't want to get into bad habits.

    Read the article

  • What design pattern shall I use in this question?

    - by iyad al aqel
    To be frank, this is a homework question, so I'll tell you my opinion. Can you let me know my mistakes rather than giving me the solution? This is the question : Assume a restaurant that only offers the following two types of meals: (a) a full meal and (b)an economic meal. The full meal consists of the following food items and is served in the following order: 1. Appetizer 2. Drink 3. Main dish 4. Dessert Meanwhile the economic meal consists of the following food items and is served in the following order: 1. Drink 2. Main dish Identify the most appropriate design pattern that can be used to allow a customer to only order using one of the two types of meals provided and that the meal components must be served in the given order. I'm confused between the Factory and the Iterator and using them both together. Using the factory Pattern we can create the two meals full and economic and provide the user with with a base object class that will decide upon. But how can we enforce the ordering of the elements, I thought of using the iterator along that will iterate through the the composite of the two created factories sort of speak. What do you think?

    Read the article

  • OracleGlobalization.SetThreadInfo() ORA-12705 Error

    - by michele
    Hi guys! I'm stuck in a problem, i cannot workaround! I have a Oracle client 11, with registry key set to AMERICAN_AMERICA.WE8ISO8859P1. I cannot edit this key, but my application must get data from Oracle in Italian culture format. So I want to edit culture info form my application only. I'm trying to using OracleGlobalization class in ODP.NET library before my Application.Run(), to set culture for my thread: OracleGlobalization og = OracleGlobalization.GetThreadInfo(); //OracleGlobalization.SetThreadInfo(OracleGlobalization.GetThreadInfo()); og.Calendar = "GREGORIAN"; og.Comparison = "BINARY"; og.Currency = "€"; og.DateFormat = "DD-MON-RR"; og.DateLanguage = "ITALIAN"; og.DualCurrency = "€"; og.ISOCurrency = "ITALY"; og.Language = "ITALIAN"; og.LengthSemantics = "BYTE"; og.NCharConversionException = false; og.NumericCharacters = ",."; og.Sort = "WEST_EUROPEAN"; og.Territory = "ITALY"; OracleGlobalization.SetThreadInfo(og); I get always the same error: ORA-12705: Cannot access NLS data files or invalid environment specified. I really don't know ho to solve this problem! Any hint? I'm working on a Win7 pc with VisualStudio 2008. Thank you in advance!

    Read the article

  • Should client-server code be written in one "project" or two?

    - by Ricket
    I've been beginning a client-server application. At first I naturally created two projects in Eclipse, two source control repositories, etc. But I'm quickly seeing that there is a bit of shared code between the two that would probably benefit to sharing instead of copying. In addition, I've been learning and trying test-driven development, and it seems to me that it would be easier to test based on real client components rather than having to set up a huge amount of code just to mock something, when the code is probably mostly in the client. My biggest concern in merging the client and server is of security; how do I ensure that the server pieces of the code do not reach an user's computer? So especially if you are writing client-server applications yourself (and especially in Java, though this can turn into a language-agnostic question if you'd like to share your experience with this in other languages), what sort of separation do you keep between your client and server code? Are they just in different packages/namespaces or completely different binaries using shared libraries, or something else entirely? How do you test the code together and yet ship separately?

    Read the article

  • Why doesn't Default route work using Html.ActionLink in this case?

    - by StuperUser
    I have a rather perculiar issue with routing. Coming back to routing after not having to worry about configuration for it for a year, I am using the default route and ignore route for resources: routes.IgnoreRoute("{resource}.axd/{*pathInfo}"); routes.MapRoute( "Default", // Route name "{controller}/{action}/{id}", // URL with parameters new { controller = "Home", action = "Index", id = UrlParameter.Optional }); I have a RulesController with an action for Index and Lorem and a Index.aspx, Lorem.aspx in Views Rules directory. I have an ActionLink aimed at Rules/Index on the maseter page: <li><div><%: Html.ActionLink("linkText", "Index", "Rules")%></div></li> The link is being rendered as http://localhost:12345/Rules/ and am getting a 404. When I type Index into the URL the application routes it to the action. When I change the default route action from "Index" to "Lorem", the action link is being rendered as http://localhost:12345/Rules/Index adding the Index as it's no longer on the default route and the application routes to the Index action correctly. I have used Phil Haack's Routing Debugger, but entering the url http://localhost:12345/Rules/ is causing a 404 using that too. I think I've covered all of the rookie mistakes, relevant SO questions and basic RTFMs. I'm assuming that "Rules" isn't any sort of reserved word in routing. Other than updating the Routes and debuugging them, what can I look at?

    Read the article

< Previous Page | 333 334 335 336 337 338 339 340 341 342 343 344  | Next Page >