Search Results

Search found 28957 results on 1159 pages for 'single instance'.

Page 382/1159 | < Previous Page | 378 379 380 381 382 383 384 385 386 387 388 389  | Next Page >

  • Pattern for performing game actions

    - by Arkiliknam
    Is there a generally accepted pattern for performing various actions within a game? A way a player can perform actions and also that an AI might perform actions, such as move, attack, self-destruct, etc. I currently have an abstract BaseAction which uses .NET generics to specify the different objects that get returned by the various actions. This is all implemented in a pattern similar to the Command, where each action is responsible for itself and does all that it needs. My reasoning for being abstract is so that I may have a single ActionHandler, and AI can just queue up different action implementing the baseAction. And the reason it is generic is so that the different actions can return result information relevant to the action (as different actions can have totally different outcomes in the game), along with some common beforeAction and afterAction implementations. So... is there a more accepted way of doing this, or does this sound alright?

    Read the article

  • Is it possible to remove a particular host key from SSH's known_hosts file?

    - by Kaustubh P
    I usually end up deleting the entire known_hosts file, which I have no problems for. But just out of curiosity, Is it possible to remove just a single entry? I opened the known_hosts file, and other han understanding that the file contains fingerprints for a given machine, I dont understand anything. Below is the message I faced, which led me to ask this question. Add correct host key in /home/wissen16/.ssh/known_hosts to get rid of this message. Offending key in /home/wissen16/.ssh/known_hosts:1 RSA host key for foo.com has changed and you have requested strict checking. Host key verification failed. Thanks.

    Read the article

  • What are good words for defining multiples?

    - by Scott Langham
    In databases you might take about one-to-many. This means there's one thing that maps to zero or more others. In this kind of style I'm looking for words that define min/max amounts of things. So far, I have: min max one 1 1 many 0 infinite optional 0 1 ??? 1 infinite Is there a single word that fits '???' to mean more than one? Do you have better alternatives for 'optional'? I'm wondering if there are any conventional names for those concepts?

    Read the article

  • What issues lead people to use Japanese-specific encodings rather than Unicode?

    - by Nicolas Raoul
    At work I come across a lot of Japanese text files in Shift-JIS and other encodings. It causes many mojibake (unreadable character) problems for all computer users. Unicode was intended to solve this sort of problem by defining a single character set for all languages, and the UTF-8 serialization is recommended for use on the Internet. So why doesn't everybody switch from Japanese-specific encodings to UTF-8? What issues with or disadvantages of UTF-8 are holding people back? EDIT: The W3C lists some known problems with Unicode, could this be a reason too?

    Read the article

  • Profiling Startup Of VS2012 &ndash; YourKit Profiler

    - by Alois Kraus
    The YourKit (v7.0.5) profiler is interesting in terms of price (79€ single place license, 409€ + 1 year support and upgrades) and feature set. You do get a performance and memory profiler in one package for which you normally need also to pay extra from the other vendors. As an interesting side note the profiler UI is written in Java because they do also sell Java profilers with the same feature set. To get all methods of a VS startup you need first to configure it to include System* in the profiled methods and you need to configure * to measure wall clock time. By default it does record only CPU times which allows you to optimize CPU hungry operations. But you will never see a Thread.Sleep(10000) in the profiler blocking the UI in this mode. It can profile as all others processes started from within the profiler but it can also profile the next or all started processes. As usual it can profile in sampling and tracing mode. But since it is a memory profiler as well it does by default also record all object allocations > 1MB. With allocation recording enabled VS2012 did crash but without allocation recording there were no problems. The CPU tab contains the time line of the application and when you click in the graph you the call stacks of all threads at this time. This is really a nice feature. When you select a time region you the CPU Usage estimation for this time window. I have seen many applications consuming 100% CPU only because they did create garbage like crazy. For this is the Garbage Collection tab interesting in conjunction with a time range. This view is like the CPU table only that the CPU graph (green) is missing. All relevant information except for GCs/s is already visible in the CPU tab. Very handy to pinpoint excessive GC or CPU bound issues. The Threads tab does show the thread names and their lifetime. This is useful to see thread interactions or which thread is hottest in terms of CPU consumption. On the CPU tab the call tree does exist in a merged and thread specific view. When you click on a method you get below a list of all called methods. There you can sort for methods with a high own time which are worth optimizing. In the Method List you can select which scope you want to see. Back Traces are the methods which did call you. Callees ist the list of methods called directly or indirectly by your method as a flat list. This is not a call stack but still very useful to see which methods were slow so you can see the “root” cause quite quickly without the need to click trough long call stacks. The last view Merged Calles is a call stacked view of the previous view. This does help a lot to understand did call each method at run time. You would get the same view with a debugger for one call invocation but here you get the full statistics (invocation count) as well. Since YourKit is also a memory profiler you can directly see which objects you have on your managed heap and which objects do hold most of your precious memory. You can in in the Object Explorer view also examine the contents of your objects (strings or whatsoever) to get a better understanding which objects where potentially allocating this stuff.   YourKit is a very easy to use combined memory and performance profiler in one product. The unbeatable single license price makes it very attractive to straightly buy it. Although it is a Java UI it is very responsive and the memory consumption is considerably lower compared to dotTrace and ANTS profiler. What I do really like is to start the YourKit ui and then start the processes I want to profile as usual. There is no need to alter your own application code to be able to inject a profiler into your new started processes. For performance and memory profiling you can simply select the process you want to investigate from the list of started processes. That's the way I like to use profilers. Just get out of the way and let the application run without any special preparations.   Next: Telerik JustTrace

    Read the article

  • MDW Reports–New Source Code ZIP File Available

    - by billramo
    In my MDW Reports series, I attached V1 of the RDL files in my post - May the source be with you! MDW Report Series Part 6–The Final Edition. Since that post, Rachna Agarwal from MSIT in India updated the RDL files that are ready to go in a single ZIP. The reports assume that they will ne uploaded to the Report Manager’s root folder and use a shared data source named MDW. The reports also integrate with the new Query Hash Statistics reports. You can download them from my SQLBlog.com download site.

    Read the article

  • Performance triage

    - by Dave
    Folks often ask me how to approach a suspected performance issue. My personal strategy is informed by the fact that I work on concurrency issues. (When you have a hammer everything looks like a nail, but I'll try to keep this general). A good starting point is to ask yourself if the observed performance matches your expectations. Expectations might be derived from known system performance limits, prototypes, and other software or environments that are comparable to your particular system-under-test. Some simple comparisons and microbenchmarks can be useful at this stage. It's also useful to write some very simple programs to validate some of the reported or expected system limits. Can that disk controller really tolerate and sustain 500 reads per second? To reduce the number of confounding factors it's better to try to answer that question with a very simple targeted program. And finally, nothing beats having familiarity with the technologies that underlying your particular layer. On the topic of confounding factors, as our technology stacks become deeper and less transparent, we often find our own technology working against us in some unexpected way to choke performance rather than simply running into some fundamental system limit. A good example is the warm-up time needed by just-in-time compilers in Java Virtual Machines. I won't delve too far into that particular hole except to say that it's rare to find good benchmarks and methodology for java code. Another example is power management on x86. Power management is great, but it can take a while for the CPUs to throttle up from low(er) frequencies to full throttle. And while I love "turbo" mode, it makes benchmarking applications with multiple threads a chore as you have to remember to turn it off and then back on otherwise short single-threaded runs may look abnormally fast compared to runs with higher thread counts. In general for performance characterization I disable turbo mode and fix the power governor at "performance" state. Another source of complexity is the scheduler, which I've discussed in prior blog entries. Lets say I have a running application and I want to better understand its behavior and performance. We'll presume it's warmed up, is under load, and is an execution mode representative of what we think the norm would be. It should be in steady-state, if a steady-state mode even exists. On Solaris the very first thing I'll do is take a set of "pstack" samples. Pstack briefly stops the process and walks each of the stacks, reporting symbolic information (if available) for each frame. For Java, pstack has been augmented to understand java frames, and even report inlining. A few pstack samples can provide powerful insight into what's actually going on inside the program. You'll be able to see calling patterns, which threads are blocked on what system calls or synchronization constructs, memory allocation, etc. If your code is CPU-bound then you'll get a good sense where the cycles are being spent. (I should caution that normal C/C++ inlining can diffuse an otherwise "hot" method into other methods. This is a rare instance where pstack sampling might not immediately point to the key problem). At this point you'll need to reconcile what you're seeing with pstack and your mental model of what you think the program should be doing. They're often rather different. And generally if there's a key performance issue, you'll spot it with a moderate number of samples. I'll also use OS-level observability tools to lock for the existence of bottlenecks where threads contend for locks; other situations where threads are blocked; and the distribution of threads over the system. On Solaris some good tools are mpstat and too a lesser degree, vmstat. Try running "mpstat -a 5" in one window while the application program runs concurrently. One key measure is the voluntary context switch rate "vctx" or "csw" which reflects threads descheduling themselves. It's also good to look at the user; system; and idle CPU percentages. This can give a broad but useful understanding if your threads are mostly parked or mostly running. For instance if your program makes heavy use of malloc/free, then it might be the case you're contending on the central malloc lock in the default allocator. In that case you'd see malloc calling lock in the stack traces, observe a high csw/vctx rate as threads block for the malloc lock, and your "usr" time would be less than expected. Solaris dtrace is a wonderful and invaluable performance tool as well, but in a sense you have to frame and articulate a meaningful and specific question to get a useful answer, so I tend not to use it for first-order screening of problems. It's also most effective for OS and software-level performance issues as opposed to HW-level issues. For that reason I recommend mpstat & pstack as my the 1st step in performance triage. If some other OS-level issue is evident then it's good to switch to dtrace to drill more deeply into the problem. Only after I've ruled out OS-level issues do I switch to using hardware performance counters to look for architectural impediments.

    Read the article

  • How should I account for the GC when building games with Unity?

    - by Eonil
    *As far as I know, Unity3D for iOS is based on the Mono runtime and Mono has only generational mark & sweep GC. This GC system can't avoid GC time which stops game system. Instance pooling can reduce this but not completely, because we can't control instantiation happens in the CLR's base class library. Those hidden small and frequent instances will raise un-deterministic GC time eventually. Forcing complete GC periodically will degrade performance greatly (can Mono force complete GC, actually?) So, how can I avoid this GC time when using Unity3D without huge performance degrade?

    Read the article

  • Why do we keep using CSV?

    - by Stephen
    Why do we keep using CSV? I recently made a shift to working the health domain and despite the wonderful work in data transfer standards, all data transfer is in CSV, both for reporting to external organisations, and for data migrations when implementing new systems. Unfortunately the use of CSV is the cause of the endless repetition of the same stupid errors, with the same waste of developer time. (bad escaping, failing to handle null fields etc.) I know we can do better, and anything between JSON and XML (depending on the instance) would be fine. (Most of the time this is data going from one MS SQLserver 2005 to another!) I feel as if each time I see this happening I am literally watching one developer waste anothers time. So why do we keep shafting each other? When will we stop?

    Read the article

  • Blender 2.63a window position unstable with dual monitor extended desktop

    - by Steve Smith
    I have two 1920 x 1200 HP monitors and running Ubuntu 12.10. If I set them up as an extended desktop, (both 1280 x 1024 positioned for 2048 x 1280, all my applications work fine EXCEPT Blender (even Inkscape). Blender works great on a single monitor but always defaults to a window size that uses all available real estate and it can't do that with 2048 x 1280. It launches, but jumps all over the screen. Does any know a way to start Blender in a fixed size window so it would run in just one monitor?

    Read the article

  • Silverlight Reporting Application Part 3.5 - Prism Background and WCF RIA [Series Intermission]

    Taking a step back before I dive into the details and full-on coding fun, I wanted to once again respond to a comment on my last post to clear up some things in regards to how I'm setting up my project and some of the choices I've made. Aka, thanks Ben. :) Prism Project Setup For starters, I'm not the ideal use case for a Prism application. In most cases where you've got a one-man team, Prism can be overkill as it is more intended for large teams who are geographically dispersed or in applications that have a larger scale than my Recruiting application in which you'll greatly benefit from modularity, delayed loading of xaps, etc. What Prism offers, though, is a manner for handling UI, commands, and events with the idea that, through a modular approach in which no parts really need to know about one another, I can update this application bit by bit as hiring needs change or requirements differ between offices without having to worry that changing something in the Jobs module will break something in, say, the Scheduling module. All that being said, here's a look at how our project breakdown for Recruit (MVVM/Prism implementation) looks: This could be a little misleading though, as each of those modules is actually another project in the overall Recruit solution. As far as what the projects actually are, that looks a bit like this: Recruiting Solution Recruit (Shell up there) - Main Silverlight Application .Web - Default .Web application to host the Silverlight app Infrastructure - Silverlight Class Library Project Modules - Silverlight Class Library Projects Infrastructure &Modules The Infrastructure project is probably something you'll see to some degree in any composite application. In this application, it is going to contain custom commands (you'll see the joy of these in a post or two down the road), events, helper classes, and any custom classes I need to share between different modules. Think of this as a handy little crossroad between any parts of your application. Modules on the other hand are the bread and butter of this application. Besides the shell, which holds the UI skeleton, and the infrastructure, which holds all those shared goodies, the modules are self-contained bundles of functionality to handle different concerns. In my scenario, I need a way to look up and edit Jobs, Applicants, and Schedule interviews, a Notification module to handle telling the user when different things are happening (i.e., loading from database), and a Menu to control interaction and moving between different views. All modules are going to follow the following pattern: The module class will inherit from IModule and handle initialization and loading the correct view into the correct region, whereas the Views and ViewModels folders will contain paired Silverlight user controls and ViewModel class backings. WCF RIA Services Since we've got all the projects in a single solution, we did not have to go the route of creating a WCR RIA Services Class Library. Every module has it's WCF RIA link back to the main .Web project, so the single Linq-2-SQL (yes, I said Linq-2-SQL, but I'll soon be switching to OpenAccess due to the new visual designer) context I'm using there works nicely with the scope of my project. If I were going for completely separating this project out and doing different, dynamically loaded elements, I'd probably go for the separate class library. Hope that clears that up. In the future though, I will be using that in a project that I've got in the "when I've got enough time to work on this" pipeline, so we'll get into that eventually- and hopefully when WCF RIA is in full release! Why Not use Silverlight Navigation/Business Template? The short answer- I'm a creature of habit, and having used Silverlight for a few years now, I'm used to doing lots of things manually. :) Plus, starting with a blank slate of a project I'm able to set up things exactly as I want them to be. In this case, rather than the navigation frame we would see in one of the templates, the MainRegion/ContentControl is working as our main navigation window. In many cases I will use theSilverlight navigation template to start things off, however in this case I did not need those features so I opted out of using that. Next time when I actually hit post #4, we're going to get into the modules and starting to get functionality into this application. Next week is also release week for the Q1 2010 release, so be sure to check out our annualWebinar Week (I might be biased, but Wednesday is my favorite out of the group). Did you know that DotNetSlackers also publishes .net articles written by top known .net Authors? We already have over 80 articles in several categories including Silverlight. Take a look: here.

    Read the article

  • Application development : method to manage backgound process

    - by Simon Dubois
    I am developing an application with different behavior depending on the arguments : - "-config" starts a Gtk window to change options, start and close the daemon. - "-daemon" starts a background process that does something every X minutes. I already know how to use fork/system/exec etc... But I would like to know the main logic of such application to : - restart or refresh the daemon when configuration change. - keep only one instance of the daemon. I have red that killing the daemon to restart it is not a clean way to do. How other applications do ? (ubuntuone, weather forecast, rss feed working with notification area) Thanks for your help. PS : I don't want to create a system-wide daemon, just a user application with a background process.

    Read the article

  • How should I handle using two databases with a legacy PHP application?

    - by Toby Allen
    I have a legacy PHP application that was written in 2004 and uses MSSQL as a database backend. At this stage MSSQL is still supported by PHP but only just via a Microsoft driver. I have looked at converting to mysql via automated tools, which work quite well, but I have quite complex views which need a lot of individual work to convert. I don't have a great deal of time to do this. Many tools I wish to use and frameworks I would like to move the application to, don't support MSSQL, so I was considering adding new features using a new mysql database and wondered if anyone had opinions on the pros and cons of using two seperate database backends in a single application?

    Read the article

  • How we call an RPC that not only calls external functions but also updates data structures?

    - by Kabumbus
    I have a simple C++ RPC that lets you have remote class instances that support live members (data structures) update as well as method calls. For example I had a class declared like this (pseudocode): class Sum{ public: RPC_FIELD(int lastSum); RPC_METHOD(int summ(int a, int b)) { lastSum = a + b; return lastSum; } }; On machine A I had its instance. On machines B and C I had created its instances and connected them to machine A. So now they actually do all processing on machine A but machines B, C get lastSum class field updates automatically (and can subscribe to update event). How to call (Nice Name) such a functionality when we have binding over network done automatically by RPC library? How RPC library creator can announce such feature?

    Read the article

  • Having a generic data type for a database table column, is it "good" practice?

    - by Yanick Rochon
    I'm working on a PHP project where some object (class member) may contain different data type. For example : class Property { private $_id; // (PK) private $_ref_id; // the object reference id (FK) private $_name; // the name of the property private $_type; // 'string', 'int', 'float(n,m)', 'datetime', etc. private $_data; // ... // ..snip.. public getters/setters } Now, I need to perform some persistence on these objects. Some properties may be a text data type, but nothing bigger than what a varchar may hold. Also, later on, I need to be able to perform searches and sorting. Is it a good practice to use a single database table for this (ie. is there a non negligible performance impact)? If it's "acceptable", then what could be the data type for the data column?

    Read the article

  • Create association between informations

    - by Andrea Girardi
    I deployed a project some days ago that allow to extract some medical articles using the results of a questionnaire completed by a user. For instance, if I reply on questionnaire I'm affected by Diabetes type 2 and I'm a smoker, my algorithm extracts all articles related to diabetes bubbling up all articles contains information about Diabetes type 2 and smoking. Basically we created a list of topic and, for every topic we define a kind of "guideline" that allows to extract and order informations for a user. I'm quite sure there are some better way to put on relationship two content but I was not able to find them on network. Could you suggest my a model, algorithm or paper to better understand this kind of problem and that helps me to find a faster, and more accurate way to extract information for an user?

    Read the article

  • Why did visual programming never take off and what future paradigms might change that?

    - by Rego
    As the number of "visual" OS's such as Android, iOS and the promised Windows 8 are becoming more popular, it does not seem to me that we programmers have new ways to code using these new technologies, due to a possible lack in new visual programming languages paradigms. I've seen several discussions about incompatibilities between the current coding development environment, and the new OS approaches from Windows 8, Android and other tablets OS's. I mean, today if we have a new tablet, it's almost a requirement for coding, to have, for instance, an external keyboard (due it seems to me it's very difficult to program using the touch screen), exactly because the coding assistance is not conceived to "write" thousands of lines of code. So, how advanced should be the "new" visual programming languages paradigms? Which characteristics these new paradigms would be required?

    Read the article

  • Broken links in content reports when tracking subdomains with Google Analytics

    - by Rob Sobers
    I have a tracking code that I use on my main site and my blog, which is on a subdomain: www.example.com blog.example.com I have a single profile in Google Analytics. I use advanced segments to look at traffic to the main site vs. traffic to the blog. Problem 1: When I'm browsing my content reports under Standard Reporting, the "Page" column doesn't show the top-level or sub-domain, so I can't differentiate www.example.com/index.html from blog.example.com/index.html easily. According to the docs, this filter is supposed to make GA prepend the hostname to the page URL in your content reports, but it doesn't seem to work. Problem 2: When I click on the little "Open in new window" icon next to a given page in a content report line, it always assumes the page lives on www.example.com, so I get 404s when the page is actually on blog.example.com. Is there a good solution for these subdomain tracking problems?

    Read the article

  • Habanero

    - by csharp-source.net
    An Enterprise Application Framework for .Net that is ideally suited for developing applications in an agile manner. The framework is used for producing an application from the data layer through to the front-end. Free open source under the LGPL license, it includes ORM, code generation and runtime UI generation to create one application for the desktop & web. Features: * ORM: Map database tables to objects in code * Persist property values to and from the database * Define all mapping in a single XML file * Switch between database vendors with one setting * Support for MySQL, MS Sql Server, MS Access, Oracle, PostgreSQL, SQLite, Firebird * FireStarter GUI class definitions xml manager * Generate user interfaces and map properties to controls * Develop for both desktop (with Windows Forms) and web (with Gizmox' Visual WebGUI) * Generate new projects and code files * Generate UI forms from templates * Reverse engineer class definitions from existing databases * Support variable data sources, including an in-memory database. Ships with Firestarter a free database reverse engineering, Domain Modelling and Code Generator.

    Read the article

  • Why does Network-Manager close when it cannot bind to any connection?

    - by Agmenor
    When Network-Manager is not able to bind my Ubuntu to any connection, after a while it closes. The icon in the notification area (or indicator area) disappears, and it does not run any more. For instance this happens when there is really no ethernet network nor wireless one. Nonetheless this is a problem, since I often plug in my ethernet cable long after having booted my computer. As a hack I therefore hit Alt + F2, and type nm-applet. But I am not satisfied with this solution, because I think it is not intuitive enough. Is there a way to keep Network-Manager always running? I have Network-Manager 0.8.1 on Ubuntu 10.10, but this has been so for a long time, at least for one year.

    Read the article

  • "Never to forget information" for programmers [closed]

    - by Milan
    Hi there! I'm new to programming and I would like to make a list of most important pieces of knowledge of programming and CS that would be useful no matter what specific programming language I would use in the future. For instance, if I would make this kind of list for Law studies, there would be stuff like Articles of Constitution etc. Those pieces of information I would put in Anki, and repeat it from time to time. Speaking in terms of CS and programming I mean on the most useful: mathematical theorems algorithms (examples of elegant solutions, comparison of two solutions etc.) pieces of code anything else that is vital (and very handy) to have in mind Do you think that making this kind of knowledge list makes sense?

    Read the article

  • Outline Shader Effect for Orthogonal Geometry in XNA

    - by Griffin
    I just recently started learning the art of shading, but I can't give an outline width to 2D, concave geometry when restrained to a single vertex/pixel shader technique (thanks to XNA). the shape I need to give an outline to has smooth, per-vertex coloring, as well as opacity. The outline, which has smooth, per-vertex coloring, variable width, and opacity cannot interfere with the original shape's colors. A pixel depth border detection algorithm won't work because pixel depth isn't a 3.0 semantic. expanding geometry / redrawing won't work because it interferes with the original shape's colors. I'm wondering if I can do something with the stencil/depth buffer outside of the shader functions since I have access to that through the graphics device. But I don't believe I'm able to manipulate actual values. How might I do this?

    Read the article

  • How would you code an AI engine to allow communication in any programming language?

    - by Tokyo Dan
    I developed a two-player iPhone board game. Computer players (AI) can either be local (in the game code) or remote running on a server. In the 2nd case, both client and server code are coded in Lua. On the server the actual AI code is separate from the TCP socket code and coroutine code (which spawns a separate instance of AI for each connecting client). I want to be able to further isolate the AI code so that that part can be a module coded by anyone in their language of choice. How can I do this? What tecniques/technology would enable communication between the Lua TCP socket/coroutine code and the AI module?

    Read the article

  • Automation at GUI or API Level in Scrum

    - by Sani Parwani
    I am a Automation Engineer. I use QTP for Automation. I wanted to know couple of things. In a scrum Project which has 2 weeks of work, how can complete automation be done in that time frame (talking only about the GUI Level)? Similarly, how can API Level of automated testing be accomplished, especially inside a single sprint? And what exactly is API level testing? How to begin with API Testing? I assume QTP is not the tool here certainly.

    Read the article

  • Breaking 1NF to model subset constraints. Does this sound sane?

    - by Chris Travers
    My first question here. Appologize if it is in the wrong forum but this seems pretty conceptual. I am looking at doing something that goes against conventional wisdom and want to get some feedback as to whether this is totally insane or will result in problems, so critique away! I am on PostgreSQL 9.1 but may be moving to 9.2 for this part of this project. To re-iterate: Does it seem sane to break 1NF in this way? I am not looking for debugging code so much as where people see problems that this might lead. The Problem In double entry accounting, financial transactions are journal entries with an arbitrary number of lines. Each line has either a left value (debit) or a right value (credit) which can be modelled as a single value with negatives as debits and positives as credits or vice versa. The sum of all debits and credits must equal zero (so if we go with a single amount field, sum(amount) must equal zero for each financial journal entry). SQL-based databases, pretty much required for this sort of work, have no way to express this sort of constraint natively and so any approach to enforcing it in the database seems rather complex. The Write Model The journal entries are append only. There is a possibility we will add a delete model but it will be subject to a different set of restrictions and so is not applicable here. If and when we allow deletes, we will probably do them using a simple ON DELETE CASCADE designation on the foreign key, and require that deletes go through a dedicated stored procedure which can enforce the other constraints. So inserts and selects have to be accommodated but updates and deletes do not for this task. My Proposed Solution My proposed solution is to break first normal form and model constraints on arrays of tuples, with a trigger that breaks the rows out into another table. CREATE TABLE journal_line ( entry_id bigserial primary key, account_id int not null references account(id), journal_entry_id bigint not null, -- adding references later amount numeric not null ); I would then add "table methods" to extract debits and credits for reporting purposes: CREATE OR REPLACE FUNCTION debits(journal_line) RETURNS numeric LANGUAGE sql IMMUTABLE AS $$ SELECT CASE WHEN $1.amount < 0 THEN $1.amount * -1 ELSE NULL END; $$; CREATE OR REPLACE FUNCTION credits(journal_line) RETURNS numeric LANGUAGE sql IMMUTABLE AS $$ SELECT CASE WHEN $1.amount > 0 THEN $1.amount ELSE NULL END; $$; Then the journal entry table (simplified for this example): CREATE TABLE journal_entry ( entry_id bigserial primary key, -- no natural keys :-( journal_id int not null references journal(id), date_posted date not null, reference text not null, description text not null, journal_lines journal_line[] not null ); Then a table method and and check constraints: CREATE OR REPLACE FUNCTION running_total(journal_entry) returns numeric language sql immutable as $$ SELECT sum(amount) FROM unnest($1.journal_lines); $$; ALTER TABLE journal_entry ADD CONSTRAINT CHECK (((journal_entry.running_total) = 0)); ALTER TABLE journal_line ADD FOREIGN KEY journal_entry_id REFERENCES journal_entry(entry_id); And finally we'd have a breakout trigger: CREATE OR REPLACE FUNCTION je_breakout() RETURNS TRIGGER LANGUAGE PLPGSQL AS $$ BEGIN IF TG_OP = 'INSERT' THEN INSERT INTO journal_line (journal_entry_id, account_id, amount) SELECT NEW.id, account_id, amount FROM unnest(NEW.journal_lines); RETURN NEW; ELSE RAISE EXCEPTION 'Operation Not Allowed'; END IF; END; $$; And finally CREATE TRIGGER AFTER INSERT OR UPDATE OR DELETE ON journal_entry FOR EACH ROW EXECUTE_PROCEDURE je_breaout(); Of course the example above is simplified. There will be a status table that will track approval status allowing for separation of duties, etc. However the goal here is to prevent unbalanced transactions. Any feedback? Does this sound entirely insane? Standard Solutions? In getting to this point I have to say I have looked at four different current ERP solutions to this problems: Represent every line item as a debit and a credit against different accounts. Use of foreign keys against the line item table to enforce an eventual running total of 0 Use of constraint triggers in PostgreSQL Forcing all validation here solely through the app logic. My concerns are that #1 is pretty limiting and very hard to audit internally. It's not programmer transparent and so it strikes me as being difficult to work with in the future. The second strikes me as being very complex and required a series of contraints and foreign keys against self to make work, and therefore it strikes me as complex, hard to sort out at least in my mind, and thus hard to work with. The fourth could be done as we force all access through stored procedures anyway and this is the most common solution (have the app total things up and throw an error otherwise). However, I think proof that a constraint is followed is superior to test cases, and so the question becomes whether this in fact generates insert anomilies rather than solving them. If this is a solved problem it isn't the case that everyone agrees on the solution....

    Read the article

< Previous Page | 378 379 380 381 382 383 384 385 386 387 388 389  | Next Page >