Search Results

Search found 2100 results on 84 pages for 'somewhat offtopic'.

Page 50/84 | < Previous Page | 46 47 48 49 50 51 52 53 54 55 56 57  | Next Page >

• How to repair an external harddrive?

  - by dodohjk

  I would like to reformat my hard disk, and if possible recover the (somewhat unimportant) contents if possible. I have a Western Digital 1TB hard drive which had a NTFS partition. I unplugged the drive without safely removing it first. At first a pop up was asking me to use a Windows OS to run the chkdsk /f command, however, in the effort to keep using a Linux OS I used the ntfsfix command on the ubuntu terminal Now, when I try to access the hard drive, it doesn't show up anymore in Nautilus. I tried reformatting it using Disk Utility, but it gives me an error message, and Gparted would hang on the "Scanning devices" step infinitely. Please comment any output that you would like to see and I will add it to my question. EDIT disk utility tells me is on /dev/sdb the command sudo fdisk -l gives dodohjk@DodosPC:~$ sudo fdisk -l [sudo] password for dodohjk: Disk /dev/sda: 250.1 GB, 250059350016 bytes 255 heads, 63 sectors/track, 30401 cylinders, total 488397168 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x0006fa8c Device Boot Start End Blocks Id System /dev/sda1 * 4094 482344959 241170433 5 Extended /dev/sda2 482344960 488396799 3025920 82 Linux swap / Solaris /dev/sda5 4096 31461127 15728516 83 Linux /dev/sda6 31463424 52434943 10485760 83 Linux /dev/sda7 52436992 62923320 5243164+ 83 Linux /dev/sda8 62924800 482344959 209710080 83 Linux Disk /dev/sdb: 1000.2 GB, 1000202043392 bytes 255 heads, 63 sectors/track, 121600 cylinders, total 1953519616 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x6e697373 This doesn't look like a partition table Probably you selected the wrong device. Device Boot Start End Blocks Id System /dev/sdb1 ? 1936269394 3772285809 918008208 4f QNX4.x 3rd part /dev/sdb2 ? 1917848077 2462285169 272218546+ 73 Unknown /dev/sdb3 ? 1818575915 2362751050 272087568 2b Unknown /dev/sdb4 ? 2844524554 2844579527 27487 61 SpeedStor Partition table entries are not in disk order I wrote something wrong here, however here the output of fsck /dev/sbd is dodohjk@DodosPC:~$ sudo fsck /dev/sdb fsck from util-linux 2.20.1 e2fsck 1.42.5 (29-Jul-2012) ext2fs_open2: Bad magic number in super-block fsck.ext2: Superblock invalid, trying backup blocks... fsck.ext2: Bad magic number in super-block while trying to open /dev/sdb The superblock could not be read or does not describe a correct ext2 filesystem. If the device is valid and it really contains an ext2 filesystem (and not swap or ufs or something else), then the superblock is corrupt, and you might try running e2fsck with an alternate superblock: e2fsck -b 8193 <device&gt;

  Read the article

• OpenGL - Cascaded shadow mapping - Texture lookup

  - by Silverlan

  I'm trying to implement cascaded shadow mapping in my engine, but I'm somewhat stuck at the last step. For testing purposes I've made sure all cascades encompass my entire scene. The result is currently this: The different intensity of the cascades is not on purpose, it's actually the problem. This is how I do the texture lookup for the shadow maps inside the fragment shader: layout(std140) uniform CSM { vec4 csmFard; // far distances for each cascade mat4 csmVP[4]; // View-Projection Matrix int numCascades; // Number of cascades to use. In this example it's 4. }; uniform sampler2DArrayShadow csmTextureArray; // The 4 shadow maps in vec4 csmPos[4]; // Vertex position in shadow MVP space float GetShadowCoefficient() { int index = numCascades -1; vec4 shadowCoord; for(int i=0;i<numCascades;i++) { if(gl_FragCoord.z < csmFard[i]) { shadowCoord = csmPos[i]; index = i; break; } } shadowCoord.w = shadowCoord.z; shadowCoord.z = float(index); shadowCoord.x = shadowCoord.x *0.5f +0.5f; shadowCoord.y = shadowCoord.y *0.5f +0.5f; return shadow2DArray(csmTextureArray,shadowCoord).x; } I then use the return value and simply multiply it with the diffuse color. That explains the different intensity of the cascades, since I'm grabbing the depth value directly from the texture. I've tried to do a depth comparison instead, but with limited success: [...] // Same code as above shadowCoord.w = shadowCoord.z; shadowCoord.z = float(index); shadowCoord.x = shadowCoord.x *0.5f +0.5f; shadowCoord.y = shadowCoord.y *0.5f +0.5f; float z = shadow2DArray(csmTextureArray,shadowCoord).x; if(z < shadowCoord.w) return 0.25f; return 1.f; } While this does give me the same shadow value everywhere, it only works for the first cascade, all others are blank: (I colored the cascades because otherwise the transitions wouldn't be visible in this case) What am I missing here?

  Read the article

• Cross-platform independent development

  - by Joe Wreschnig

  Some years ago, if you wrote in C and some subset of C++ and used a sufficient number of platform abstractions (via SDL or whatever), you could run on every platform an indie could get on - Linux, Windows, Mac OS of various versions, obscure stuff like BeOS, and the open consoles like the GP2X and post-death Dreamcast. If you got a contract for a closed platform at some point, you could port your game to that platform with "minimal" code changes as well. Today, indie developers must use XNA to get on the Xbox 360 (and upcoming Windows phone); must not use XNA to work anywhere else but Windows; until recently had to use Java on Android; Flash doesn't run on phones, HTML5 doesn't work on IE. Unlike e.g. DirectX vs. OpenGL or Windows vs. Unix, these are changes to the core language you write your code in and can't be papered over without, basically, writing a compiler. You can move some game logic into scripts and include an interpreter - except when you can't, because the iPhone SDK doesn't allow it, and performance suffers because no one allows JIT. So what can you do if you want a really cross-platform portable game, or even just a significant body of engine and logic code? Is this not a problem because the platforms have fundamentally diverged - it's just plain not worthwhile to try to target both an iPhone and the Xbox 360 with any shared code because such a game would be bad? (I find this very unlikely. I can easily see wanting to share a game between a Windows Mobile phone and an Android, or an Xbox 360 and an iPad.) Are interfaces so high-level now that porting time is negligible? (I might believe this for business applications, but not for games with strict performance requirements.) Is this going to become more pronounced in the future? Is the split going to be, somewhat scarily, still down vendor lines? Will we all rely on high-level middleware like Flash or Unity to get anything cross-platform done? tl;dr - Is porting a problem, is it going to be a bigger problem in the future, and if so how do we solve it?

  Read the article

• Achieving Zero Downtime Deployment

  - by MattW

  I am trying to achieve zero downtime deployments so I can deploy less during off hours and more during "slower" hours - or anytime, in theory. My current setup, somewhat simplified: Web Server A (.NET App) Web Server B (.NET App) Database Server (SQL Server) My current deployment process: "Stop" the sites on both Web Server A and B Upgrade the database schema for the version of the app being deployed Update Web Server A Update Web Server B Bring everything back online Current Problem This leads to a small amount of downtime each month - about 30 mins. I do this during off hours, so it isn't a huge problem - but it is something I'd like to get away from. Also - there is no way to really go 'back'. I don't generally make rollback DB scripts - only upgrade scripts. Leveraging The Load Balancer I'd love to be able to upgrade one Web Server at a time. Take Web Server A out of the load balancer, upgrade it, put it back online, then repeat for Web Server B. The problem is the database. Each version of my software will need to execute against a different version of the database - so I am sort of "stuck". Possible Solution A current solution I am considering is adopting the following rules: Never delete a database table. Never delete a database column. Never rename a database column. Never reorder a column. Every stored procedure must be versioned. Meaning - 'spFindAllThings' will become 'spFindAllThings_2' when it is edited. Then it becomes 'spFindAllThings_3' when edited again. Same rule applies to views. While, this seems a bit extreme - I think it solves the problem. Each version of the application will be hitting the DB in a non breaking way. The code expects certain results from the views/stored procedures - and this keeps that 'contract' valid. The problem is - it just seeps sloppy. I know I can clean up old stored procedures after the app is deployed for awhile, but it just feels dirty. Also - it depends on all of the developers following these rule, which will mostly happen, but I imagine someone will make a mistake. Finally - My Question Is this sloppy or hacky? Is anybody else doing it this way? How are other people solving this problem?

  Read the article

• Keeping up with New Releases

  - by Jeremy Smyth

  You can keep up with the latest developments in MySQL software in a number of ways, including various blogs and other channels. However, for the most correct (if somewhat dry and factual) information, you can go directly to the source.  Major Releases  For every major release, the MySQL docs team creates and maintains a "nutshell" page containing the significant changes in that release. For the current GA release (whatever that is) you'll find it at this location: https://dev.mysql.com/doc/mysql/en/mysql-nutshell.html  At the moment, this redirects to the summary notes for MySQL 5.6. The notes for MySQL 5.7 are also available at that website, at the URL http://dev.mysql.com/doc/refman/5.7/en/mysql-nutshell.html, and when eventually that version goes GA, it will become the currently linked notes from the URL shown above. Incremental Releases  For more detail on each incremental release, you can have a look at the release notes for each revision. For MySQL 5.6, the release notes are stored at the following location: http://dev.mysql.com/doc/relnotes/mysql/5.6/en/ At the time I write this, the topmost entry is a link for MySQL 5.6.15. Each linked page shows the changes in that particular version, so if you are currently running 5.6.11 and are interested in what bugs were fixed in versions since then, you can look at each subsequent release and see all changes in glorious detail. One really clever thing you can do with that site is do an advanced Google search to find exactly when a feature was released, and find out its release notes. By using the preceding link in a "site:" directive in Google, you can search only within those pages for an entry. For example, the following Google search shows pages within the release notes that reference the --slow-start-timeout option:     site:http://dev.mysql.com/doc/relnotes/mysql/ "--slow-start-timeout" By running that search, you can see that the option was added in MySQL 5.6.5 and also rolled into MySQL 5.5.20.   White Papers Also, with each major release you can usually find a white paper describing what's new in that release. In MySQL 5.6 there was a "What's new" whitepaper at this location: http://www.mysql.com/why-mysql/white-papers/whats-new-mysql-5-6/ You'll find other white papers at: http://www.mysql.com/why-mysql/white-papers/ Search the page for "5.6" to see any papers dealing specificallly with that version.

  Read the article

• Interviews: Going Beyond the Technical Quiz

  - by Tony Davis

  All developers will be familiar with the basic format of a technical interview. After a bout of CV-trawling to gauge basic experience, strengths and weaknesses, the interview turns technical. The whiteboard takes center stage and the challenge is set to design a function or query, or solve what on the face of it might seem a disarmingly simple programming puzzle. Most developers will have experienced those few panic-stricken moments, when one’s mind goes as blank as the whiteboard, before un-popping the marker pen, and hopefully one’s mental functions, to work through the problem. It is a way to probe the candidate’s knowledge of basic programming structures and techniques and to challenge their critical thinking. However, these challenges or puzzles, often devised by some of the smartest brains in the development team, have a tendency to become unnecessarily ‘tricksy’. They often seem somewhat academic in nature. While the candidate straight out of IT school might breeze through the construction of a Markov chain, a candidate with bags of practical experience but less in the way of formal training could become nonplussed. Also, a whiteboard and a marker pen make up only a very small part of the toolkit that a programmer will use in everyday work. I remember vividly my first job interview, for a position as technical editor. It went well, but after the usual CV grilling and technical questions, I was only halfway there. Later, they sat me alongside a team of editors, in front of a computer loaded with MS Word and copy of SQL Server Query Analyzer, and my task was to edit a real chapter for a real SQL Server book that they planned to publish, including validating and testing all the code. It was a tough challenge but I came away with a sound knowledge of the sort of work I’d do, and its context. It makes perfect sense, yet my impression is that many organizations don’t do this. Indeed, it is only relatively recently that Red Gate started to move over to this model for developer interviews. Now, instead of, or perhaps in addition to, the whiteboard challenges, the candidate can expect to sit with their prospective team, in front of Visual Studio, loaded with all the useful tools in the developer’s kit (ReSharper and so on) and asked to, for example, analyze and improve a real piece of software. The same principles should apply when interviewing for a database positon. In addition to the usual questions challenging the candidate’s knowledge of such things as b-trees, object permissions, database recovery models, and so on, sit the candidate down with the other database developers or DBAs. Arm them with a copy of Management Studio, and a few other tools, then challenge them to discover the flaws in a stored procedure, and improve its performance. Or present them with a corrupt database and ask them to get the database back online, and discover the cause of the corruption.

  Read the article

• Voxel terrain rendering with marching cubes

  - by JavaJosh94

  I was working on making procedurally generated terrain using normal cubish voxels (like minecraft) But then I read about marching cubes and decided to convert to using those. I managed to create a working marching cubes class and cycle through the densities and everything in it seemed to be working so I went on to work on actual terrain generation. I'm using XNA (C#) and a ported libnoise library to generate noise for the terrain generator. But instead of rendering smooth terrain I get a 64x64 chunk (I specified 64 but can change it) of seemingly random marching cubes using different triangles. This is the code I'm using to generate a "chunk": public MarchingCube[, ,] getTerrainChunk(int size, float dMultiplyer, int stepsize) { MarchingCube[, ,] temp = new MarchingCube[size / stepsize, size / stepsize, size / stepsize]; for (int x = 0; x < size; x += stepsize) { for (int y = 0; y <size; y += stepsize) { for (int z = 0; z < size; z += stepsize) { float[] densities = {(float)terrain.GetValue(x, y, z)*dMultiplyer, (float)terrain.GetValue(x, y+stepsize, z)*dMultiplyer, (float)terrain.GetValue(x+stepsize, y+stepsize, z)*dMultiplyer, (float)terrain.GetValue(x+stepsize, y, z)*dMultiplyer, (float)terrain.GetValue(x,y,z+stepsize)*dMultiplyer,(float)terrain.GetValue(x,y+stepsize,z+stepsize)*dMultiplyer,(float)terrain.GetValue(x+stepsize,y+stepsize,z+stepsize)*dMultiplyer,(float)terrain.GetValue(x+stepsize,y,z+stepsize)*dMultiplyer }; Vector3[] corners = { new Vector3(x,y,z), new Vector3(x,y+stepsize,z),new Vector3(x+stepsize,y+stepsize,z),new Vector3(x+stepsize,y,z), new Vector3(x,y,z+stepsize), new Vector3(x,y+stepsize,z+stepsize), new Vector3(x+stepsize,y+stepsize,z+stepsize), new Vector3(x+stepsize,y,z+stepsize)}; if (x == 0 && y == 0 && z == 0) { temp[x / stepsize, y / stepsize, z / stepsize] = new MarchingCube(densities, corners, device); } temp[x / stepsize, y / stepsize, z / stepsize] = new MarchingCube(densities, corners); } } } (terrain is a Perlin Noise generated using libnoise) I'm sure there's probably an easy solution to this but I've been drawing a blank for the past hour. I'm just wondering if the problem is how I'm reading in the data from the noise or if I may be generating the noise wrong? Or maybe the noise is just not good for this kind of generation? If I'm reading it wrong does anyone know the right way? the answers on google were somewhat ambiguous but I'm going to keep searching. Thanks in advance!

  Read the article

• Transparent Technology from Amazon

  - by David Dorf

  Amazon has been making some interesting moves again, this time in the augmented humanity area.  Augmented humanity is about helping humans overcome their shortcomings using technology.  Putting a powerful smartphone in your pocket helps you in many ways like navigating streets, communicating with far off friends, and accessing information.  But the interface for smartphones is somewhat limiting and unnatural, so companies have been looking for ways to make the technology more transparent and therefore easier to use. When Apple helped us drop the stylus, we took a giant leap forward in simplicity.  Using touchscreens with intuitive gestures was part of the iPhone's original appeal.  People don't want to know that technology is there -- they just want the benefits.  So what's the next leap beyond the touchscreen to make smartphones even easier to use? Two natural ways we interact with the world around us is by using sight and voice.  Google and Apple have been using both in their mobile platforms for limited uses cases.  Nobody actually wants to type a text message, so why not just speak it?  Any if you want more information about a book, why not just snap a picture of the cover?  That's much more accurate than trying to key the title and/or author. So what's Amazon been doing?  First, Amazon released a new iPhone app called Flow that allows iPhone users to see information about products in context.  Yes, its an augmented reality app that uses the phone's camera to view products, and overlays data about the products on the screen.  For the most part it requires the barcode to be visible to correctly identify the product, but I believe it can also recognize certain logos as well.  Download the app and try it out but don't expect perfection.  Its good enough to demonstrate the concept, but its far from accurate enough.  (MobileBeat did a pretty good review.)  Extrapolate to the future and we might just have a heads-up display in our eyeglasses. The second interesting area is voice response, for which Siri is getting lots of attention.  Amazon may have purchased a voice recognition company called Yap, although the deal is not confirmed.  But it would make perfect sense, especially with the Kindle Fire in Amazon's lineup. I believe over the next 3-5 years the way in which we interact with smartphones will mature, and they will become more transparent yet more important to our daily lives.  This will, of course, impact the way we shop, making information more readily accessible than it already is.  Amazon seems to be positioning itself to be at the forefront of this trend, so we should be watching them carefully.

  Read the article

• How to implement an email unsubscribe system for a site with many kinds of emails?

  - by Mike Liu

  I'm working on a website that features many different types of emails. Users have accounts, and when logged in they have access to a setting page that they can use to customize what types of emails they receive. However, I'd like to also give users an easy way to unsubscribe directly in the emails they receive. I've looked into list unsubscribe headers as well as creating some type of one click link that would unsubscribe a user from that type of email without requiring login or further action. The later would probably require me to break convention and make changes to the database in response to a GET on the link. However, am I incorrect in thinking that either of these would require me to generate and permanently store a unique identifier in my database for every email I ever send, really complicating email delivery? Without that, I'm not sure how I would be able to uniquely identify a user and a type of email in order to change their email preferences, and this identifier would need to be stored forever as a user could have an email sitting in their inbox for a long time before they decide to act on it. Alternatively, I was considering having a no-login page for managing email preferences. In contrast to above where I would need one of these identifiers for each email, this would only need one identifier per user, with no generation or other action required on sending an email. All of these raise security issues, and they could potentially be used by people to tamper with others' email preferences. This could be mitigated somewhat by ensuring that the identifier is really difficult to guess. For the once per user identifier approach, I was considering generating the identifier by passing a user's ID through some type of encryption algorithm, is this a sound approach? For the per-email identifiers, perhaps I could use a user's ID appended to the time. However, even this would not eliminate the problem entirely, as this would really just be security through obscurity, and anyone with the URL could tamper, and in the end the main defense would have to be that most people aren't so bored as to tamper with other people's email preferences. Are there any other alternatives I've missed, or issues or solutions with these that anyone can provide insight on? What are best practices in this area?

  Read the article

• Why does Chrome video performance substantially degrade after waking from suspend in 10.10?

  - by Grant Heaslip

  Note: For some more details, some of which may not be true given what I've figured out, see this post. When I first boot my computer, video performance (both native H.264 HTML5 in YouTube and Vimeo, and in Flash) in Chrome is perfectly reasonable. CPU usage stays slow, everything works correctly, and the video is silky-smooth. But for whatever reason, if I suspend my computer then wake it up, video performance plummets. Full screen HTML5 video is choppy at best, and full-screen Flash video basically brings my computer to its knees (I'm talking less than a frame a second, and a 5 second lead time to leave full-screen after hitting Esc). Restarting Chrome doesn't fix this — I need to completely restart my machine before performance goes back to normal. Video performance in other applications, such as Movie Player, doesn't seem to be affected at all by the suspend cycle — it's only Chrome. I'm using a Lenovo X201, with an Intel GMA HD graphics chipset, and Intel compnents all around (I don't need any proprietary drivers). This didn't happen in 10.04, and I haven't anything that I think would have caused this to happen. It's possible that a Chrome release could have caused this, but it seems less likely than a regression between 10.04 and 10.10. Any ideas? EDIT: In response Georg's comment, logging in and out doesn't fix it. Restarting Compiz or switching to Metacity (at least by using "compiz/metacity --replace & disown" — am I doing it right?) doesn't help (actually, it seemed to help somewhat with Flash once, but I haven't been able to reproduce this). I'm not sure about GDM — when I use "sudo restart gdm" I get kicked back to the Linux shell (?), which I have no idea how to get out of. Also, I want to make very clear that this isn't just a case of Flash sucking (it does,but that's beside the point). I"m seeing the same general problem with HTML5 videos, and Flash is performing better on my Nexus One than it does on my Core i5 laptop. There's something screwy going on with Chrome and/or 10.10.

  Read the article

• ArchBeat Link-o-Rama for 2012-06-22

  - by Bob Rhubart

  Guide to integration architecture | Stephanie Mann "The landscape of integration architecture is shifting as service-oriented and cloud-based architecture take the fore," says Stephanie Mann. "To ensure success, enterprise architects and developers are turning to lighter-weight infrastructure to support more complex integration projects." FY13 Oracle PartnerNetwork Kickoff - Tues June 26, 2012 Join us for a one-hour live online event hosted by the Oracle PartnerNetwork team as we kickoff FY13. Other dates/times for EMEA/LAD/JAPAN/APAC. Click the link for details. Why should you choose Oracle WebLogic 12c instead of JBoss EAP 6? | Ricardo Ferreira Okay, you would expect an Oracle guy to make this argument. But Ferreira takes a very deep, very detailed technical dive into the issue. So hear the man out, will ya? Hibernate4 and Coherence | Rene van Wijk According to Oracle ACE Rene van Wijk, "there are two ways to integrate Hibernate and Coherence." In this post he illustrates one of them. Simple Made Easy | Rich Hickey Rich Hickey discusses simplicity, why it is important, how to achieve it in design and how to recognize its absence in the tools, language constructs and libraries in this presentation from QCon London 2012. Starting a cluster | Mark Nelson Fusion Middleware A-Team blogger Mark Nelson looks at Oracle SOA Suite, Oracle BPM, and Oracle Coherence, three products that are " commonly clustered, and which have somewhat different requirements." Why building SaaS well means giving up your servers | GigaOM The biggest benefit to PaaS, reports GigaOM's Derrick Harris, "might be a better product because the company is able to focus on building the app rather than managing servers." Personas - what, why & how | Mascha van Oosterhout "To be able to create a successful, user-friendly website or application," says Mascha van Oosterhout, "every decision you take, whether you are part of the marketing team, the design team or the development team, should be based on what you know about the user." Thought for the Day "Machines take me by surprise with great frequency." — Alan Turing(June 23, 1912 - June 7, 1954) Source: Brainy Quote

  Read the article

• Stagnating in programming

  - by Coder

  Time after time this question came up in my mind, but up until today I wasn't thinking about it much. I have been programming for maybe around 8 years now, and for the last two years it seems I'm not as keen to pick up new technologies anymore. Maybe that's a burnout or something, but I'd say it's experience and what I like, that's stopping me from running after the latest and greatest. I'm C++ developer, by this I mean, I love close to metal programming. I have no problems tracing problems through assembly, using tools like WinDbg or HexView. When I use constructs, I think about how they are realized underneath, how the bits are set and unset under the hood. I love battling with complex threading problems and doing everything hardcore way, even by hand if the regular solutions seem half baked. But I also love the C++0x stuff, and use it a lot. And all C++ code as long as it's not cumbersome compared to C counterparts, sometimes I also fall back to sort of "Super C" if the C++ way is ugly. And then there are all other developers who seem to be way more forward looking, .Net 4.0 MVC, WPF, all those Microsoft X#s, LINQ languages, XML and XSLT, mobile devices and so on. I have done a considerable amount of .NET, SQL, ASPX programming, but the further I go, the less I want to try those technologies. Is that bad? Almost every day I hear people saying that managed code is the only way forward, WPF is the way to go. I hear that C++ is godawful, and you can't code anything in it that's somewhat stable. But I don't buy it. With the experience I have, and the knowledge of how native code is compiled and executes, I can say I find it extremely rare that C++ code is unstable, or leaks, or causes crashes that takes more than 30 seconds to identify and fix. And to tell the truth, I've seen enough problems with other "cool" languages that I'd say C++ is even more stable and production proof than the safe languages, at least for me. The only thing that scares me in C++ is new frameworks, I don't trust them, and I use them extra sparingly. STL - yes, ATL - very sparingly, everything else... Well, not very keen on it. Most huge problems I've ran into, all were related to frameworks, not the language itself. Some overrided operator here, bad hierarchy there, poor class design here, mystical castings there. Other than that, C/C++ (yes, I use them together) still seems a very controlled and stable way to develop applications. Am I stagnating? Should I switch a profession, or force myself in all that marketing hype? Are there more developers who feel the same way?

  Read the article

• A Community Cure for a String Splitting Headache

  - by Tony Davis

  A heartwarming tale of dogged perseverance and Community collaboration to solve some SQL Server string-related headaches. Michael J Swart posted a blog this week that had me smiling in recognition and agreement, describing how an inquisitive Developer or DBA deals with a problem. It's a three-step process, starting with discomfort and anxiety; a feeling that one doesn't know as much about one's chosen specialized subject as previously thought. It progresses through a phase of intense research and learning until finally one achieves breakthrough, blessed relief and renewed optimism. In this case, the discomfort was provoked by the mystery of massively high CPU when searching Unicode strings in SQL Server. Michael explored the problem via Stack Overflow, Google and Twitter #sqlhelp, finally leading to resolution and a blog post that shared what he learned. Perfect; except that sometimes you have to be prepared to share what you've learned so far, while still mired in the phase of nagging discomfort. A good recent example of this recently can be found on our own blogs. Despite being a loud advocate of the lightning fast T-SQL-based string splitting techniques, honed to near perfection over many years by Jeff Moden and others, Phil Factor retained a dogged conviction that, in theory, shredding element-based XML using XQuery ought to be even more efficient for splitting a string to create a table. After some careful testing, he found instead that the XML way performed and scaled miserably by comparison. Somewhat subdued, and with a nagging feeling that perhaps he was still missing "something", he posted his findings. What happened next was a joy to behold; the community jumped in to suggest subtle changes in approach, using an attribute-based rather than element-based XML list, and tweaking the XQuery shredding. The result was performance and scalability that surpassed all other techniques. I asked Phil how quickly he would have arrived at the real breakthrough on his own. His candid answer was "never". Both are great examples of the power of Community learning and the latter in particular the importance of being brave enough to parade one's ignorance. Perhaps Jeff Moden will accept the string-splitting gauntlet one more time. To quote the great man: you've just got to love this community! If you've an interesting tale to tell about being helped to a significant breakthrough for a problem by the community, I'd love to hear about it. Cheers, Tony.

  Read the article

• Attunity Oracle CDC Solution for SSIS - Beta

  We in no way work for Attunity but we were asked to test drive a beta version of their Oracle CDC solution for SSIS.  Everybody should know that moving more data than you need to takes too much time and uses resources that may better be employed doing something else.  Change data Capture is a technology that is designed to help you identify only the data that has had something done to it and you can therefore move only what is needed.  Microsoft have implemented this exact functionality into SQL server 2008 and I really like it there.  Attunity though are doing it on Oracle. DISCLAIMER: This is a BETA release and some of the parts are a bit ugly/difficult to work with.  The idea though is definitely right and the product once working does exactly what it says on the tin.  They have always been helpful to me when I have had a problem with the product and if that continues then beta testing pain should be eased somewhat. In due course I am going to be doing some videos around me using the product.  If you use Oracle and SSIS then give it a go. Here is their product description.   Attunity is a Microsoft SQL Server technology partner and the creator of the Microsoft Connectors for Oracle and Teradata, currently available in SQL Server 2008 Enterprise Edition. Attunity released a beta version of the Attunity Oracle-CDC for SSIS, a product that integrates continually changing Oracle data into SSIS, efficiently and in real-time. Attunity designed the product and integrated it into SSIS to create the simple creation of change data capture (CDC) solutions, accelerate implementation time, and reduce resources and costs. They also utilize log-based CDC so the solution has minimal impact on the Oracle source system. You can use the product to implement enterprise-class data replication, synchronization, and real-time business intelligence (BI) and data warehousing projects, quickly and efficiently, leveraging their existing SQL Server investments and resource skills. Attunity architected the product specifically for the Microsoft SSIS developer community and the product is available for both SQL Server 2005 and SQL Server 2008. It offers the following key capabilities: · Log-based, non-intrusive Oracle CDC · Full integration into SSIS and the Business Intelligence Developer Studio · Automatic generation of SSIS packages for CDC as well as full-loads of Oracle data · Filtering of Oracle tables and columns at the source · Monitoring and control of CDC processing Click to learn more and download the beta.

  Read the article

• Configurable Objects - Introduction

  - by Anthony Shorten

  One of the interesting facilities in the framework is Configurable Object functionality (it is also known as Task Optimization and also known as Cool Tools). The idea is that any implementation can create their own views of the base product objects and services and implement functionality against those new views. For example, in Oracle Utilities Customer Care and Billing, there is a Person object. That object is used to store and manage information about individuals as well as companies. In the base product you would use the Person Maintenance screen and fill in some of the screen when you wanted to register or maintain and individual as well and fill out other parts of the screen when you wanted to register or maintain a company. This can be somewhat confusing to some customers. Using Configurable Objects this can be simplified. A business object can be created that is a view of the any object. For example, you could create a Human business object which would cover the aspects of the Person object pertaining to an individual and a Company business object to cover the aspects unique to a company. Even the tag names (i.e. Field Names) in the object can be changed to be more what the implementation is familiar with. The object can also restructure the object. For example, a common identifier for an individual in the USA is the Social Security number, this value is a Person Identifier (as this varies in each country). In the new Human object you can remap the Person Identifier as a Social Security number. To define a Business Object you use a schema editor built into the browser user interface and use a mapping language to setup the business objects. An example of the language is shown below in an extract of the schema for the Human business object. As you can see there are mapping as well as formatting and other tags. This information can be built manually or using a wizard which generates the base structure for you to alter. This is all stored as meta data when saved. Once a Business object is built it can be used as basis for code, other business objects (we support inheritance), called by a screen (called a UI Map) or even as a Web Service. This is just a start with Configurable Objects as you can also create views of base services called Business Services, Service Scripts used for non-object or complex object processing (as well as other things), UI Maps used for screens and Data Areas to reuse definitions across multiple objects. Configurable Objects are powerful and I only really touched on them here. Over the next few months I hope to add lots more entries about them.

  Read the article

• Handy SQL Server Function Series: Part 1

  - by Most Valuable Yak (Rob Volk)

  I've been preparing to give a presentation on SQL Server for a while now, and a topic that was recommended was SQL Server functions.  More specifically, the lesser-known functions (like @@OPTIONS), and maybe some interesting ways to use well-known functions (like using PARSENAME to split IP addresses)  I think this is a veritable goldmine of useful information, and researching for the presentation has confirmed that beyond my initial expectations.I even found a few undocumented/underdocumented functions, so for the first official article in this series I thought I'd start with 2 of each, COLLATIONPROPERTY() and COLLATIONPROPERTYFROMID().COLLATIONPROPERTY() provides information about (wait for it) collations, SQL Server's method for handling foreign character sets, sort orders, and case- or accent-sensitivity when sorting character data.  The Books Online entry for  COLLATIONPROPERTY() lists 4 options for code page, locale ID, comparison style and version.  Used in conjunction with fn_helpcollations():SELECT *, COLLATIONPROPERTY(name,'LCID') LCID, COLLATIONPROPERTY(name,'CodePage') CodePage, COLLATIONPROPERTY(name,'ComparisonStyle') ComparisonStyle, COLLATIONPROPERTY(name,'Version') Version FROM fn_helpcollations()You can get some excellent information. (c'mon, be honest, did you even know about fn_helpcollations?)Collations in SQL Server have a unique name and ID, and you'll see one or both in various system tables or views like syscolumns, sys.columns, and INFORMATION_SCHEMA.COLUMNS.  Unfortunately they only link the ID and name for collations of existing columns, so if you wanted to know the collation ID of Albanian_CI_AI_WS, you'd have to declare a column with that collation and query the system table.While poking around the OBJECT_DEFINITION() of sys.columns I found a reference to COLLATIONPROPERTYFROMID(), and the unknown property "Name".  Not surprisingly, this is how sys.columns finds the name of the collation, based on the ID stored in the system tables.  (Check yourself if you don't believe me)Somewhat surprisingly, the "Name" property also works for COLLATIONPROPERTY(), although you'd already know the name at that point.  Some wild guesses and tests revealed that "CollationID" is also a valid property for both functions, so now:SELECT *, COLLATIONPROPERTY(name,'LCID') LCID, COLLATIONPROPERTY(name,'CodePage') CodePage, COLLATIONPROPERTY(name,'ComparisonStyle') ComparisonStyle, COLLATIONPROPERTY(name,'Version') Version, COLLATIONPROPERTY(name,'CollationID') CollationID FROM fn_helpcollations() Will get you the collation ID-name link you…probably didn't know or care about, but if you ever get on Jeopardy! and this question comes up, feel free to send some of your winnings my way. :)And last but not least, COLLATIONPROPERTYFROMID() uses the same properties as COLLATIONPROPERTY(), so you can use either one depending on which value you have available.Keep an eye out for Part 2!

  Read the article

• Networking Guidelines

  - by ACShorten

  One of the things I have noticed in my years in IT is the changes in networking. In the past networking was pretty simple with the host name and name resolution (via DNS) being pretty simple. Some sites still use this simple networking setup. These days, more complex name resolution, proxies, firewalls, demarcation nd virtualization, can make networking more complex. This can cause issues when installing products with in built networking that can frustrate even seasoned veterans. I have put together a few basic guidelines to hopefully help along with product installation and getting a product to operate in a somewhat complex network setup. All the components of the product (including the infrastructure) need to communicate via a network (even it is within a local machine/host). Ensure any host names referred to within configuration files are accessible via your networking setup. This may mean defining the hosts to the machines, to the DNS for name resolution and even your firewall to allow machines to communicate within your network. Make sure the ports used for any of the infrastructure are accessible (even through your firewall) and are unique within the host. Host duplication can cause the product to fail on startup as the port is already in use. If there are still issues, consider using localhost as your host name. I have used this in so many situations that I tend to use it now as a default anytime I install anything myself. Most Oracle products suggest to use localhost when using dynamic host or dynamic IP addresses and this is no different for the Oracle Utilities Application Framework. If you do use localhost then installing a Loopback Adapter for the operating system is recommended to force networking to a minimum. Usually localhost resolves to 127.0.0.1. When using multiple network connections, especially in a virtualized environment, ensure the host and ports used are relevent for the network cards you have setup. One of the common issues is finding the product is using a vierualized network card only to find that it is not setup for correct networking. If you are using the batch component, do not forget to ensure that the multicast protocol is enabled on your host and that the multicast address and port number specified are valid and accessible from all machines in the batch cluster (if clustering used). The same advice applies if you are using unicast where each host/port combination should be accessible. Hopefully these basic networking recommendations will help minimize any networking issues you might encounter.

  Read the article

• Rights and use of developed software

  - by Nils Munch

  I have been working on a piece of software for a company, that they wish to resell. There was an mail-based agreement upon a flat hourly rate for my work, and eager me chose to accept a rather low fee. Due to the stress and tempo of the task, a direct contract was never formed or signed. The software was developed locally on my machine, and I was pretty much alone with it, except by excellent help from StackOverflow when I got stuck. Now, the software is nearing completion, I suddenly hear that they have hired a new developer to make the same piece of software as me, and that I was expected to resign within long. Confused I ask around, and realize that the CEO of the company had informed the rest of the company that I was terminally ill and had cancer, and was expected to leave the company soon. Since I'm perfectly healthy, this confused me even more, until I realized what was going on. When I confronted my boss with this, I was no longer seen as a member of the company, and I left the same day, never to return. Later, I raised the question about my missing pay, since I had been working for quite a bit, and not received any payment for my software. I saw that they had already sold a fair copy of my software, and since it's not exactly sold cheap, the company should have plenty of gold to pay me. The company refused, and said that they owned the software, and everything it contained. That was a lot of drama, but my question is this: Who has the rights to the software ? The source code had my personal watermarks and copyrights inprinted, but they have since simply deleted it. The company claim that they have all the rights, because they have a website made about the product, where they write that they have "All rights reserved" in the bottom. My instinct tells me that if a company buys a service like this, and then refuses to pay their developer, then they should not be allowed to keep, and much less resell the product. I have not signed any agreements about giving the company the use of this product, I have made it in my own time and without help from the rest of the company. This all takes place in Denmark, Europe, but I would guess that the rules about this is somewhat universal. Im not the strongest person to legal-talk, so I might be wrong.

  Read the article

• Use CSS Selectors with HtmlUnit

  - by kerry

  HtmlUnit is a great library for performing web integration tests in Java.  But sometimes node traversal can be somewhat cumbersome. Fear not fellow automated tester (good for you!).  I found a great little project on Github that will allow you to query your document for elements via css selectors similar to jQuery. The project is located at https://github.com/chrsan/css-selectors.  You can use Maven to build it, or download 1.0.2 here.  Beware.  I will not be updating this link so I suggest you download the latest code. In any case, you can use it like so: // from HtmlUnit getting started final WebClient webClient = new WebClient(); final HtmlPage page = webClient.getPage("http://htmlunit.sourceforge.net"); final DOMNodeSelector cssSelector = new DOMNodeSelector(page.getDocumentElement()); final Set elements = cssSelector.querySelectorAll("div.section h2"); final Node first = elements.iterator().next(); assertThat(first.getTextContent(), equalTo("HtmlUnit")); The only problem here is that the querySelectAll returns a Set<Node>.  Not HtmlElement like we may want in some cases.   However, if you were to reflect on the Set, you would find that it is indeed a Set of HtmlElement objects. Typically, I like to create a base class for my web tests.  Just for fun, I am using the $ method similar to jQuery. public class WebTestBase { protected WebClient webClient; protected HtmlPage htmlPage; protected void goTo(final String url){ return (HtmlPage)webClient.getPage(url); } protected List $(final String cssSelector) { final DOMNodeSelector cssSelector = new DOMNodeSelector(htmlPage.getDocumentElement()); final Set nodes = cssSelector.querySelectorAll("div.section h2"); // for some reason Set cannot be cast to Set? final List elements = new ArrayList(nodes.size()); for (final Node node : nodes) { elements.add((HtmlElement)node); } return elements; } } Now we can write tests like this: public class LoginWebTest extends WebTestBase { @Test public void login_page_has_instructions() throws Exception { goTo(baseUrl + "/login") assertThat( $("p.instructions").size(), equalTo(1) ); } }

  Read the article

• WPF: Reloading app parts to handle persistence as well as memory management.

  - by Ingó Vals

  I created a app using Microsoft's WPF. It mostly handles data reading and input as well as associating relations between data within specific parameters. As a total beginner I made some bad design decision ( not so much decisions as using the first thing I got to work ) but now understanding WPF better I'm getting the urge to refactor my code with better design principles. I had several problems but I guess each deserves it's own question for clarity. Here I'm asking for proper ways to handle the data itself. In the original I wrapped each row in a object when fetched from database ( using LINQ to SQL ) somewhat like Active Record just not active or persistence (each app instance had it's own data handling part). The app has subunits handling different aspects. However as it was setup it loaded everything when started. This creates several problems, for example often it wouldn't be neccesary to load a part unless we were specifically going to work with that part so I wan't some form of lazy loading. Also there was problem with inner persistance because you might create a new object/row in one aspect and perhaps set relation between it and different object but the new object wouldn't appear until the program was restarted. Persistance between instances of the app won't be huge problem because of the small amount of people using the program. While I could solve this now using dirty tricks I would rather refactor the program and do it elegantly, Now the question is how. I know there are several ways and a few come to mind: 1) Each aspect of the program is it's own UserControl that get's reloaded/instanced everytime you navigate to it. This ensures you only load up the data you need and you get some persistancy. DB server located on same LAN and tables are small so that shouldn't be a big problem. Minor drawback is that you would have to remember the state of each aspect so you wouldn't always start at beginners square. 2) Having a ViewModel type object at the base level of the app with lazy loading and some kind of timeout. I would then propegate this object down the visual tree to ensure every aspect is getting it's data from the same instance 3) Semi active record data layer with static load methods. 4) Some other idea What in your opinion is the most practical way in WPF, what does MVVM assume?

  Read the article

• Organizing Git repositories with common nested sub-modules

  - by André Caron

  I'm a big fan of Git sub-modules. I like to be able to track a dependency along with its version, so that you can roll-back to a previous version of your project and have the corresponding version of the dependency to build safely and cleanly. Moreover, it's easier to release our libraries as open source projects as the history for libraries is separate from that of the applications that depend on them (and which are not going to be open sourced). I'm setting up workflow for multiple projects at work, and I was wondering how it would be if we took this approach a bit of an extreme instead of having a single monolithic project. I quickly realized there is a potential can of worms in really using sub-modules. Supposing a pair of applications: studio and player, and dependent libraries core, graph and network, where dependencies are as follows: core is standalone graph depends on core (sub-module at ./libs/core) network depdends on core (sub-module at ./libs/core) studio depends on graph and network (sub-modules at ./libs/graph and ./libs/network) player depends on graph and network (sub-modules at ./libs/graph and ./libs/network) Suppose that we're using CMake and that each of these projects has unit tests and all the works. Each project (including studio and player) must be able to be compiled standalone to perform code metrics, unit testing, etc. The thing is, a recursive git submodule fetch, then you get the following directory structure: studio/ studio/libs/ (sub-module depth: 1) studio/libs/graph/ studio/libs/graph/libs/ (sub-module depth: 2) studio/libs/graph/libs/core/ studio/libs/network/ studio/libs/network/libs/ (sub-module depth: 2) studio/libs/network/libs/core/ Notice that core is cloned twice in the studio project. Aside from this wasting disk space, I have a build system problem because I'm building core twice and I potentially get two different versions of core. Question How do I organize sub-modules so that I get the versioned dependency and standalone build without getting multiple copies of common nested sub-modules? Possible solution If the the library dependency is somewhat of a suggestion (i.e. in a "known to work with version X" or "only version X is officially supported" fashion) and potential dependent applications or libraries are responsible for building with whatever version they like, then I could imagine the following scenario: Have the build system for graph and network tell them where to find core (e.g. via a compiler include path). Define two build targets, "standalone" and "dependency", where "standalone" is based on "dependency" and adds the include path to point to the local core sub-module. Introduce an extra dependency: studio on core. Then, studio builds core, sets the include path to its own copy of the core sub-module, then builds graph and network in "dependency" mode. The resulting folder structure looks like: studio/ studio/libs/ (sub-module depth: 1) studio/libs/core/ studio/libs/graph/ studio/libs/graph/libs/ (empty folder, sub-modules not fetched) studio/libs/network/ studio/libs/network/libs/ (empty folder, sub-modules not fetched) However, this requires some build system magic (I'm pretty confident this can be done with CMake) and a bit of manual work on the part of version updates (updating graph might also require updating core and network to get a compatible version of core in all projects). Any thoughts on this?

  Read the article

• Adjusting the Score on Oracle Text search results

  - by Kyle Hatlestad

  When you sort the results of a search by Score using OracleTextSearch as the search engine in WebCenter Content, the results coming back are based on the relevancy on the document.  In theory, the more relevant the search term is to the document, the higher ranked Score it should receive.  But in practice, the relevancy score can seem somewhat of a mystery.  It's not entirely clear how it ranks the importance of some documents over others based on the search term.  And often times, once a word appears a certain number of times within a document, the Score simply maxes out at 100 and the top results can be difficult to discern from one another.  Take for example the search for 'vacation' on this set of documents: Out of 7 results, 6 of them have a Score of '100' which means they are basically ranked the same.  This doesn't make the sort by Score very meaningful.   Besides sorting by relevance, you can also tell Oracle Text to sort by occurrence.  In that case, it is a much more predictable result in how they would be ranked. And for many cases provide a more meaningful sorting of results then relevance. To change this takes a small component change to the SearchOperatorMap resource.  By default, the query used for full-text searching looks like: <td>(ORACLETEXTSEARCH)fullText</td> <td>DEFINESCORE((%V), RELEVANCE * .1)</td> <td>text</td> Overriding this resource and changing it to: <td>(ORACLETEXTSEARCH)fullText</td> <td>DEFINESCORE((%V), OCCURRENCE * .01)</td>  <td>text</td> will force it to now use occurrence (note the change in scale to .01 as well).  So running the same search and sort options as the example above, the results come out quite a bit differently: In this case, there is a clear understanding of how the items rank.   And generally, if the search term appears 3 times more in one document then another, it's got a better chance of being a document I'm interested in.  You may or may not feel the relevance ranking is better then the search term occurrence, but this provides the opportunity to try an alternate method that might work better for your results.  A pre-built component is available for download here. There is one caveat in using this method.  The occurrence ranking also maxes out at 100, so if a search term is in the document more then that, the Score result will stay at 100.

  Read the article

• GNU/Linux interactive table content GUI editor?

  - by sdaau

  I often find myself in the need to gather data (say from the internet), into a table, for comparison reasons. I usually need the final table output in HTML or MediaWiki mostly, but often times also Latex. My biggest problem is that I often forget the correct table syntax for these markup languages, as well as what needs to be properly escaped in the inline data, for the table to render correctly. So, I often wish there was a GUI application, which provides a tabular framework - which I could stick "Always on Top" as a desktop window, and I could paste content into specific cells - before finally exporting the table as a code in the correct language. One application that partially allows this is Open/LibreOffice calc: The good thing here is that: I can drag and drop browser content into a specifically targeted table cell (here B2) "Rich" text / HTML code gets pasted For long content, the cell (column) width stays put as it originally was The bad thing is, that: when the cell height (due to content size) becomes larger than the calc window, it becomes nearly impossible to scroll calc contents up and down (at least with the mousewheel), as the view gets reset to top-right corner of the selected cell calc shows an "endless"/unlimited field of cells, so not exactly a "table" - which I find visually very confusing (and cognitively taxing) Can only export table to HTML What I would need is an application that: Allows for a limited size table, but with quick adding of rows and columns (e.g. via corresponding + buttons) Allows for quick setup of row and column height and width (as well as table size) Stays put at those sizes, regardless of size of content pasted in; if cell content overflows, cell scrollbars are shown (cell content could be possibly re-edited in a separate/new window); if table overflows over window size, window scrollbars are shown Exports table in multiple formats (I'd need both HTML and mediawiki), properly escaping cell content for each (possibility to strip HTML tags from content pasted in cells, to get plain text, is a plus) Targeting a specific cell in the table for the content paste operation is a must - it doesn't have to be drag'n'drop though, a right click over a cell with "Paste content" is enough. I'd also want the ability to click in a specific cell and type in (plain text) content immediately. So, my question is: is there an application out there that already does something like this? The reason I'm asking is that - as the screenshots show - for instance Libre/OpenOffice allows it, but only somewhat (as using it for that purpose is tedious). I know there exist some GUI editors for Linux (both for UI like guile or HTML like amaya); but I don't know them enough to pinpoint if any of them would offer this kind of functionality (and at least in my searches, that kind of functionality, if present in diverse software, seems not to be advertised). Note I'm not interested in styling an HTML table, which is why I haven't used "table designer" in the title, but "table editor" (in lack of better terms) - I'm interested in (quickly) adjusting row/column size of the table, and populating it with pasted data (which is possibly HTML) in a GUI; and finally exporting such a table as self-contained HTML (or other) code.

  Read the article

• Northwind now available on SQL Azure

  - by jamiet

  Two weeks ago I made available a copy of [AdventureWorks2012] on SQL Azure and published credentials so that anyone from the SQL community could connect up and experience SQL Azure, probably for the first time. One of the (somewhat) popular requests thereafter was to make the venerable Northwind database available too so I am pleased to say that as of right now, Northwind is up there too. You will notice immediately that all of the Northwind tables (and the stored procedures and views too) have been moved into a schema called [Northwind] – this was so that they could be easily differentiated from the existing [AdventureWorks2012] objects. I used an SQL Server Data Tools (SSDT) project to publish the schema and data up to this SQL Azure database; if you are at all interested in poking around that SSDT project then I have made it available on Codeplex for your convenience under the MS-PL license – go and get it from https://northwindssdt.codeplex.com/. Using SSDT proved particularly useful as it alerted me to some aspects of Northwind that were not compatible with SQL Azure, namely that five of the tables did not have clustered indexes: The beauty of using SSDT is that I am alerted to these issues before I even attempt a connection to SQL Azure. Pretty cool, no? Fixing this situation was of course very easy, I simply changed the following primary keys from being nonclustered to clustered: [PK_Region] [PK_CustomerDemographics] [PK_EmployeeTerritories] [PK_Territories] [PK_CustomerCustomerDemo]   If you want to connect up then here are the credentials that you will need: Server mhknbn2kdz.database.windows.net Database AdventureWorks2012 User sqlfamily Password sqlf@m1ly You will need SQL Server Management Studio (SSMS) 2008R2 installed in order to connect or alternatively simply use this handy website: https://mhknbn2kdz.database.windows.net which provides a web interface to a SQL Azure server. Do remember that hosting this database is not free so if you find that you are making use of it please help to keep it available by visiting Paypal and donating any amount at all to [email protected]. To make this easy you can simply hit this link and the details will be completed for you – all you have to do is login and hit the “Send” button. If you are already a PayPal member then it should take you all of about 20 seconds! I hope this is useful to some of you folks out there. Don’t forget that we also have more data up there than in the conventional [AdventureWorks2012], read more at Big AdventureWorks2012. @Jamiet  AdventureWorks on Azure - Provided by the SQL Server community, for the SQL Server community!

  Read the article

• Liskov principle: violation by type-hinting

  - by Elias Van Ootegem

  According to the Liskov principle, a construction like the one below is invalid, as it strengthens a pre-condition. I know the example is pointless/nonsense, but when I last asked a question like this, and used a more elaborate code sample, it seemed to distract people too much from the actual question. //Data models abstract class Argument { protected $value = null; public function getValue() { return $this->value; } abstract public function setValue($val); } class Numeric extends Argument { public function setValue($val) { $this->value = $val + 0;//coerce to number return $this; } } //used here: abstract class Output { public function printValue(Argument $arg) { echo $this->format($arg); return $this; } abstract public function format(Argument $arg); } class OutputNumeric extends Output { public function format(Numeric $arg)//<-- VIOLATION! { $format = is_float($arg->getValue()) ? '%.3f' : '%d'; return sprintf($format, $arg->getValue()); } } My question is this: Why would this kind of "violation" be considered harmful? So much so that some languages, like the one I used in this example (PHP), don't even allow this? I'm not allowed to strengthen the type-hint of an abstract method but, by overriding the printValue method, I am allowed to write: class OutputNumeric extends Output { final public function printValue(Numeric $arg) { echo $this->format($arg); } public function format(Argument $arg) { $format = is_float($arg->getValue()) ? '%.3f' : '%d'; return sprintf($format, $arg->getValue()); } } But this would imply repeating myself for each and every child of Output, and makes my objects harder to reuse. I understand why the Liskov principle exists, don't get me wrong, but I find it somewhat difficult to fathom why the signature of an abstract method in an abstract class has to be adhered to so much stricter than a non-abstract method. Could someone explain to me why I'm not allowed to hind at a child class, in a child class? The way I see it, the child class OutputNumeric is a specific use-case of Output, and thus might need a specific instance of Argument, namely Numeric. Is it really so wrong of me to write code like this?

  Read the article

< Previous Page | 46 47 48 49 50 51 52 53 54 55 56 57  | Next Page >