Search Results

Search found 20347 results on 814 pages for 'control adapters'.

Page 58/814 | < Previous Page | 54 55 56 57 58 59 60 61 62 63 64 65  | Next Page >

  • How can I adjust system volume from within XBMC?

    - by d3vid
    While running XBMC, I can adjust the volume of the XBMC application itself. However, this volume is limited by the current system volume. For example, if the system volume is at 80% and XBMC is at 100%, I am effectively at 80% and cannot go higher. Or if the sound is too soft and needs a boost, I would normally increase the system volume beyond 100%. XBMC takes over the whole screen, so the system volume is not accessible. Pressing the Super key brings up the dash and top menu, but clicking on it is difficult and inconsistent, and very quickly XBMC takes over the screen again. How can I adjust the system volume without having to quit XBMC?

    Read the article

  • Github Organization Repositories, Issues, Multiple Developers, and Forking - Best Workflow Practices

    - by Jim Rubenstein
    A weird title, yes, but I've got a bit of ground to cover I think. We have an organization account on github with private repositories. We want to use github's native issues/pull-requests features (pull requests are basically exactly what we want as far as code reviews and feature discussions). We found the tool hub by defunkt which has a cool little feature of being able to convert an existing issue to a pull request, and automatically associate your current branch with it. I'm wondering if it is best practice to have each developer in the organization fork the organization's repository to do their feature work/bug fixes/etc. This seems like a pretty solid work flow (as, it's basically what every open source project on github does) but we want to be sure that we can track issues and pull requests from ONE source, the organization's repository. So I have a few questions: Is a fork-per-developer approach appropriate in this case? It seems like it could be a little overkill. I'm not sure that we need a fork for every developer, unless we introduce developers who don't have direct push access and need all their code reviewed. In which case, we would want to institute a policy like that, for those developers only. So, which is better? All developers in a single repository, or a fork for everyone? Does anyone have experience with the hub tool, specifically the pull-request feature? If we do a fork-per-developer (or even for less-privileged devs) will the pull-request feature of hub operate on the pull requests from the upstream master repository (the organization's repository?) or does it have different behavior? EDIT I did some testing with issues, forks, and pull requests and found that. If you create an issue on your organization's repository, then fork the repository from your organization to your own github account, do some changes, merge to your fork's master branch. When you try to run hub -i <issue #> you get an error, User is not authorized to modify the issue. So, apparently that work flow won't work.

    Read the article

  • How do I identify mouse clicks versus mouse down in games?

    - by Tristan
    What is the most common way of handling mouse clicks in games? Given that all you have in way of detecting input from the mouse is whether a button is up or down. I currently just rely on the mouse down being a click, but this simple approach limits me greatly in what I want to achieve. For example I have some code that should only be run once on a mouse click, but using mouse down as a mouse click can cause the code to run more then once depending on how long the button is held down for. So I need to do it on a click! But what is the best way to handle a click? Is a click when the mouse goes from mouse up to down or from down to up or is it a click if the button was down for less then x frames/milliseconds and then if so, is it considered mouse down and a click if its down for x frames/milliseconds or a click then mouse down? I can see that each of the approaches can have their uses but which is the most common in games? And maybe i'll ask more specifically which is the most common in RTS games?

    Read the article

  • Simultaneous remote connections

    - by user3215
    Is there any package for Ubuntu which allows me to connect to remote computer without disturbing or terminating others who have already connected to that server or who are accessing the computer directly?. I remember I did this in Windows Server 2003 installing Terminal Server when I was doing my course which allowed multiple clients to connect to the server simultaneously with no disturbance to clients and even direct access to server. Is there any package like this for Ubuntu desktops?

    Read the article

  • What is an effective git process for managing our central code library?

    - by Mathew Byrne
    Quick background: we're a small web agency (3-6 developers at any one time) developing small to medium sized Symfony 1.4 sites. We've used git for a year now, but most of our developers have preferred Subversion and aren't used to a distributed model. For the past 6 months we've put a lot of development time into a central Symfony plugin that powers our custom CMS. This plugin includes a number of features, helpers, base classes etc. that we use to build custom functionality. This plugin is stored in git, but branches wildly as the plugin is used in various products and is pulled from/pushed to constantly. The repository is usually used as a submodule within a major project. The problems we're starting to see now are a large number of Merge conflicts and backwards incompatible changes brought into the repository by developers adding custom functionality in the context of their own project. I've read Vincent Driessen's excellent git branching model and successfully used it for projects in the past, but it doesn't seem to quite apply well to our particular situation; we have a number of projects concurrently using the same core plugin while developing new features for it. What we need is a strategy that provides the following: A methodology for developing major features within the code repository. A way of migrating those features into other projects. A way of versioning the core repository, and of tracking which version each major project uses. A plan for migrating bug fixes back to older versions. A cleaner history that's easier to see where changes have come from. Any suggestions or discussion would be greatly appreciated.

    Read the article

  • How and where do you store your private work/sourcecode?

    - by Amir Rezaei
    I have worked as a developer for over 10 years now. During that time I have had my own small projects where I have developed tools, applications and games. I have not found any robust solution to store my work. It’s always fun to get back to your code and see how you did before and how you would do it now. It’s just work that is unfortunate to lose. There are SVN solution such as Google’s Project Hosting. However I’m not interested in sharing my code or making it open source. Currently I’m hosting my own SVN server. So here comes my question: How and where do you store your private work/sourcecode? Requirements: Sourcecode versioning Backup Prefers free Edit: Remote access Edit: I have used Dropbox + TrueCrypt + SVN. Unfortunately you are limited to 5gb.

    Read the article

  • Rendering an image from an embedded Web Browser (C# WPF application)

    - by The Official Microsoft IIS Site
    How is all started So this week I was working on an extension for WebMatrix , Luke Sampson of http://StudioStyle.es just integrate a cool piece of code from Matt MCElheny . The news is that the studiostyle.es website now supports converting the over 1,000 themes uploaded for Visual Studio 2010 into the WebMatrix format, and hence we automatically got a very large load of themes to choose from. Still we aspired for an even better experience, currently the WebMatrix user will have to install the ColorThemeEditor...(read more)

    Read the article

  • how to integrate plastic scm with jira? [closed]

    - by bilal fazlani
    I am trying to migrate from VSS to Plastic SCM and want to use it with JIRA. I have reached this far. http://i.stack.imgur.com/h1wSw.png I tried referring to their help documentation. but that did not help. Does someone know how to link a new branch to an issue in JIRA ? I tried to giving same name to Issue and Branch. That din't work. If the Issue key is : "DEMO-7", what should be the "Branch Prefix" & "Branch Name" in Plastic SCM ? I am sure I am missing something.

    Read the article

  • How can I best manage making open source code releases from my company's confidential research code?

    - by DeveloperDon
    My company (let's call them Acme Technology) has a library of approximately one thousand source files that originally came from its Acme Labs research group, incubated in a development group for a couple years, and has more recently been provided to a handful of customers under non-disclosure. Acme is getting ready to release perhaps 75% of the code to the open source community. The other 25% would be released later, but for now, is either not ready for customer use or contains code related to future innovations they need to keep out of the hands of competitors. The code is presently formatted with #ifdefs that permit the same code base to work with the pre-production platforms that will be available to university researchers and a much wider range of commercial customers once it goes to open source, while at the same time being available for experimentation and prototyping and forward compatibility testing with the future platform. Keeping a single code base is considered essential for the economics (and sanity) of my group who would have a tough time maintaining two copies in parallel. Files in our current base look something like this: > // Copyright 2012 (C) Acme Technology, All Rights Reserved. > // Very large, often varied and restrictive copyright license in English and French, > // sometimes also embedded in make files and shell scripts with varied > // comment styles. > > > ... Usual header stuff... > > void initTechnologyLibrary() { > nuiInterface(on); > #ifdef UNDER_RESEARCH > holographicVisualization(on); > #endif > } And we would like to convert them to something like: > // GPL Copyright (C) Acme Technology Labs 2012, Some rights reserved. > // Acme appreciates your interest in its technology, please contact [email protected] > // for technical support, and www.acme.com/emergingTech for updates and RSS feed. > > ... Usual header stuff... > > void initTechnologyLibrary() { > nuiInterface(on); > } Is there a tool, parse library, or popular script that can replace the copyright and strip out not just #ifdefs, but variations like #if defined(UNDER_RESEARCH), etc.? The code is presently in Git and would likely be hosted somewhere that uses Git. Would there be a way to safely link repositories together so we can efficiently reintegrate our improvements with the open source versions? Advice about other pitfalls is welcome.

    Read the article

  • HG: fork web app project to separate API code from app code

    - by cs_brandt
    I have a web app thats been in active development for about 8 months now and its becoming apparent that the project has a need to maintain a separation between app specific code and our OO Javascript API. What I would like to do is have another repository with the following general structure of the js API code. repo_name | +---build | +---build_tools | +---doc | +---src | +---js Of course this structure is different from the original web app directory structure. If I make changes to this new repository how could I pull in those changes to the web app repository without unintentionally removing files or modifying the directory structure of the web app repository?

    Read the article

  • When to delete a branch in Git

    - by Jo-Herman Haugholt
    I have a script project I've been managing with Git. Besides two main branches, several minor branches have been introduced over time to cover minor features, tweaks or temporary changes. Some of these branches are nearing end-of-life, and I won't be updating them any more. What's the different philosophies for handling branches like this? Should they be removed, or left in the repository unmaintained? If I do, won't I end up with a cluttered repository?

    Read the article

  • Combining Code Review with Trust Metrics

    - by DragonFax
    I don't get the chance to partake of it at work. But I love the idea of code review. Especially of online open source code review like Gerrit Code Review. I love what Trust Metrics have done for forums and collective intelligences sites on the internet like stackexchange, reddit, and wikipedia. Would it be possible to combine the two and come up with an open source project management system. Something that ends up being mostly community driven. Perhaps a kind of wikipedia of code for a project. Where submitters become popular/trusted by having lots of patches reviewed favoriably by others, and accepted into the trunk. And popular/trusted submitters get their patchs accepted faster/easier. I'm looking for some opinions on the idea, or perhaps pointers to where its been done before, if thats the case. This might leave the lead maintiner little more to do than: wrangle the direction of the project by fast-tracking or vetoing specific patches. settling disputes when the CI tests break, or fixing it himself. Is design by community worse than design by committee?

    Read the article

  • How do I structure code and builds for continuous delivery of multiple applications in a small team?

    - by kingdango
    Background: 3-5 developers supporting (and building new) internal applications for a non-software company. We use TFS although I don't think that matters much for my question. I want to be able to develop a deployment pipeline and adopt continuous integration / deployment techniques. Here's what our source tree looks like right now. We use a single TFS Team Project. $/MAIN/src/ $/MAIN/src/ApplicationA/VSSOlution.sln $/MAIN/src/ApplicationA/ApplicationAProject1.csproj $/MAIN/src/ApplicationA/ApplicationAProject2.csproj $/MAIN/src/ApplicationB/... $/MAIN/src/ApplicationC $/MAIN/src/SharedInfrastructureA $/MAIN/src/SharedInfrastructureB My Goal (a pretty typical promotion pipeline) When a code change is made to a given application I want to be able to build that application and auto-deploy that change to a DEV server. I may also need to build dependencies on Shared Infrastructure Components. I often also have some database scripts or changes as well If developer testing passes I want to have an manually triggered but automated deploy of that build on a STAGING server where end-users will review new functionality. Once it's approved by end users I want to a manually triggered auto-deploy to production Question: How can I best adopt continuous deployment techniques in a multi-application environment? A lot of the advice I see is more single-application-specific, how is that best applied to multiple applications? For step 1, do I simply setup a separate Team Build for each application? What's the best approach to accomplishing steps 2 and 3 of promoting latest build to new environments? I've seen this work well with web apps but what about database changes

    Read the article

  • Set default pulseaudio volume

    - by MrUser
    When I reboot, the volume on the PulseAudio sinks is set to 100%. I know how to change the volume using pactl set-sink-volume, but that is only set temporarily. I could write a script that is run at startup, but it will not be applied if I connect a sink later. (i.e. a Bluetooth speaker) Is there a config file where I can set the default volume that is given to a device when it is first loaded by PulseAudio? Or even the master volume at startup? Clarification: The application is for a headless linux box, so the system shouldn't have to be rebooted and should be able to accept new devices without having to connect via ssh to adjust volumes, etc.

    Read the article

  • Odd Android touch event problem

    - by user22241
    Overview When testing my game I came across a bizarre problem with my touch controls. Note this isn't related to multi-touch as I completely removed my ACTION_POINTER_UP and ACTION_POINTER_DOWN along with my ACTION_MOVE code. So I'm simply working with ACTION_UP and ACTION_DOWN now and still get the problem. The problem I have a left and right button on the left of the screen and a jump button on the right. Everything works as it should but if I touch a large area of my hand (the fleshy part at the base of the thumb for instance) onto the screen, then release it and then press one of my arrows, the sprite moves in that direction for a few seconds, and then ACTION_UP is mysteriously triggered. The sprite stops and then if I release my finger and re-apply it to an arrow, the same thing happens. This goes on and on and eventually (randomly??) stops and everything work OK again. Test device & OS Google Nexus 10 Tablet running Jellybean 4.2.2 Code //Action upon which to switch actionMask = event.getActionMasked(); //Pointer Index of the currently touching pointer pointerIndex = event.getActionIndex(); //Number of pointers (for multi-touch) pointerCount = event.getPointerCount(); //ID of the pointer currently being processed (Multitouch) pointerID = event.getPointerId(pointerIndex); switch (actionMask){ //Primary pointer down case MotionEvent.ACTION_DOWN: { //if pressing left button then set moving left if (isLeftPressed(event.getX(), event.getY())){ renderer.setSpriteLeft(); } //if pressing right button then set moving right else if (isRightPressed(event.getX(), event.getY())){ renderer.setSpriteRight(); } //if pressing jump button then set sprite jumping else if (isJumpPressed(event.getX(),event.getY())){ renderer.setSpriteState('j', true); } break; }//End of case //Primary pointer up case MotionEvent.ACTION_UP:{ //When finger leaves the screen, stop sprite's horizontal movement renderer.setSpriteStopped(); break; }

    Read the article

  • How do you version/track changes to SQL tables?

    - by gabe.
    When working in a team of developers, where everyone is making changes to local tables, and development tables, how do you keep all the changes in sync? A central log file where everyone keeps their sql changes? A wiki page to track alter table statements, individual .sql files that the devs can run to bring their local db's to the latest version? I've used some of these solutions, and I'm tyring to get a good solid solution together that works, so I'd appreciate your ideas.

    Read the article

  • Git workflow for small teams

    - by janos
    I'm working on a git workflow to implement in a small team. The core ideas in the workflow: there is a shared project master that all team members can write to all development is done exclusively on feature branches feature branches are code reviewed by a team member other than the branch author the feature branch is eventually merged into the shared master and the cycle starts again The article explains the steps in this cycle in detail: https://github.com/janosgyerik/git-workflows-book/blob/small-team-workflow/chapter05.md Does this make sense or am I missing something?

    Read the article

  • Why are tools like git-svn that allow git to integrate with svn useful? [closed]

    - by Wes
    I have read these related questions: I'm a Subversion geek, why should I consider or not consider Mercurial or Git or any other DVCS? git for personal (one-man) projects. Overkill? ...and I understand why git is useful. What I don't understand is why tools like git-svn that allow git to integrate with svn are useful. When, for example, a team is working with svn, or any other centralised SCM, why would a member of the team opt to use git-svn? Are there any practical advantages for a developer that has to synchronize with a centralized repository?

    Read the article

  • Single Responsibility Principle - How Can I Avoid Code Fragmentation?

    - by Dean Chalk
    I'm working on a team where the team leader is a virulent advocate of SOLID development principles. However, he lacks a lot of experience in getting complex software out of the door. We have a situation where he has applied SRP to what was already quite a complex code base, which has now become very highly fragmented and difficult to understand and debug. We now have a problem not only with code fragmentation, but also encapsulation, as methods within a class that may have been private or protected have been judged to represent a 'reason to change' and have been extracted to public or internal classes and interfaces which is not in keeping with the encapsulation goals of the application. We have some class constructors which take over 20 interface parameters, so our IoC registration and resolution is becoming a monster in its own right. I want to know if there is any 'refactor away from SRP' approach we could use to help fix some of these issues. I have read that it doesn't violate SOLID if I create a number of empty courser-grained classes that 'wrap' a number of closely related classes to provide a single-point of access to the sum of their functionality (i.e. mimicking a less overly SRP'd class implementation). Apart from that, I cannot think of a solution which will allow us to pragmatically continue with our development efforts, while keeping everyone happy. Any suggestions ?

    Read the article

  • Unable to wake display with remote

    - by Eugene
    I'm running an HTPC (xbmc) without a keyboard/mouse attached, running oneiric. After some indeterminate amount of time, sometime between 1 and 12 hours, the display goes to sleep. The computer itself is not sleeping, I can still SSH to it from another computer. The remote will not wake the display. The IR receiver is working, as irw will show me the remote key presses. The only way to get my display back is to restart the display manager, lightdm in this case. Does anybody know a way to keep the display from going to sleep? I don't really need any power management at all considering that it connects to my TV and when I want my display to go to sleep, I turn off my TV.

    Read the article

  • How and where do you store your private work/source codes?

    - by Amir Rezaei
    I have worked as developer for over 10 years now. Over the time I have had my own small projects where I have developed tool/application and games. I have not found any robust solution to store my work. It’s always fun to get back to your code and see how you did before and how you would do it now. It’s just a work that is unfortunate to get lost. There are SVN solution such as Google’s Project Hosting. However I’m not interested in sharing my code or making it open source. Currently I’m hosting my own SVN server. So here comes my question. How and where do you store your private work/source codes? Requirements: Source code versioning Backup Prefers free

    Read the article

  • Mock RequireJS define dependencies with config.map

    - by Aligned
    Originally posted on: http://geekswithblogs.net/Aligned/archive/2014/08/18/mock-requirejs-define-dependencies-with-config.map.aspxI had a module dependency, that I’m pulling down with RequireJS that I needed to use and write tests against. In this case, I don’t care about the actual implementation of the module (it’s simple enough that I’m just avoiding some AJAX calls). EDIT: make sure you look at the bottom example after the edit before using the config.map approach. I found that there is an easier way. I did not want to change the constructor of the consumer as I had a chain of changes that would have to be made and that would have been to invasive for this task. I found a question on StackOverflow with a short, but helpful answer from “Artem Oboturov”. We can use the config.map from RequireJs to achieve this. Here is some code: A module example (“usefulModule” in Common/Modules/usefulModule.js): define([], function() { "use strict"; var testMethod = function() { ... }; // add more functionality of the module return { testMethod; } }); A consumer of usefulModule example: define([ "Commmon/Modules/usefulModule" ], function(usefulModule) { "use strict"; var consumerModule = function(){ var self = this; // add functionality of the module } }); Using config.map in the html of the test runner page (and in your Karma config –> I’m still trying to figure this out): map: {'*': { // replace usefulModule with a mock 'Common/Modules/usefulModule': '/Tests/Specs/Common/usefulModuleMock.js' } } With the new mapping, Require will load usefulModuleMock.js from Tests/Specs/Common instead of the real implementation. Some of the answers on StackOverflow mentioned Squire.js, which looked interesting, but I wasn’t ready to introduce a new library at this time. That’s all you need to be able to mock a depency in RequireJS. However, there are many good cases when you should pass it in through the constructor instead of this approach.   EDIT: After all that, here’s another, probably better way: The consumer class, updated: define([ "Commmon/Modules/usefulModule" ], function(UsefulModule) { "use strict"; var consumerModule = function(){ var self = this; self.usefulModule = new UsefulModule(); // add functionality of the module } }); Jasmine test: define([ "consumerModule", "/UnitTests/Specs/Common/Mocks/usefulModuleMock.js" ], function(consumerModule, UsefulModuleMock){ describe("when mocking out the module", function(){ it("should probably just override the property", function(){ var consumer = new consumerModule(); consumer.usefulModule = new UsefulModuleMock(); }); }); });   Thanks for letting me think out loud :-).

    Read the article

  • Why fork a library for your own application?

    - by Mr. Shickadance
    Why should a programmer ever fork a library for inclusion in a widely used application? I ask this question because I was reading an article about why Chromium isn't packaged for many Linux distros like Fedora. Apparently its largely due to the fact that Google has forked a number of libraries, modified them, and included them in Chromium. This has driven up the complexity of packaging releases. There are a number of reasons why this can be a bad thing, but how strong a case can you actually make for doing so in a large widely used application such as Chromium? The original article: http://ostatic.com/blog/making-projects-easier-to-package-why-chromium-isnt-in-fedora Isn't it usually worth the effort to make slight modifications to your own program in order to use a popular and well developed library?

    Read the article

  • Convincing my coworkers to use Hudson CI

    - by in0de
    Im really aware of some benefits of using Hudson as CI server. But, im facing the problem to convince my coworkers to install and use it. To put some context, we are developing two different products (one is an enterprise search engine based on Apache Solr) and several enterprise search projects. We are facing a lot of versioning issues and i think Hudson will solve this problems. They argued about its productivity and learning curve What Hudson's benefits would you spotlight?

    Read the article

  • What sort of security method is this called (if it has a name)?

    - by loosebruce
    I have thought of a way of securing access to an application interacting with another application. Using this method Application 1 - "What is the sum of 1+1?" Application 2 - "3" Application 1 - "Access granted" Is this method used a lot, does it have a classification in the programming world? The advantages for me of using this is that I do not have to spend more effort implementing security keys/certificates. Any unauthorized machine trying to interpret it would give the correct result and identify itself as untrusted. What sort of weaknesses are there to doing this?

    Read the article

< Previous Page | 54 55 56 57 58 59 60 61 62 63 64 65  | Next Page >