build-automation

问题 I've been trying to understand what Ant is used for but i still don't get it. Can someone give me one use case for which Ant is used for, something I can try to see why Ant is useful and what kind of things I can use it for? I do Java development in Eclipse and I'm just getting started with servlets and Google Web Toolkit. 回答1: Ant is a build tool. Say for example you have several projects in your Eclipse workspace, all of which are pieces of a larger application. To build it all into a jar

问题 I have an issue running xunit.net tests in VSTS . When running the Build plan, the Test assemblies step produces the following log: 2017-03-21T12:11:39.3302859Z ##[section]Starting: Test Assemblies 2017-03-21T12:11:39.3382932Z ============================================================================== 2017-03-21T12:11:39.3382932Z Task : Visual Studio Test 2017-03-21T12:11:39.3382932Z Description : Run tests with Visual Studio test runner 2017-03-21T12:11:39.3382932Z Version : 1.0.84 2017

问题 How to make this possible ? I had a set of jar files which are to be included to CLASSPATH variable. I don't want to give the command SET CLASSPATH=xxx.jar;xx.jar;.. as part of the build step. I dont' want to manually set the Environment variable CLASSPATH as part of system properties. I tried by copying a set of jar files into Jenkins_HOME/war/WEB-INF/lib and had started the Jenkins server. But couldn't make it possible... Any Solution ? 回答1: You can try by additional class-path elements

问题 I am trying to write a shell script that will install all of our development tools & dependencies onto a clean OSX machine. Does anybody know the best approach to automate the installation of Xcode? I am doing this to: Document the development environment. Speed up the on-boarding process for new developers. Follow the automate-everything principle. 回答1: Fully Automated XCode Installation First, login to the Apple Developer Downloads Site and get the version of XCode.dmg that works for your

问题 I'm trying to automate a program I made with a test suite via a .cmd file. I can get the program that I ran's return code via %errorlevel%. My program has certain return codes for each type of error. For example: 1 - means failed for such and such a reason 2 - means failed for some other reason ... echo FAILED: Test case failed, error level: %errorlevel% >> TestSuite1Log.txt Instead I'd like to somehow say: echo FAILED: Test case failed, error reason: lookupError(%errorlevel%) >>

问题 When you are monitoring the TFS build from Visual Studio (2008 or 2005), you can see where it is up to. The issue is that I have some Post-Build custom steps I would like the developer to be able to see directly throught the UI. Those steps take some times and we can also get a "timing" of the build step. Any idea how to have it displayed? 回答1: This is the pattern that I normally use for adding steps to the build report in TFS 2008. (See http://code.msdn.microsoft.com/buildwallboard/ for the

问题 When you are monitoring the TFS build from Visual Studio (2008 or 2005), you can see where it is up to. The issue is that I have some Post-Build custom steps I would like the developer to be able to see directly throught the UI. Those steps take some times and we can also get a "timing" of the build step. Any idea how to have it displayed? 回答1: This is the pattern that I normally use for adding steps to the build report in TFS 2008. (See http://code.msdn.microsoft.com/buildwallboard/ for the

问题 I would like to modify the string in a .h file with NAnt before building the solution. There is a macro in .h file: #define SERVER_ADDRESS "www.customserver.net" and I would like to modify the string before deploying software so each build could be made for custom address by passing the address in command line. Does anyone know how this could be done? Thanks! 回答1: One could use the loadfile task to help with this. This task loads the given file into a property. What is really useful is when

问题 We just published parts of our library (Mango) which is a Scala wrapper around Google Guava common libraries for Java. The library currently depends on Google Gauva 14.0, but we would like to add support for other versions as well. Is there a way in sbt, a build-automation tool for Scala and Java projects, to create maven like profiles, where each profile compiles against a different version of the respective Guava dependency, so we can include it in the continuous integration test matrix?

问题 I have a cross platform C++ project that targets and successfully compiles on Linux, OSX and Windows. I'm using GNU Make to handle the building on all platforms, gcc for compiling under Linux & OSX and cl.exe to compile under Cygwin on windows. My current workflow consists of coding under OSX and then building on each individual platform to test code portability. This process is somewhat time consuming and I was wondering if it is possible to automatically build on all platforms in one step?