We have a custom Skylark rule that invokes jar after using the Thrift compiler to generate .java files, modeled after genproto.bzl.
What is the recommended way to invoke an external jar command in a hermetic way? Should we use new_http_archive to pull in the JDK archive (and how would this work with the DMG Oracle provides for OSX) with something like jdk.BUILD?
You have #local_jdk//:jar that expose the jar binary coming from the jdk used by Bazel.
Related
I have a groovy/grails jar file. I want to decompile it into groovy files. Is there any tool to do that. I have tried some online decopilers , they all decmpile the files into java not groovy. I have created the jar using gradle jar command. I'm using grails version 3.1.11 and groovy version: 2.4.7 and gradle verion: 2.13
You can use any Java decompiler, but you will not get Groovy code back, as no such tool exists (to my knowledge)
This is one of the weaknesses of non-Java JVM languages. Groovy does a lot of reflection and other such trickery in creating its bytecode, so if you decompile the .class file, you'll have a lot of noise. This is similarly reflected in Groovy stacktraces. You have to train yourself to find the signal despite all of the noise. The good news is that both Eclipse and IntelliJ can debug Groovy code nicely, so you can step through to find the issue you're dealing with.
I'm able to successfully build the Maven assembly plugin in my project and generate a jar file with all the needed dependencies. But now I also want to instruct Maven after building me the jar file with dependencies, go into the target folder where the jar file with dependencies is located and run my main program.
Should I consider looking into the Maven Exec Plugin for what I want to acheive?
Yes the exec-maven-plugin is the right choice. The question is if you like to start the assembled jar archive or just a java class with it's dependencies.
I am trying to use the Ant <scp> task in my buildfile. It gives me an error every time I run the target:
cannot be found: com/jcraft/jsch/UserInfo
I have read some posts that talk about having this issue with Ant 1.7. Some of the posts are talking about removing the ant-jsch.jar files from the ant_home lib. Unfortunately I can't do that as the ant_home is common across lot of other projects and I can't remove or update any jar files from the lib directory.
Is there a way that I can solve this issue purely using Ant configuration?
ant-jsch.jar contains ssh related tasks developed by Apache Ant team, while jsch.jar contains the jsch library which is the dependency of those ssh tasks.
So do a little check: check if jsch.jar is in your $ANT_HOME/lib and also check if the jar contains UserInfo.class.
If one of above happens, download the latest version of jsch.jar and put it in $ANT_HOME/lib.
I am currently using Ant 1.8.4 with a jsch.jar downloaded last year without any problem.
Update:
Check this page first:
http://ant.apache.org/manual/install.html#optionalTasks
Please note that <scp> and other ssh tasks (the ant-jsch.jar) are included in the Ant binary distribution by default. You don't need to write <taskdef> to use <scp> because it is included in Ant's lib directory by default. I recommend you to put jsch.jar in that directory with the following points:
These ssh tasks don't support classpath def, that is to say, you can't define a <classpath> pointing to jsch.jar and then reference to it in an ssh task.
Also you can put jsch.jar in any directory included in your system's CLASSPATH environment variable, but it is not a good practice.
We usually put a task and its dependency in the same directory.
What approach has worked well for you combining IVY + msbuild?
Our goal is to integrate IVY into the C#/C++ build process for dependency resolution and publishing. We have tried adding it to custom tasks at the beginning and end of the build and we have tried wrapping the msbuild calls with ant+ apache-ant-dotnet.
Other options might be gradle, buildr, rake.
What do you use?
Thanks
Peter
Most build technologies can use libraries found in a local directory. I'd suggest using the command-line ivy program to populate this, at the start of your build:
java -jar ivy.jar -ivy ivy.xml -settings ivysettings.xml -retrieve "lib/[conf]/[artifact].[ext]"
Your dependencies are listed in a standard ivy file called ivy.xml. The protocol, location and layout of your remote repository is described in ivysettings.xml
The advantage of this approach (as opposed to switching to Gradle, etc) is that you're not trying to replace your existing build tool. Ivy is solely concerned with managing dependencies.
My team has been using Ivy for .NET for a couple of years very successfully. I know several more that give it a vote of confidence.
Use it standalone. Wrap calls into msbuild tasks. No need to use Ant integration.
I have a Java project that is dependent on other Java projects that are siblings and there is a chain of dependencies. Each individual project has a build script written in Ant. For clarity find below a sample of the same.
EARProject depends on WebProject and EJBProject: The war file that is generated by the WebProject build and jar file that is generated by the EJBProject are needed to build the EARProject.
WebProject depends on ComponentOneProject: The jar file that is generated by the ComponentOneProject build is needed to build WebProject.
EJBProject depends on ComponentTwoProject: The jar file that is generated by the ComponentTwoProject build is needed to build EJBProject.
So, when I build the EARProject build, if the dependent war and jar have not been built yet, then it should kick-off the WebProject build and EJBProject build and if the ComponentOneProject is yet to be built, the build of ComponentOneProject needs to be kicked-off and so on.
Can someone suggest a clean method by which we can accomplish this?
Facing the same problem we at our company wrote a custom Groovy script that explores the full dependency tree ant generates the Ant build scripts based on all the .project, .classpath, .settings/* files. This wasn't as difficult as it might seem as first. This way we can build our products without (My)Eclipse on a clean CVS+JDK+Groovy virtual machine. Hope it helps..