On Jan 18, 2011, at 1036AM, Sim IJskes - QCG wrote: > On 18-01-11 15:40, Dennis Reedy wrote: >> Referenced attachment >> https://issues.apache.org/jira/secure/attachment/12468648/river-modularization-overview.odt > > Thanks, > > do you have views/remarks/caveats on: > - hudson integration
There are plugins for Gradle and Maven to enable Hudson integration. I guess it depends on what you want to do. > - verification of api signatures (animal-sniffer) You may not need this because each module can have it's own JVM source & target. If you find you do require this utility, Animal Sniffer is a maven plugin, so its relatively easy to add it to a module if required. > - complete javadoc An aggregation task that just needs to be added into either the pom(Maven) or gradle build files. > - integration of QA framework Thats part of the document. > - existing Junit work in QA Easily added into the lifecycle of building the project. > - switchable JDK version See above, specific to each module, or if desired project wide default > - reproducable builds (from svn) Not sure I understand what this is > - scripted tree reorganisation Or this > > Do you mind if we lift the modularisation part, and put this on the website > under discussions? Or is this heavily dependend on the build solution? I think we should keep it in Jira and discuss it as part of RIVER-300 > > Do you have a favorite mvm or gradle, based on which reason?
