I totally agree with Stephen here, the pipeline script engine is a lock down groovy with many quirks and approval mess. Making it a painful experience. Do as much as possible into your build scripts (.vcxproj, makefile, shell script, python...). Just launch them and take light decision into pipeline. If you shell scripts are working, you may want to clean them to just launch the build and run sequence and let Jenkins pipeline deceide which node should be launch and do the tests on each platform/OS. The declarative pipeline seem to make it easier to do things like this, but from what I have seen into this forum so far, many things ain't working so well yet, but it's definitely the future thing with blue ocean.
I for one do use jenkins pipeline (non declarative one) and it make only 243 lines (with a lots of comments and even file headers) that does all the following: 1. checkout repos 2. checkout an helper repos for scripts (python, executable that does specific jobs) 3. build whole solution with msbuild 4. build unit tests 5. run the tests 6. extract results results 7. extract benchmarks data and graph them between build (into an HTML web graph) 8. Syntax checking for qmllint, Qml dir inspection, .vcxproj, .csproj 9. Generate the Doxygen doc Many of those things are done into python outside pipeline script, the pipeline only launch them (using either sh or bat) back to back as different stage. -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/663ec671-39c1-4708-8c06-0af75396428d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
