You can use a multi-configuration (matrix) job to build all the 'flavors' you 
need in a single job, archive the artifacts from each configuration, then use 
them in downstream jobs as you see fit. Your deployment job will be deploying 
binaries built from the same source code as the unit-test/coverage job binaries 
were built from, but of course won't be deploying the same binaries.

----- Original Message -----
From: [email protected]
To: [email protected]
At: Aug 26 2013 13:16:00

Hi, I currently have three jobs running one after the other: 
build->unittest->regression.

I want to add code coverage results to the unittest & regression I have. For 
that end, I have to build with specific debug flags. I see few alternatives:


  1)Create a parallel chain of jobs: build with code coverage->unittest with 
code coverage->regression with code coverage. Pro - it's clean, Cons - it 
duplicates the jobs I have.

  2)Use the same jobs I have, but build with code coverage->unittest with code 
coverage->regression witrh code coverage. Pro - I don't have duplication of 
jobs. Cons - I'm testing a version different than production candidate.
  3)Add more logic to the current 3 jobs, so I'll build and run a release 
version, and a code coverage version. Pro - I don't have duplication of jobs. 
Cons -seems less clean to me.


Thanks.
-- 
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].
For more options, visit https://groups.google.com/groups/opt_out.

-- 
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].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to