my eclipse workspace contains about 15 projects; and all of them I need basically to be open. Those projects contain insane amounts of code ... which of course, ofr anything I do, only 0.0001% are of interest.
In other words:
a) I have some production project P, that contains ONE package x.y.z
b) I have some test project P-test, that contains ONE package x.y.z
Now I want to run the JUnit test in P-test x.y.z; and afterwards I want eclemma to show me coverage ... only for the classes from P x.y.z (and of course, the main thing is that I want eclemma to only compute coverage for that one package; and to ignore anything else).
I thought this would be thing that everybody does everyday,
but the FAQ and the user guide ... gave me ideas ... which didn't help.
No matter what I do, eclemma sits there for 2 minutes, computing coverage
for ALL stuff in my workspace.
And other people seem have the same problems:
Long story short: how can I make sure that eclemma isn't spending 2 minutes to compute 0% coverage for millions of LOC I am not interested in at all?
Regards,
eg
> you can limit the analysis scope on project level. This can be done in
> the coverage launch dialog:
>
> http://www.eclemma.org/userdoc/launching.html
>
> We have a feature request to specify the scope on package granularity:
>
> https://github.com/jacoco/eclemma/issues/39
Thanks for the quick reply.
I will have to figure if the "launch dialog" settings are a good way when dealing with so many different jars and projects as I am (gut feeling: this is probably not a sufficient solution).
That feature request is obviously a step in the correct direction in my eyes, but dare me asking: how much of chance exists that this feature request will ever make it into eclemma? You know, the request is 3 years old, and doesn't show much activity around it. Is that request high on anybody's priority list? Or just sitting there to prove Leblancs law (later equals never)?