Hi,
I introduced this point of view with the native JaCoCo reports. The
benefit comes if you compare projects (packages etc): What does it tell
you if A has 80% code coverage and B has 60%? B is "better"? No way, for
two reasons:
1) Just because you have code coverage it does not mean the code is
correct. So all it says is that A has 20% uncovered code and B has 40%
uncovered code.
2) How do you compare this figures if A has a total of 10'000 LOC and B
100 LOC? This means A has 2'000 lines if untested code while B only has
40 untested lines. So even if A has a way better coverage percentage it
is definitely the project to start testing with.
Best regards,
-marc
>> Eclipse developer at JBoss, by Red Hat [1]
>> My blog [2] - My Tweets [3]
>
> --
> Azért kapta ezt az üzenetet, mert feliratkozott a Google Csoportok
> szolgáltatásbeli Jenkins JaCoCo plugin mailing list csoportra.
> Az erről a csoportról és az ahhoz kapcsolódó e-mailekről való
> leiratkozáshoz küldjön egy levelet a következő címre:
>
jenkins-jacoco-plugin-m...@googlegroups.com.
> Ha üzenetet szeretne küldeni ebbe a csoportba, küldjön egy
> e-mailt a(z)
jenkins-jacoco-pl...@googlegroups.com
> címre.
> További lehetőségekért látogasson el a(z)
>
https://groups.google.com/groups/opt_out [4] címre.
>
>
>
> Links:
> ------
> [1]
http://www.jboss.org/tools
> [2]
http://mickaelistria.wordpress.com
> [3]
http://twitter.com/mickaelistria
> [4]
https://groups.google.com/groups/opt_out