Hello Marc,
I could try to detect the surefire setting, though I have no clue whether
this is a trivial task. Documenting the behavior is a simple workaround.
Regards
Mirko
--
Sent from my mobile
On Apr 14, 2014 8:30 PM, Marc R. Hoffmann hoffm...@mountainminds.com
wrote:
Hi,
thanks for this
Silvio, the java compiler is probably creating multiple branches in byte
code for your source. The number of branches depend on the jdk version.
Just search in your favorite search engine for jacoco try finally.
Regards
Mirko
--
Sent from my mobile
On Dec 22, 2014 9:36 AM,
Hello Leela,
please just google for jacoco and powermock, both modify the loaded
classes, so depending on your scenario different approaches are valid.
Regards
Mirko
--
Sent from my mobile
On May 6, 2015 7:05 PM, leela kumili lkum...@gmail.com wrote:
Have a multi-module maven project.
Each
Hello Michael
are you using mvn site? Just create a file src/sites/resources/.nojekyll
otherwise GitHub will take a gh-pages branch as to be processed by the
Jekyll blog software.
Regards
Mirko
--
Sent from my mobile
Am 20.10.2015 15:49 schrieb "Michael Remijan" :
> It
Hello,
if you only execute one module the other modules will be pulled from your
maven repository manager and the class files will probably be different. So
this is to be expected.
Regards
Mirko
--
Sent from my mobile
Am 07.06.2016 20:31 schrieb "Thanuku Krishna Chaitanya" <
Hello Thanuku,
* you have to include all modules for which you want to have the aggregated
report as dependencies of the aggregating module
* in your aggregating module you only need the report-aggregate execution.
* you have to bind the execution to a phase, e.g. verify.
See
Hello Thanuku,
I have seen this as well, when I had scope for a dependency. Just to be
sure: the single module reports are created correctly? Aggregate only
collects previously created jacoco.exec files.
Regards
Mirko
--
Sent from my mobile
Am 07.06.2016 17:19 schrieb "Thanuku Krishna
Hello Sahana,
it looks like your build server is running on Windows. Back in the day I
always set my basic workspace directory to sth. like C:\\ws to make sure I
do not run into the 255 character path limit.
Regards
Mirko
--
Sent from my mobile
Am 04.03.2016 20:36 schrieb
Hello Florian,
did you put the reporting section into the profile as well? This should
just work AFAIK.
Regards
Mirko
--
Sent from my mobile
Am 19.04.2016 13:42 schrieb :
> Hi,
>
> In our project we use offline instrumentation as we are doing AspectJ
> compile-time weaving.
eport-plugin:2.19.1:report < [surefire]test @
>>> PROJECT <<<
>>>
>>>
>>> So, surefire-reports reruns surefire:test with all the preparation
>>> steps. Whereas surefire:test detects that it was already ran,
>>> jacoco:instrument doe
Hello,
you could try to use Maven 3.3.9 as well in Travis, see
https://github.com/mojohaus/mojo-parent/blob/master/.travis.yml for how to
use the takari wrapper plugin.
Regards
Mirko
--
Sent from my mobile
Am 04.07.2016 22:15 schrieb :
> Hello,
>
> I am trying to
Output here:
https://gist.githubusercontent.com/patrickpilch/21d32e367b312443c98cbf2bcaa687fa/raw/fac1d3372bb24d18f3df7fcba5a8acc9a0c4296a/wrapper_test.txt
On Monday, July 4, 2016 at 4:36:02 PM UTC-5, Mirko Friedenhagen wrote:
> Hello,
>
> you could try to use Maven 3.3.9 as well in Travis
Hello,
I am playing with Kotlin and tried to do a POC for a SonarQube plugin[0]
While doing this I encountered a strange behavior (at least for me)
SonarQube iterates on iSourceFileCoverage and checks that the number
of lines in iSourceFileCoverage do not exceed the number of lines of
the
Hello,
just for the record: But in the kotlin-tracker
https://youtrack.jetbrains.com/issue/KT-16002
Regards
Mirko
Regards Mirko
--
http://illegalstateexception.blogspot.com/
https://github.com/mfriedenhagen/ (http://osrc.dfm.io/mfriedenhagen)
https://bitbucket.org/mfriedenhagen/
On Thu, Jan
Hello,
one hint: reports using the Maven plugin may report different coverages
than the ones found in SonarQube because aggregation done by the tools is
different.
So if you have hard limits for a failing QA gate in SonarQube you may have
discussions with developers.
Regards
Mirko
--
Sent from
t be i miss something here?
>
> For the skip - do you run these projects with skipTests set?
>
> Kind Regards,
> Andreas.
>
>
> Am Samstag, 29. Oktober 2016 12:59:52 UTC+2 schrieb Mirko Friedenhagen:
> > Hello,
> >
> > you mean skip when maven.test.
Hello,
you mean skip when maven.test.skip is set?
You may achieve this easily by referencing this property in the
default-prepare-agent execution. There you may easily redefine the bound
phase as well.
I use jacoco in projects which have no tests run by surefire or failsafe
but the invoker
Hello, at my company we use jacoco with mockito all the time,
power-mockito or any framework which manipulates the classes when loading
them is more tricky.
Regards
Mirko
--
Sent from my mobile
Am 05.10.2017 03:37 schrieb :
> I've banging my head against the wall
Hello,
another caveat: which metric to use for this. I would go for complexity, as
it is the most valuable IMO.
Best regards
Mirko
--
Sent from my mobile
Am 11.02.2018 22:42 schrieb "Marc Hoffmann" :
> Hi Andy,
>
> If the coverage is greater than or equal to the
Hello,
a bit off-topic, please hop to the sonarqube forum for further details.
Basically you have to create the coverage beforehand by running e.g. “mvn
Test” after configuring the jacoco-maven-plugin in your project.
mvn sonar:sonar will then pick up the coverage data and show you detailed
Hello,
I do not think Jacoco offers this out of the box. What you would need is
historisation of data.
I would recommend to setup sonarqube for this which will not only historise
coverage but additionally will do a lot of other useful analysises of your
code which will help you to find and fix
Hello Aswak,
this is something you would better ask on the jenkins user mailing list, as
the jacoco developers do not track which versions of derived software are
using which version of jacoco itself.
That said I do not think there is a wiki-page, however you could go to the
github page of the
Hello Mike,
if you only configure the goals "prepare-agent" and "report" as executions
the coverage should be automatically measured for tests run by surefire and
failsafe combined.
Just *do not* define executions for prepare-agent-integration resp.
report-integration.
Could you show the
23 matches
Mail list logo