does anybody know if Cobertura provides possibility to automatically compare code coverage reports of two builds? If needed, I am always doing it manually in separate tabs in browser, which takes some time for larger modules
Related
I have an lcov report which produces a pretty html report and an lcov.info file. I want to see on jenkins whether our code coverage is improving or degrading. Is it possible to do this with an lcov report? I can do it using cobertura, but for various reasons we need to stick with lcov. We also need to see statements, branches, functions and lines.
The coverage report is being produced by istanbul/nyc.
I'm trying to get Cobertura code coverage reports integrated into Jenkins for my company's AngularJS project. It's working, however, packages files and classes with es5 AngularJS project are always 100%. What I'm more focused on obviously, are the methods lines and conditionals.
Since the first 3 metrics are always 100% the graph leaves little to be desired as I can't zoom in close enough to notice slight changes in the metrics.
Is there a way to exclude Packages, Files, and Classes from the report?
I'm calculating code coverage on every build done in Jenkins and producing a coverage XML report. That's recorded really nicely inside of Jenkins with Cobertura, but what I'd really like is to be able to somehow get at the total branch coverage number so I can automatically publish to a medium the rest of my team can easily consume (i.e. Slack).
As a bonus, getting the difference in coverage from the last run would be even better. I don't see any environment variables that hold this, and haven't found anything detailing a simple way to do this in Cobertura docs. I know I can hack some code together to do this myself, but if there's a simple way I'm missing or something someone else has already built, I'd much rather do that.
I believe you'll have to do this yourself.
The existing Slack plugin just sends build start/success/failure notifications etc.
The Cobertura plugin unfortunately doesn't seem to be built on top of the static code analysis plugin, so there probably isn't much in the way of graphs, difference reports and all that.
You could try adding /api/json to the end of a Cobertura report URL for a Jenkins build — most endpoints reveal some information in this way. If there's some useful information, that could be a basis for whatever you want to hack together.
Is there a way to aggregate code coverage data from jmockit-coverage and emma coverage? I can run the two different coverage steps in two separate junit ant tasks and generate the coverage data in two directories. Just not sure if the coverage outputs from these two are compatible and can be merged to display together.
No, they can't be aggregated into a single HTML report, since these are two different code coverage tools, which know nothing about each other. Of course, someone could create yet another tool which would do that; personally, I don't see much value in it, though.
I am using sonar for the last few months and want to know that do sonar works in an incremental way or not i.e if i do soanr analysis for the first time on my project code it will definitely analyze all the code but if i do some enhancement on my core source code and update some files then after updation do sonar analysis again on the same code then will sonar analyze all the files OR only analyze files which i have updated?
I am using "Sonar way with Findbugs" as my default quality profile.
Is there any way to use sonar in an incremental way, to analyze only updated files?
Is this possible in sonar or not?
Kindly revert your help will be appreciated..
Thanks in advance..
Sonar does not currently handle incremental analyses.
If you want, you can watch and vote for the following ticket: http://jira.codehaus.org/browse/SONAR-2815
If your task if code quality metrics in a general way, then you can use these tools directly on Jenkins. See FindBugs Plugin , PMD plugin or Checkstyle Plugin. They independently give what you need in the form of "incremental code" reports. But if you have to use SONAR only, then Fabrice's answer is the correct one.
Here the post where the ticket that #fabrice-sonar-team comments was born:
does Sonar support Incremental code quality analysis
You can read Freddy Mallet's explanation about why this functionality is not trivial to be implemented. It requires lot of effort, so as Fabrice said, vote for the ticket! :)
Just adding on since this is the first google result for 'Sonar incremental preview analysis' in stack, and the answers are way too old.
Sonar has plugins available for IDEs such as Eclipse and NetBeans that can run incremental analysis on the changed files alone. This still needs a connection the SonarQube server though.
You may also run the analysis right from Jenkins by passing an additional sonar analysis property - sonar.analysis.mode=incrementalin your Sonar scan build step. This will report the code quality in a full report - will all code issues as well as a light report containing only new issues (since last full scan as recorded in server).
To take one step further use sonar.issuesReport.html.enable=true and publish the generated html reports to your Jenkins build page - Neat and Trim
Sonar documentation here