I am using Coverity 7.6.0 for our source code (C/C++). I did an initial Coverity run on our source code:
I can see all the results in a web interface. At this moment a lot of the issues were fixed and I need to run again Coverity in order to verify that what we did really solves the reported issues.
My concern is: how to avoid having duplicate entries for the issues that weren't yet addressed.
Do we need to use the same intermediate directory?
Can we use different directories and only push to server what is not already reported?
At what stage can we do this sort of filtering?
How are already reported issues affected if the place in the code changed?
Is there a workflow to do this daily?
My experience with Coverity as an administrator is work in progress but please don't try to be gentle :)
It seems that Coverity has this thing already figured out. You can run Coverity as much as you like as it merges all the findings and also removes the old issues from the display.
So no need to stress, just run Coverity on the code as much as you like.