If there is a project that has tests executed as part of the build process on the build machine, if any set test Fail, will the whole build fail?
What should anyone consider when answering that question? Is this the point of which tests are failing?
Background information that inspires this question:
Currently I am working on a project that has been done as part of the construction of the test. Process and execution on our build machine.
This project was used to set up so that any test fails, the build fails. The argument is that if the test fails, it means that the product is not working / is not complete / it is a project failure, and hence the build should fail.
We've added some tests, though they fail, they are not important for the project (see below for more details). So, if those tests fail, then the project is not completely unsuccessful, and we still want to make it.
One of the passage tests confirms that there is an exception to the wrong arguments, but the test pass is one which checks that all the permissions logic not , so the class dismisses all invalid cases, but some legitimate ones too. This is not a problem for the project, as the rejected legitimate logic is Fringe, which the application will not trust.
If this is possible in any way, do so. It becomes very less:
With any system (visible) defects, it is generally considered to be a very bad idea to start a small defect. So if you have got a project with a green status (no unit tests fail) and you start the first unsuccessful test, you (and / or your colleagues) will be prompted to fix the problem.
If, on the other hand, there are known-unsuccessful tests, then only adding another broken test is seen keeping in mind the situation of the situation.
So you should always try to keep all the test is going on (and not just "most of them") and treating every unsuccessful test because the build fails Due to being that leads a long way towards that goal.
Comments
Post a Comment