How to run "Failed Tests" only from previous nunit .xml report file.

[constructor]

Could you help me to solve my problem?
Each night we run tests and store report to .xml report file.
When a programmer fixes failed test we want to re run only failed
tests immediately.
So, we need a possibility somehow to say to nunit to run failed
(include error) tests from stored .xml report file.

[constructor (igor-zc)]

we need the feature in command line nunit.

[Charlie Poole (charlie.poole)]

OK...

I imagined a situation where the developer would use the gui to load the saved
result file, which would provide the info as to what had failed. The
actual project
file and assemblies would be found using info in the result file.

This is analogous to what we now do with the VisualState.xml file that stores
the latest results between gui runs.

Can you describe how you see it working using the console runner?

Charlie

On Sun, Feb 20, 2011 at 12:13 PM, constructor <email address hidden> wrote:
> we need the feature in command line nunit.
>
> --
> You received this bug notification because you are subscribed to NUnit
> Extended Testing Platform.
> https://bugs.launchpad.net/bugs/722207
>
> Title:
>  How to run "Failed Tests" only from previous nunit .xml report file.
>

[constructor (igor-zc)]

Right now, we run all regression tests each night (with help of Final Builder and NUnit console application). The process can take 5 hours.
When we have a failed tests, our programmer should fix the failed tests. But, unfortunately, I have not automatic way to check it. And now I should manually to run failed (fixed) tests to be sure if the tests really fixed.

So, I want to FinalBuilder script, which will check failed tests (with help of last Nunit test report file) and send me report with status of all failed tests: what tests were fixed, what tests were not fixed.

Scenario:
1. Run all tests (5 hours). Result of the process: report.xml file with all tests.
sample: nunit-console-x86.exe MyAssembly /XML=report.xml

2. When we have failed tests, programmers should fix the tests.
3. When I receive email from programmer, I want to run NUnit console and check new status of all failed tests from file report.xml.
sample: nunit-console-x86.exe MyAssembly /Fillter=error+failures;report.xml /XML=newReport.xml

thank you,
Igor.

[Charlie Poole (charlie.poole)]

While this is not the right place for a discussion, I can't help but ask...
how do you know that the fix didn't break other tests?

We'll look at trying to support this scenario in some way but since
NUnit is first and foremost a tool for programmers, the ability for
the developer to run the tests on the desktop has to have first
priority.

I'm assuming that you would want to pass the report file in as
an argument to NUnit-console and have NUnit figure out what
 to run from that point on.

Charlie

On Sun, Feb 20, 2011 at 1:29 PM, constructor <email address hidden> wrote:
> Right now, we run all regression tests each night (with help of Final Builder and NUnit console application). The process can take 5 hours.
> When we have a failed tests, our programmer should fix the failed tests. But, unfortunately, I have not automatic way to check it. And now I should manually to run failed (fixed) tests to be sure if the tests really fixed.
>
> So, I want to FinalBuilder script, which will check failed tests (with
> help of last Nunit test report file)  and send me report with status of
> all failed tests: what tests were fixed, what tests were not fixed.
>
> Scenario:
> 1. Run all tests (5 hours). Result of the process: report.xml file with all tests.
> sample: nunit-console-x86.exe MyAssembly /XML=report.xml
>
> 2. When we have failed tests, programmers should fix the tests.
> 3. When I receive email from programmer, I want to run NUnit console and check new status of all failed tests from file report.xml.
> sample: nunit-console-x86.exe MyAssembly /Fillter=error+failures;report.xml /XML=newReport.xml
>
> thank you,
> Igor.
>
> --
> You received this bug notification because you are subscribed to NUnit
> Extended Testing Platform.
> https://bugs.launchpad.net/bugs/722207
>
> Title:
>  How to run "Failed Tests" only from previous nunit .xml report file.
>

[Charlie Poole (charlie.poole)]

Moved issue to https://github.com/nunit/nunit-gui/issues/15