<div class="gmail_quote">On Fri, Jun 8, 2012 at 11:39 AM, Christoph Grüninger <span dir="ltr"><<a href="mailto:christoph.grueninger@iws.uni-stuttgart.de" target="_blank">christoph.grueninger@iws.uni-stuttgart.de</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi David,<br>
thanks for your answer and the bug tracker link. I don't have any new arguments besides that the arguments and use cases stated in the bug tracker apply to our use case, too.<br>
<br>
I see it like a poor man's code coverage. Once I forgot to enable MPI support in autotools and it reported 47 tests passed, 1 skipped. I cannot memorize the absolute number of tests for every program but the mentioned skipped tests made me investigate which test was skipped. After reconfiguring all 48 tests passed and I was sure, my MPI related change did no harm. If the skipped test was reported as passed, I actually would falsely assume all tests passed.<br>
<br>
As already stated in the bug tracker, I would report skipped tests as "not run" and leave CDash as it is. Only CTest needs changes, so no "can of worms" to be opened.<br>
<br>
Thanks for your answer,<br>
Christoph<div class="HOEnZb"><div class="h5"><br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
This has been requested before:<br>
<br>
<a href="http://public.kitware.com/Bug/view.php?id=8466" target="_blank">http://public.kitware.com/Bug/<u></u>view.php?id=8466</a><br>
<br>
It's in the backlog, would require changes to both ctest and cdash, and<br>
I'm not enthusiastic about it. Perhaps I just need convincing that it's<br>
a good idea. For now, there are no plans to implement anything like<br>
this, and I would recommend that you use a wrapper script approach,<br>
printing out "skipped" and returning success when you don't actually run<br>
your test.<br>
<br>
<br>
HTH,<br>
David<br>
<br>
</blockquote>
<br>
-- <br>
People who are more than casually interested in computers should have<br>
at least some idea of what the underlying hardware is like. Otherwise<br>
the programs they write will be pretty weird. -- Donald Knuth<br>
--<br>
<br>
Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
<br>
Visit other Kitware open-source projects at <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/<u></u>opensource/opensource.html</a><br>
<br>
Please keep messages on-topic and check the CMake FAQ at: <a href="http://www.cmake.org/Wiki/CMake_FAQ" target="_blank">http://www.cmake.org/Wiki/<u></u>CMake_FAQ</a><br>
<br>
Follow this link to subscribe/unsubscribe:<br>
<a href="http://www.cmake.org/mailman/listinfo/cmake" target="_blank">http://www.cmake.org/mailman/<u></u>listinfo/cmake</a><br>
</div></div></blockquote></div><br><div><br></div><div>There is a test property called REQUIRED_FILES, which is a list of files that must exist in order to run a test. According to the ctest source code, we will return a status of "Not run" if one of those files does not exist.</div>
<div><br></div><div>So one way to achieve your goal would be to list the test via add_test always, but then require a file for the test to run, and make sure that file does not exist before ctest_test is run.</div><div><br>
</div><div>So it seems like you should be able to achieve your goal using that if you would like. As long as you're ok with the "Not run" status.</div><div><br></div><div><br></div><div>HTH,</div><div>David</div>
<div><br></div>