[1.34.0] Optional Regressions
Fernando, Optional still fails on various msvc flavours. Any idea what's going on? Thomas -- Thomas Witt witt@acm.org
Thomas Witt wrote:
Fernando,
Optional still fails on various msvc flavours. Any idea what's going on?
Stale results? It appears the turnaround time is much higher that we'd like. Of all optional failures, only optional_test / msvc-6.5_stlport4 is relatively recent, and we don't care about that configuration much. - Volodya
Vladimir Prus wrote:
Stale results? It appears the turnaround time is much higher that we'd like.
I have to second that. I am running as fast as I can! What puzzles me most is: Does it really take almost 24 hours to rebuild the status pages? E.g.: Status page last update Report Time: Sat, 10 Mar 2007 19:25:43 +0000 Upload of fresh results: Sun, 11 Mar 2007 02:47:?? +0000 Now (Sun 11 Mar 2007 10:54 +0000) I am looking on the page and still seeing only results from Saturday. Is the page build process that slow, or would it just needed to be scheduled more often? Roland
Vladimir Prus wrote:
Thomas Witt wrote:
Fernando,
Optional still fails on various msvc flavours. Any idea what's going on?
Stale results? It appears the turnaround time is much higher that we'd like. Of all optional failures, only
optional_test / msvc-6.5_stlport4
is relatively recent, and we don't care about that configuration much.
Hmm... it's probably my fault, but when I look at the page I have msvc-7.1 msvc-8.0 Rudbek Asociates V2 from Sat 10. What am I missing? Thomas -- Thomas Witt witt@acm.org
Thomas Witt wrote:
Vladimir Prus wrote:
Thomas Witt wrote:
Fernando,
Optional still fails on various msvc flavours. Any idea what's going on?
Stale results? It appears the turnaround time is much higher that we'd like. Of all optional failures, only
optional_test / msvc-6.5_stlport4
is relatively recent, and we don't care about that configuration much.
Hmm... it's probably my fault, but when I look at the page I have msvc-7.1 msvc-8.0 Rudbek Asociates V2 from Sat 10. What am I missing?
As I said in my first post about these, "optional_test_ref_fail2" is no longer in the Jamfile. So these "fail" results for msvc-7.1 and msvc-8.0 are wrong. The test shouldn't execute, and if it still does, it should just be ignored. That leaves msvc-6.5 (from various testers) and msvc-7.0, which should be marked as expected failures IMO. Best
Fernando Cacciola wrote:
As I said in my first post about these, "optional_test_ref_fail2" is no longer in the Jamfile. So these "fail" results for msvc-7.1 and msvc-8.0 are wrong. The test shouldn't execute, and if it still does, it should just be ignored.
As I wrote several times: Please, everyone, inform the testers about changes that involve renaming or removing tests! Several of the testers run tests incrementally. Until there's support for removing old tests in Boost.Build (which might never be the case) old tests have to get deleted manually. Regards, m Send instant messages to your online friends http://au.messenger.yahoo.com
Martin Wille wrote:
Fernando Cacciola wrote:
As I said in my first post about these, "optional_test_ref_fail2" is no longer in the Jamfile. So these "fail" results for msvc-7.1 and msvc-8.0 are wrong. The test shouldn't execute, and if it still does, it should just be ignored.
As I wrote several times:
Please, everyone, inform the testers about changes that involve renaming or removing tests!
Oh. We should post to this list saying so I guess.
Several of the testers run tests incrementally. Until there's support for removing old tests in Boost.Build (which might never be the case) old tests have to get deleted manually.
OK. I didn't know that. I expected it to be as simple as removing them from the bjam file. Best Fernando Cacciola
Hello testers, The following tests have been removed from optional/test/Jamfile.v2 [ compile-fail optional_test_fail2.cpp ] [ compile-fail optional_test_ref_fail2.cpp ] Best Fernando Cacciola
participants (5)
-
Fernando Cacciola -
Martin Wille -
Roland Schwarz -
Thomas Witt -
Vladimir Prus