[test] Unrecognized option -u
Hi, I'm seeing test errors about unrecognized option -u presumably from Boost.Test, see here for example: http://www.boost.org/development/tests/develop/developer/output/teeks99-09b-... The test in question is not mine, and I don't know what the -u option is supposed to be for, but I assume it was supported by Boost.Test at some point and now it isn't. Is this an intended change in Boost.Test, an oversight or just a mistake in the Jamfile? If it was supported by Boost.Test, what did the option do and is it ok to just remove it?
Le 02/09/16 à 17:48, Andrey Semashev a écrit :
Hi,
I'm seeing test errors about unrecognized option -u presumably from Boost.Test, see here for example:
http://www.boost.org/development/tests/develop/developer/output/teeks99-09b-...
Hi, Thanks for pointing that out!
The test in question is not mine, and I don't know what the -u option is supposed to be for, but I assume it was supported by Boost.Test at some point and now it isn't.
I am not aware of any -u option. I believe this option was silently ignored for at least a decade. What boost.test added in 1.60 is a stricter CLA checking (http://www.boost.org/doc/libs/1_61_0/libs/test/doc/html/boost_test/change_lo...)
Is this an intended change in Boost.Test, an oversight or just a mistake in the Jamfile? If it was supported by Boost.Test, what did the option do and is it ok to just remove it?
What bugs me there is that the failing test is using the deprecated test_execution_monitor (which is fine), which is not supposed to be using the new CLA (and this is not fine). As a quick workaround, you may remove the "-u" option. However, something that would help me debugging: is this error new to 1.62? Thanks, Raffi
On Fri, Sep 2, 2016 at 11:46 PM, Raffi Enficiaud
Le 02/09/16 à 17:48, Andrey Semashev a écrit :
What bugs me there is that the failing test is using the deprecated test_execution_monitor (which is fine), which is not supposed to be using the new CLA (and this is not fine).
As a quick workaround, you may remove the "-u" option.
Thanks. It looks like the test is passing without it, so I'll settle with that.
However, something that would help me debugging: is this error new to 1.62?
Frankly, I don't know. I only stumbled upon this because I ran Boost.Utility tests to verify my changes to string_view. Judging by git history, the command line option has been there from the start, since 2006. I probably ran the tests before, but it was long time ago (6+ months maybe) and I don't think I've seen this error before.
Am 02.09.2016 um 22:46 schrieb Raffi Enficiaud:
As a quick workaround, you may remove the "-u" option. However, something that would help me debugging: is this error new to 1.62?
It's been there since at least 1.57, and probably even 1.55 but that's a long time ago and I don't have any logs around from those first testsuite run attempts of mine. Ciao Dani
Am 03.09.2016 um 06:27 schrieb Daniela Engert:
Am 02.09.2016 um 22:46 schrieb Raffi Enficiaud:
As a quick workaround, you may remove the "-u" option. However, something that would help me debugging: is this error new to 1.62?
It's been there since at least 1.57, and probably even 1.55 but that's a long time ago and I don't have any logs around from those first testsuite run attempts of mine.
I reran 'utility' unittests from 1.55 and 1.57 with msvc-12 and msvc-14. My memory served me wrong: both versions are fine.
participants (3)
-
Andrey Semashev -
Daniela Engert -
Raffi Enficiaud