Re: [boost] [testing][regression] Heads up: compilation warnings are available for passing tests.

Subject: Re: [boost] [testing][regression] Heads up: compilation warnings are available for passing tests.
From: Andrzej Krzemienski (akrzemi1_at_[hidden])
Date: 2015-06-08 03:23:15

2015-06-05 20:34 GMT+02:00 Adam Wulkiewicz <adam.wulkiewicz_at_[hidden]>:

> 2015-06-05 15:28 GMT+02:00 Andrzej Krzemienski <akrzemi1_at_[hidden]>:
>
> > 2015-06-05 0:14 GMT+02:00 Adam Wulkiewicz <adam.wulkiewicz_at_[hidden]>:
> >
> > > 2015-06-03 17:27 GMT+02:00 Andrzej Krzemienski <akrzemi1_at_[hidden]>:
> > >
> > > > 2015-06-03 14:26 GMT+02:00 Adam Wulkiewicz <
> adam.wulkiewicz_at_[hidden]
> > >:
> > > >
> > > > > Hi,
> > > > >
> > > > > For passing tests which are expected to pass, if warnings are
> > generated
> > > > > during the compilation they can be seen after clicking on the
> "pass"
> > > > link.
> > > > > If the link isn't there it means that there were no warnings. I
> hope
> > it
> > > > > will allow us to improve the quality of the libraries even more.
> > > > >
> > > >
> > > > This is really good.
> > > > How about changing the tint of the green color a bit for these cases?
> > > >
> > >
> > > Initially I thought that it might be distracting. But indeed it's more
> > > clear this way. So it's done.
> > >
> > > Regards,
> > > Adam
> > >
> >
> > Adam,
> > Would it be possible to keep the passing tests with warnings green, and
> > only *slightly* change its tint? Currently they are white, and it sends a
> > strong signal that something is wrong with all these tests. I tried to
> > follow the links, but the majority of warnings are silly, buggy,
> > false-positive, or inadequate. For instance "warning: rvalue references
> are
> > a C++11 extension" I never intend to fix them, and at the same time, I do
> > want to send the message that the library is (or becomes) bug-free with
> the
> > help of green color.
> >
> > Sorry for changing my mind like that.
> >
>
> The warnings should be marked with yelowish green color. You were probably
> seing the tests for master baranch. For a reason unknown to me more time is
> needed to update the CSS file for master regression results. Please see it
> again.
>
> Is the color ok for you? Or is it too different from the lightgreen used
> for passing tests?
> You could also create a PR in Regression setting the color in the CSS file.
>

Ok, after F5 in my browser, I can see the green color. It looks perfect!
Thank you for doing it, I have already found one genuine warning in my
library.

Regards,
&rzej