[Buildroot] [autobuild.buildroot.net] Your build results for 2017-02-09

[Buildroot] [autobuild.buildroot.net] Your build results for 2017-02-09

[Thomas Petazzoni]

Hello,

On Fri, 10 Feb 2017 09:52:09 +0000, Naumann Andreas wrote:

> I've actually sent a patch for this earlier in '[Buildroot] [PATCH v4
> 10/10] poppler: Compile with -std=c++11'. Unfortunately I forgot it
> in v5. Is this enough or do you want me to resend it?

OK, I found it again in patchwork:

  http://patchwork.ozlabs.org/patch/708239/

But this patch has an issue: if a package needs C++11 support then it
needs to have a BR2_TOOLCHAIN_GCC_AT_LEAST_XYZ dependency.

But this raises an even more fundamental issue: there is *nothing* in
poppler that indicates it can depend on Qt5. There is some logic for
Qt4, but nothing for Qt5.

So: the dependency on Qt5 should be made explicit, and should take care
of the fact that Qt5 needs C++11 support, and therefore a reasonably
recent C++ compiler.

Could you work on this?

Alternatively, if you don't care/use Qt5 support in Poppler, you can
also just pass the appropriate --disable-<foo> option to
unconditionally disable Qt5 support.

Thanks!

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[Buildroot] [autobuild.buildroot.net] Your build results for 2017-02-09

[Andreas Naumann]

Hi,

Am 10.02.2017 um 11:26 schrieb Thomas Petazzoni:
> Hello,
>
> On Fri, 10 Feb 2017 09:52:09 +0000, Naumann Andreas wrote:
>
>> I've actually sent a patch for this earlier in '[Buildroot] [PATCH v4
>> 10/10] poppler: Compile with -std=c++11'. Unfortunately I forgot it
>> in v5. Is this enough or do you want me to resend it?
>
> OK, I found it again in patchwork:
>
>   http://patchwork.ozlabs.org/patch/708239/
>
> But this patch has an issue: if a package needs C++11 support then it
> needs to have a BR2_TOOLCHAIN_GCC_AT_LEAST_XYZ dependency.
>
> But this raises an even more fundamental issue: there is *nothing* in
> poppler that indicates it can depend on Qt5. There is some logic for
> Qt4, but nothing for Qt5.

Ok, I see that we have some local patch that makes this explicit. I 
guess in the given failure it must have been by chance that qt5 was 
built before poppler and popplers configure automatically detected this.

>
> So: the dependency on Qt5 should be made explicit, and should take care
> of the fact that Qt5 needs C++11 support, and therefore a reasonably
> recent C++ compiler.
>
> Could you work on this?
>
> Alternatively, if you don't care/use Qt5 support in Poppler, you can
> also just pass the appropriate --disable-<foo> option to
> unconditionally disable Qt5 support.

Quite the opposite. I'll try to come up with something in the afternoon.


regards,
Andreas

>
> Thanks!
>
> Thomas
>

[Buildroot] [autobuild.buildroot.net] Your build results for 2017-02-09

[Thomas Petazzoni]

Hello,

On Fri, 10 Feb 2017 11:47:03 +0100, Andreas Naumann wrote:

> > But this raises an even more fundamental issue: there is *nothing* in
> > poppler that indicates it can depend on Qt5. There is some logic for
> > Qt4, but nothing for Qt5.  
> 
> Ok, I see that we have some local patch that makes this explicit. I 
> guess in the given failure it must have been by chance that qt5 was 
> built before poppler and popplers configure automatically detected this.

Exactly.

> > So: the dependency on Qt5 should be made explicit, and should take care
> > of the fact that Qt5 needs C++11 support, and therefore a reasonably
> > recent C++ compiler.
> >
> > Could you work on this?
> >
> > Alternatively, if you don't care/use Qt5 support in Poppler, you can
> > also just pass the appropriate --disable-<foo> option to
> > unconditionally disable Qt5 support.  
> 
> Quite the opposite. I'll try to come up with something in the afternoon.

Excellent, thanks!

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com