[Development] Header diff for QtQml

[Alan Alpert]

On Wed, Jun 26, 2013 at 5:28 PM, Thiago Macieira
<thiago.macieira at intel.com> wrote:
> On quarta-feira, 26 de junho de 2013 17.23.56, Alan Alpert wrote:
>> Also qqml.h now includes qdebug.h. While not harmful, do we have a
>> convention of not including that in public headers?
>
> Does it have a reason for the include? For example, does it declare debugging
> operators for anything? If not, it's probably accidental and can be safely
> removed.

It uses qWarning() in the definition of a new inline function (the url
form of qmlRegisterType). Sounds like it's okay to leave it there,
although the function could be un-inlined if the include is a problem.

> Removing includes is *technically* a source-incompatible change, but one we
> tolerate. glibc and GCC have been cleaning up their headers for the past half
> a dozen releases; every upgrade turns up missing includes that people relied
> on.

The include is new in 5.1. Although you might just be warning me to be
careful about what includes get in in the first place.

--
Alan Alpert

On Wed, Jun 26, 2013 at 5:28 PM, Thiago Macieira
<thiago.macieira at intel.com> wrote:
> On quarta-feira, 26 de junho de 2013 17.23.56, Alan Alpert wrote:
>> Also qqml.h now includes qdebug.h. While not harmful, do we have a
>> convention of not including that in public headers?
>
> Does it have a reason for the include? For example, does it declare debugging
> operators for anything? If not, it's probably accidental and can be safely
> removed.
>
> Removing includes is *technically* a source-incompatible change, but one we
> tolerate. glibc and GCC have been cleaning up their headers for the past half
> a dozen releases; every upgrade turns up missing includes that people relied
> on.
>
> --
> Thiago Macieira - thiago.macieira (AT) intel.com
>   Software Architect - Intel Open Source Technology Center
>
> _______________________________________________
> Development mailing list
> Development at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/development
>