[vtk-developers] [ANNOUNCE] New module system landing

[Sean McBride]

On Wed, 23 Jan 2019 14:58:29 -0500, Ben Boeckel said:

>On Wed, Jan 23, 2019 at 14:42:07 -0500, Sean McBride wrote:
>> And is it your intention to not bring back this ignoring?
>
>If there are warnings we cannot otherwise fix, yes.
>
>> Getting all the many third party libs as warning-clean as VTK proper
>> is a pretty Herculean task.  Meanwhile, all those third party warnings
>> stop us from seeing VTK warnings.  If no one is doing the former, why
>> should we lose the latter?
>
>Looking at the warnings for one of the rogue machines:
>
>    https://open.cdash.org/viewBuildError.php?type=1&buildid=5719051
>
>it looks like, we just need to ignore the `[0-9]* warnings? generated`
>regex to exclude those that are matched yet. 

So for example:

[CTest: warning suppressed] /Users/builder/external/VTK/ThirdParty/libproj/vtklibproj/src/projects.h:609:25: warning: this function declaration is not a prototype [-Wstrict-prototypes]
void pj_deallocate_grids();
                        ^
                         void
[CTest: warning matched] 2 warnings generated.

I hadn't noticed the "CTest: warning suppressed/matched" part... you're saying it's the 'matched' part that's making all the rest appear too?

>Does adding that to
>`CMake/CTestCustom.cmake.in` (in the general section), help with the
>output?

I'll try an Experimental submission...

Sean