OS X Problems After last Git pull on 2017-03-31

[dlhenke]

I compile every day if there is some modifications on source.
but today i have some problems..

[ 12%] Building CXX object src/Gui/CMakeFiles/FreeCADGui.dir/ActionFunction.cpp.o
/Users/duhenke/FreeCAD/source/src/Gui/ActionFunction.cpp:124:23: error: implicit instantiation of undefined template 'QPointer<QObject>'
QPointer<QObject> argQObject;
^
/usr/local/Cellar/qt@5.6/5.6.2/include/QtCore/qmetatype.h:1333:1: note: template is declared here
QT_FOR_EACH_AUTOMATIC_TEMPLATE_SMART_POINTER(QT_FORWARD_DECLARE_SHARED_POINTER_TYPES_ITER)
^
/usr/local/Cellar/qt@5.6/5.6.2/include/QtCore/qmetatype.h:209:7: note: expanded from macro 'QT_FOR_EACH_AUTOMATIC_TEMPLATE_SMART_POINTER'
F(QPointer)
^
1 error generated.
make[2]: *** [src/Gui/CMakeFiles/FreeCADGui.dir/ActionFunction.cpp.o] Error 1
make[1]: *** [src/Gui/CMakeFiles/FreeCADGui.dir/all] Error 2
make: *** [all] Error 2
Could someone help to fix ?

OS: Mac OS X
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.17.10664 (Git)
Build type: Release
Branch: master
Hash: 784edd3f82e272350b340af76d843603a78550c3
Python version: 2.7.13
Qt version: 5.6.2
Coin version: 4.0.0a
OCC version: 7.1.0

[ian.rees]

dlhenke - the issue you're seeing is due to commit shown below. Going forward, it's better to refer to commit hashes rather than times - there are often multiple commits per day, and then there are some of us on the other side of the date line, etc.

Code: Select all

...
Ians-MacBook-Pro:freecad-code irees$ git bisect good
47847513a85ff6615774ef628230f79e37471daf is the first bad commit
commit 47847513a85ff6615774ef628230f79e37471daf
...

At any rate, I imagine Werner is aware of the situation ( https://travis-ci.org/FreeCAD/FreeCAD/branches ), and it looks like a straightforward fix. To get your build working again, just check out the previous commit. Expect this sort of thing to happen occasionally, this is the tip of development after all! -Ian-

[dlhenke]

dlhenke - the issue you're seeing is due to commit shown below. Going forward, it's better to refer to commit hashes rather than times - there are often multiple commits per day, and then there are some of us on the other side of the date line, etc.

Code: Select all

...
Ians-MacBook-Pro:freecad-code irees$ git bisect good
47847513a85ff6615774ef628230f79e37471daf is the first bad commit
commit 47847513a85ff6615774ef628230f79e37471daf
...

At any rate, I imagine Werner is aware of the situation ( https://travis-ci.org/FreeCAD/FreeCAD/branches ), and it looks like a straightforward fix. To get your build working again, just check out the previous commit. Expect this sort of thing to happen occasionally, this is the tip of development after all! -Ian-

Ok thank you Ian, sorry i have no experience with git and others things.
But besides from my Old Age ( 60 ) will try to learn .

[dlhenke]

Thank you .. the compilation was successful now.

[ian.rees]

Ok thank you Ian, sorry i have no experience with git and others things.
But besides from my Old Age ( 60 ) will try to learn .

No need to apologise - we all have to start somewhere no matter what age. There are plenty younger people who don't understand this stuff, and plenty older who do. -Ian-

[simonvanderveldt]

Expect this sort of thing to happen occasionally, this is the tip of development after all! -Ian-

These kinds of breakages can easily be prevented by not directly committing to master and just doing a PR with the included "free" validation/checks as performed by CI.
Do you know if there's a specific reason that's not happening?

[ian.rees]

If given the option I'd certainly vote for maintaining Werner's (huge) throughput over an increased chance that the latest master will compile cleanly. -Ian-