Re: [Development] rebuild Qt 5 on Mac OS

2012-09-22 Thread Thiago Macieira
On sábado, 22 de setembro de 2012 15.55.58, Chris Meyer wrote: > make[4]: *** No rule to make target > `inspector/InspectorWebGLAgent.cpp', needed by > `obj/debug/inspector/InspectorAllInOne.o'. Stop. > make[3]: *** [debug-all] Error 2 > make[2]: *** [sub-Target-pri-make_first-ordered] Error 2 > m

[Development] rebuild Qt 5 on Mac OS

2012-09-22 Thread Chris Meyer
I'm rebuilding Qt 5 from source on Mac OS. I did git pull git submodule update --recursive make I get an error: make[4]: *** No rule to make target `inspector/InspectorWebGLAgent.cpp', needed by `obj/debug/inspector/InspectorAllInOne.o'. Stop. make[3]: *** [debug-all] Error 2 make[2]: *** [sub

Re: [Development] [Releasing] Beta 2 release

2012-09-22 Thread Jiang Jiang
Hi, On Thu, Sep 20, 2012 at 9:05 PM, Stephen Chu wrote: > And also this one for consideration: > At least this issue is fixed by 75cd2dec71d81868e1c4d5df32eb602faeb165fc. Looks like "automatic mouse grabbing" is the culprit of many widget b

Re: [Development] Any plans on updating harfbuzz?

2012-09-22 Thread Konstantin Ritt
See https://bugreports.qt-project.org/browse/QTBUG-18980 The work is in progress and would probably be targetted to 5.1 or 5.2. Konstantin 2012/9/22 Mark : > On Sat, Sep 22, 2012 at 3:03 PM, Mark wrote: >> Hi, >> >> I actually came across this one by accident. I was playing with >> qtperf4 [1]

Re: [Development] Any plans on updating harfbuzz?

2012-09-22 Thread Mark
On Sat, Sep 22, 2012 at 3:03 PM, Mark wrote: > Hi, > > I actually came across this one by accident. I was playing with > qtperf4 [1] and was profiling the "QTextEdit - add text" since that > seems a bit slow and equally slow on raster and native. So, after > profiling it turns out that the vast ma

[Development] Any plans on updating harfbuzz?

2012-09-22 Thread Mark
Hi, I actually came across this one by accident. I was playing with qtperf4 [1] and was profiling the "QTextEdit - add text" since that seems a bit slow and equally slow on raster and native. So, after profiling it turns out that the vast majority of that benchmark is spending it's time in qShapeI