PyQt4-4.7 fails to compile while creating pyqtconfig.py . It reports : "cannot access w_qpy*.pro: No such file or directory" & "can't read QtDesigner/Makefile: No such file or directory". I will attach the requested files separately. Reproducible: Always Steps to Reproduce: 1. 'emerge PyQt' 2. wait briefly 3. read error msgs Actual Results: emerge failed Expected Results: emerge should succeed
Created attachment 222351 [details] result of 'emerge --info =dev-python/PyQt4-4.7'
Created attachment 222353 [details] result of 'emerge -pqv =dev-python/PyQt4-4.7'
Created attachment 222355 [details] build log
Created attachment 222359 [details] ebuild environment
The problem seems to be that while '--enable=QtGui --enable=QtDesigner' is among configure options, configure tests decide that those modules should not be built. Could you check why ?
I hope you don't mean me (smile): I have no idea how to check into configure options or tests. Hopefully, there is a dev who can help.
(In reply to comment #6) > I hope you don't mean me (smile): I have no idea how to check into configure > options or tests. Hopefully, there is a dev who can help. > No, I meant the other guy with your problem... Of course I mean you - it fails for you, so you're the only one who can tell, what's your problem is, otherwise this will head straight into WORKSFORME. It seems you need to pass '--verbose' to configure.py options, at least that much I can tell looking at PyQt4 docs.
So just tell me how to do that & I'll test it for you. Where are "configure.py options" & how do I pass '--verbose' to them ? What I see is : root:567 ~> emerge -pv PyQt4 ... [ebuild U ] dev-python/PyQt4-4.7 [4.6.2] USE="X -assistant dbus -debug -doc -examples kde -multimedia% opengl -phonon sql svg webkit -xmlpatterns" I don't see anything resembling "--verbose".
I looked into the .../work/... dirs & ran 'python2.6 configure.py --verbose' (there are 2 subdirs & I did the same in each with the same result). Then I ran 'make' & eventually reached an error (same in each subdir): I will attach the final lines of the make output. In Comment 5 , you refer to the configure tests for QtGui QtDesigner . the .cpp test pgms are very short : cfgtest_QtDesigner.cpp : #include <QExtensionFactory> int main(int, char **) { new QExtensionFactory(); } cfgtest_QtGui.cpp : #include <qwidget.h> int main(int, char **) { new QWidget(); } Please remember that I'm a user, not a developer: I can investigate & test as much as you need, but you have to give me fairly precise instructions.
Created attachment 222495 [details] final lines of output of 'make'
I don't consider myself a programmer either. What's the output of 'emerge -1pv qt-core qt-gui' ?
root:590 ~> emerge -1pv qt-core qt-gui These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] x11-libs/qt-core-4.6.2-r1 USE="(-aqua) -debug -doc -exceptions* -glib -iconv -optimized-qmake -pch qt3support ssl" 0 kB [ebuild R ] x11-libs/qt-gui-4.6.2 USE="accessibility (-aqua) cups dbus -debug -exceptions -glib* gtk -mng -nas -nis -pch qt3support -raster tiff -xinerama" 0 kB
(In reply to comment #12) > [ebuild R ] x11-libs/qt-core-4.6.2-r1 USE="(-aqua) -debug -doc > -exceptions* -glib -iconv -optimized-qmake -pch qt3support ssl" 0 kB > [ebuild R ] x11-libs/qt-gui-4.6.2 USE="accessibility (-aqua) cups dbus > -debug -exceptions -glib* gtk -mng -nas -nis -pch qt3support -raster tiff > -xinerama" 0 kB > Me not likey that output - seems you've managed somehow to circumvent the requirement that 'glib' useflag setting in qt-gui should match one for qt-core. Same for exceptions. Also, did you have any *good* reason to disable 'exceptions' in the first place. I'd rebuild all of qt splits to make sure nothing else is messed up.
Philip, it appears that you have a non-working QtGui library... somehow you managed to screw up your Qt installation :P Please re-emerge all x11-libs/qt-* packages you have currently installed.
Thanks to both commentators! I remerged all 11 Qt pkgs with USE="exceptions glib" & PyQt4 & kdelibs have compiled successfully. This bug can be closed & I will comment separately on the other bug 308113 , which you can check, if you're still interested.