From chgans at gmail.com Fri Sep 1 08:37:37 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 1 Sep 2017 18:37:37 +1200 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) Message-ID: Hi there, Back on my 32 bits build of Qt I'm having a build failure on tst_qatomicinteger for qlonglong. Static failure: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); Sounds very much like a 32/64 bits issue. Any idea on what is going on? Thanks in advance, Chris -------------------- KUbuntu 17.04, gcc (Ubuntu 6.3.0-12ubuntu2) 6.3.0 20170406 $ git show commit 4b5af2de682541e0077298f22a05f36f80784151 Author: Qt Submodule Update Bot Date: Tue Apr 11 16:03:07 2017 +0200 Update submodules on '5.9' in qt5 Change-Id: Id6c112e68c465ae52d49f4b35bee648894a35fef Reviewed-by: Liang Qi -------------------- cd qlonglong/ && ( test -e Makefile || /home/krys/Projects/qt5/qtbase/bin/qmake -o Makefile /home/krys/Projects/qt5/qtbase/tests/auto/corelib/thread/qatomicinteger/qlonglong/qlonglong.pro ) && make -f Makefile make[7]: Entering directory '/home/krys/Projects/qt5/qtbase/tests/auto/corelib/thread/qatomicinteger/qlonglong' g++ -c -m32 -pipe -O2 -Wall -W -D_REENTRANT -fPIC -DQATOMIC_TEST_TYPE=qlonglong -Dtst_QAtomicIntegerXX=tst_QAtomicInteger_qlonglong -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_NO_DEBUG -DQT_TESTLIB_LIB -DQT_CORE_LIB -DQT_TESTCASE_BUILDDIR='"/home/krys/Projects/qt5/qtbase/tests/auto/corelib/thread/qatomicinteger/qlonglong"' -I. -I../../../../../../include -I../../../../../../include/QtTest -I../../../../../../include/QtCore -I.moc -I../../../../../../mkspecs/linux-g++-32 -o .obj/tst_qatomicinteger.o ../tst_qatomicinteger.cpp In file included from ../../../../../../include/QtCore/qglobal.h:1:0, from ../../../../../../include/QtCore/QtCore:4, from ../../../../../../include/QtTest/QtTestDepends:3, from ../../../../../../include/QtTest/QtTest:3, from ../tst_qatomicinteger.cpp:53: ../tst_qatomicinteger.cpp: In member function ‘void tst_QAtomicInteger_qlonglong::static_checks()’: ../../../../../../include/QtCore/../../src/corelib/global/qglobal.h:732:36: error: static assertion failed: Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct) #define Q_STATIC_ASSERT(Condition) static_assert(bool(Condition), #Condition) ^ ../tst_qatomicinteger.cpp:193:5: note: in expansion of macro ‘Q_STATIC_ASSERT’ Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); ^~~~~~~~~~~~~~~ Makefile:1495: recipe for target '.obj/tst_qatomicinteger.o' failed -------------------- From chgans at gmail.com Fri Sep 1 09:21:43 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 1 Sep 2017 19:21:43 +1200 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: References: Message-ID: On 1 September 2017 at 18:37, Christian Gagneraud wrote: > Hi there, > > Back on my 32 bits build of Qt > > I'm having a build failure on tst_qatomicinteger for qlonglong. > > Static failure: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == > Q_ALIGNOF(TypeInStruct)); > > Sounds very much like a 32/64 bits issue. Actually i have more test build failure: 2 about qlonglong and qulonglong, and a bunch of other that look like this (simple) one: g++ -m32 -Wl,-O1 -fuse-ld=gold -Wl,--enable-new-dtags -Wl,-rpath,/home/krys/Projects/qt5/qtbase/lib -o tst_qline .obj/tst_qline.o -lm -L/home/krys/Projects/qt5/qtbase/lib -lQt5Test -lQt5Core -lpthread .obj/tst_qline.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status Should I disabled the gold linker? I've heard of problems with it while cross building Qt... I am able to execute some test suites, eg tst_qvariant yields: Totals: 131 passed, 0 failed, 0 skipped, 0 blacklisted, 40822ms Full list of build failures: $ make -j10 -k 1> /dev/null .obj/tst_qmouseevent.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qmouseevent] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qmouseevent-make_first] Error 2 .obj/tst_qwindow.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qwindow] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qwindow-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-kernel-make_first] Error 2 .obj/tst_qbrush.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qbrush] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qbrush-make_first] Error 2 .obj/tst_qcssparser.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qcssparser] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qcssparser-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-painting-make_first] Error 2 .obj/tst_qtextlayout.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qtextlayout] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qtextlayout-make_first] Error 2 .obj/tst_gestures.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[5]: *** [tst_gestures] Error 1 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-gestures-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-text-make_first] Error 2 make[4]: Target 'first' not remade because of errors. make[3]: *** [sub-gui-make_first] Error 2 .obj/tst_qpixmapfilter.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qpixmapfilter] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qpixmapfilter-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-effects-make_first] Error 2 make[4]: Target 'first' not remade because of errors. make[3]: *** [sub-other-make_first] Error 2 .obj/tst_qgraphicsobject.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qgraphicsobject] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qgraphicsobject-make_first] Error 2 .obj/tst_qgraphicspixmapitem.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qgraphicspixmapitem] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qgraphicspixmapitem-make_first] Error 2 .obj/tst_qgraphicstransform.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qgraphicstransform] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qgraphicstransform-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-graphicsview-make_first] Error 2 .obj/tst_qline.o(.qtversion+0x0): error: unexpected reloc 3 against global symbol qt_version_tag without base register in object file when generating a position-independent output file collect2: error: ld returned 1 exit status make[6]: *** [tst_qline] Error 1 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qline-make_first] Error 2 In file included from ../../../../../../include/QtCore/qglobal.h:1:0, from ../../../../../../include/QtCore/QtCore:4, from ../../../../../../include/QtTest/QtTestDepends:3, from ../../../../../../include/QtTest/QtTest:3, from ../tst_qatomicinteger.cpp:53: ../tst_qatomicinteger.cpp: In member function ‘void tst_QAtomicInteger_qlonglong::static_checks()’: ../../../../../../include/QtCore/../../src/corelib/global/qglobal.h:732:36: error: static assertion failed: Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct) #define Q_STATIC_ASSERT(Condition) static_assert(bool(Condition), #Condition) ^ ../tst_qatomicinteger.cpp:193:5: note: in expansion of macro ‘Q_STATIC_ASSERT’ Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); ^~~~~~~~~~~~~~~ In file included from ../../../../../../include/QtCore/qglobal.h:1:0, from ../../../../../../include/QtCore/QtCore:4, from ../../../../../../include/QtTest/QtTestDepends:3, from ../../../../../../include/QtTest/QtTest:3, from ../tst_qatomicinteger.cpp:53: ../tst_qatomicinteger.cpp: In member function ‘void tst_QAtomicInteger_qulonglong::static_checks()’: ../../../../../../include/QtCore/../../src/corelib/global/qglobal.h:732:36: error: static assertion failed: Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct) #define Q_STATIC_ASSERT(Condition) static_assert(bool(Condition), #Condition) ^ ../tst_qatomicinteger.cpp:193:5: note: in expansion of macro ‘Q_STATIC_ASSERT’ Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); ^~~~~~~~~~~~~~~ make[4]: Target 'first' not remade because of errors. make[3]: *** [sub-widgets-make_first] Error 2 make[7]: *** [.obj/tst_qatomicinteger.o] Error 1 make[7]: Target 'first' not remade because of errors. make[6]: *** [sub-qlonglong-make_first] Error 2 make[7]: *** [.obj/tst_qatomicinteger.o] Error 1 make[7]: Target 'first' not remade because of errors. make[6]: *** [sub-qulonglong-make_first] Error 2 make[6]: Target 'first' not remade because of errors. make[5]: *** [sub-qatomicinteger-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-thread-make_first] Error 2 make[5]: Target 'first' not remade because of errors. make[4]: *** [sub-tools-make_first] Error 2 make[4]: Target 'first' not remade because of errors. make[3]: *** [sub-corelib-make_first] Error 2 make[3]: Target 'first' not remade because of errors. make[2]: *** [sub-auto-make_first] Error 2 make[2]: Target 'first' not remade because of errors. make[1]: *** [sub-tests-make_first] Error 2 make[1]: Target 'first' not remade because of errors. make: *** [module-qtbase-make_first] Error 2 make: Target 'first' not remade because of errors. $ Chris > > Any idea on what is going on? > > Thanks in advance, > Chris > > -------------------- > KUbuntu 17.04, gcc (Ubuntu 6.3.0-12ubuntu2) 6.3.0 20170406 > $ git show > commit 4b5af2de682541e0077298f22a05f36f80784151 > Author: Qt Submodule Update Bot > Date: Tue Apr 11 16:03:07 2017 +0200 > > Update submodules on '5.9' in qt5 > > Change-Id: Id6c112e68c465ae52d49f4b35bee648894a35fef > Reviewed-by: Liang Qi > -------------------- > cd qlonglong/ && ( test -e Makefile || > /home/krys/Projects/qt5/qtbase/bin/qmake -o Makefile > /home/krys/Projects/qt5/qtbase/tests/auto/corelib/thread/qatomicinteger/qlonglong/qlonglong.pro > ) && make -f Makefile > make[7]: Entering directory > '/home/krys/Projects/qt5/qtbase/tests/auto/corelib/thread/qatomicinteger/qlonglong' > g++ -c -m32 -pipe -O2 -Wall -W -D_REENTRANT -fPIC > -DQATOMIC_TEST_TYPE=qlonglong > -Dtst_QAtomicIntegerXX=tst_QAtomicInteger_qlonglong > -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_NO_DEBUG > -DQT_TESTLIB_LIB -DQT_CORE_LIB > -DQT_TESTCASE_BUILDDIR='"/home/krys/Projects/qt5/qtbase/tests/auto/corelib/thread/qatomicinteger/qlonglong"' > -I. -I../../../../../../include -I../../../../../../include/QtTest > -I../../../../../../include/QtCore -I.moc > -I../../../../../../mkspecs/linux-g++-32 -o .obj/tst_qatomicinteger.o > ../tst_qatomicinteger.cpp > In file included from ../../../../../../include/QtCore/qglobal.h:1:0, > from ../../../../../../include/QtCore/QtCore:4, > from ../../../../../../include/QtTest/QtTestDepends:3, > from ../../../../../../include/QtTest/QtTest:3, > from ../tst_qatomicinteger.cpp:53: > ../tst_qatomicinteger.cpp: In member function ‘void > tst_QAtomicInteger_qlonglong::static_checks()’: > ../../../../../../include/QtCore/../../src/corelib/global/qglobal.h:732:36: > error: static assertion failed: Q_ALIGNOF(QAtomicInteger) == > Q_ALIGNOF(TypeInStruct) > #define Q_STATIC_ASSERT(Condition) static_assert(bool(Condition), #Condition) > ^ > ../tst_qatomicinteger.cpp:193:5: note: in expansion of macro ‘Q_STATIC_ASSERT’ > Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); > ^~~~~~~~~~~~~~~ > Makefile:1495: recipe for target '.obj/tst_qatomicinteger.o' failed > -------------------- From chgans at gmail.com Fri Sep 1 10:10:09 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 1 Sep 2017 20:10:09 +1200 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: References: Message-ID: On 1 September 2017 at 19:21, Christian Gagneraud wrote: > On 1 September 2017 at 18:37, Christian Gagneraud wrote: >> Hi there, >> >> Back on my 32 bits build of Qt >> >> I'm having a build failure on tst_qatomicinteger for qlonglong. >> >> Static failure: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == >> Q_ALIGNOF(TypeInStruct)); >> >> Sounds very much like a 32/64 bits issue. > > Actually i have more test build failure: > > 2 about qlonglong and qulonglong, and a bunch of other that look like > this (simple) one: > > g++ -m32 -Wl,-O1 -fuse-ld=gold -Wl,--enable-new-dtags > -Wl,-rpath,/home/krys/Projects/qt5/qtbase/lib -o tst_qline > .obj/tst_qline.o -lm -L/home/krys/Projects/qt5/qtbase/lib -lQt5Test > -lQt5Core -lpthread > .obj/tst_qline.o(.qtversion+0x0): error: unexpected reloc 3 against > global symbol qt_version_tag without base register in object file when > generating a position-independent output file > collect2: error: ld returned 1 exit status > > Should I disabled the gold linker? I've heard of problems with it > while cross building Qt... OK, so disabling the gold linker fixed the weird symbols issues. So I'm down to only 2 build failure, static asserts about alignment of QAtomicInteger using qlonglong and qulonglong Looking at tst_qatomicinteger.cpp I'm passing the first static assert: Q_STATIC_ASSERT(sizeof(QAtomicInteger) == sizeof(T)); But fail the second one: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); TypeInStruct is, as the name suggest, as simple as a variable of the tested type inside a struct: struct TypeInStruct { TEST_TYPE type; }; So QAtomicInteger has the same size as qlonglong, but not the same alignment!?! Looking at qglobal.h, i've found this interesting piece of code: // x86 ABI weirdness // Alignment of naked type is 8, but inside struct has alignment 4. template <> struct AlignOf : AlignOf_WorkaroundForI386Abi {}; template <> struct AlignOf : AlignOf_WorkaroundForI386Abi {}; template <> struct AlignOf : AlignOf_WorkaroundForI386Abi {}; And my Qt configure says: [...] Checking for target architecture... i386s [...] Checking for 64 bit atomics... yes [...] So does that mean that QAtomicInteger is broken or cannot be used on linux-32 ? Which by the way is not supported anymore since Qt-5.6... Any insight, comment or suggestion on how to get that fixed are very welcome! ;) Thanks, Chris From cmakshin at gmail.com Fri Sep 1 17:10:18 2017 From: cmakshin at gmail.com (Constantin Makshin) Date: Fri, 1 Sep 2017 18:10:18 +0300 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: References: Message-ID: <56491f3e-7c4b-7310-c556-4db8f12eeaaa@gmail.com> Hi Christian. May be changing template class QBasicAtomicInteger in "src/corelib/thread/qbasicatomic.h" to template class Q_DECL_ALIGN(Q_ALIGNOF(typename QAtomicOps::Type)) QBasicAtomicInteger will fix the problem. On 09/01/2017 11:10 AM, Christian Gagneraud wrote: > On 1 September 2017 at 19:21, Christian Gagneraud wrote: >> On 1 September 2017 at 18:37, Christian Gagneraud wrote: >>> Hi there, >>> >>> Back on my 32 bits build of Qt >>> >>> I'm having a build failure on tst_qatomicinteger for qlonglong. >>> >>> Static failure: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == >>> Q_ALIGNOF(TypeInStruct)); >>> >>> Sounds very much like a 32/64 bits issue. >> >> Actually i have more test build failure: >> >> 2 about qlonglong and qulonglong, and a bunch of other that look like >> this (simple) one: >> >> g++ -m32 -Wl,-O1 -fuse-ld=gold -Wl,--enable-new-dtags >> -Wl,-rpath,/home/krys/Projects/qt5/qtbase/lib -o tst_qline >> .obj/tst_qline.o -lm -L/home/krys/Projects/qt5/qtbase/lib -lQt5Test >> -lQt5Core -lpthread >> .obj/tst_qline.o(.qtversion+0x0): error: unexpected reloc 3 against >> global symbol qt_version_tag without base register in object file when >> generating a position-independent output file >> collect2: error: ld returned 1 exit status >> >> Should I disabled the gold linker? I've heard of problems with it >> while cross building Qt... > > OK, so disabling the gold linker fixed the weird symbols issues. > > So I'm down to only 2 build failure, static asserts about alignment of > QAtomicInteger using qlonglong and qulonglong > > Looking at tst_qatomicinteger.cpp > > I'm passing the first static assert: > Q_STATIC_ASSERT(sizeof(QAtomicInteger) == sizeof(T)); > > But fail the second one: > Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == Q_ALIGNOF(TypeInStruct)); > > TypeInStruct is, as the name suggest, as simple as a variable of the > tested type inside a struct: > struct TypeInStruct { TEST_TYPE type; }; > > So QAtomicInteger has the same size as qlonglong, but not > the same alignment!?! > > Looking at qglobal.h, i've found this interesting piece of code: > // x86 ABI weirdness > // Alignment of naked type is 8, but inside struct has alignment 4. > template <> struct AlignOf : > AlignOf_WorkaroundForI386Abi {}; > template <> struct AlignOf : > AlignOf_WorkaroundForI386Abi {}; > template <> struct AlignOf : > AlignOf_WorkaroundForI386Abi {}; > > And my Qt configure says: > [...] > Checking for target architecture... i386s > [...] > Checking for 64 bit atomics... yes > [...] > > So does that mean that QAtomicInteger is broken or cannot be used on linux-32 ? > Which by the way is not supported anymore since Qt-5.6... > > Any insight, comment or suggestion on how to get that fixed are very welcome! ;) > > Thanks, > Chris -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 862 bytes Desc: OpenPGP digital signature URL: From konstantin at podsvirov.pro Fri Sep 1 17:28:54 2017 From: konstantin at podsvirov.pro (Konstantin Podsvirov) Date: Fri, 01 Sep 2017 18:28:54 +0300 Subject: [Interest] [VOTES REQUEST] https://bugreports.qt.io/browse/QTIFW-948 (more freedom in name and version) Message-ID: <490281504279734@web16j.yandex.ru> You are using QtIFW and you do not like rigid limitations on the name and version of packages - vote for optional freedom! Suggestion: https://bugreports.qt.io/browse/QTIFW-948 Apropriate change request: https://codereview.qt-project.org/#/c/203958 -- Regards, Konstantin Podsvirov From thiago.macieira at intel.com Fri Sep 1 18:42:29 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Fri, 1 Sep 2017 09:42:29 -0700 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: References: Message-ID: <4003809.5ffsG9a4Zf@tjmaciei-mobl1> On Thursday, 31 August 2017 23:37:37 PDT Christian Gagneraud wrote: > Hi there, > > Back on my 32 bits build of Qt > > I'm having a build failure on tst_qatomicinteger for qlonglong. > > Static failure: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == > Q_ALIGNOF(TypeInStruct)); > > Sounds very much like a 32/64 bits issue. > > Any idea on what is going on? It's a GCC bug. https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71660 -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From chgans at gmail.com Sat Sep 2 01:42:08 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Sat, 2 Sep 2017 11:42:08 +1200 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: <4003809.5ffsG9a4Zf@tjmaciei-mobl1> References: <4003809.5ffsG9a4Zf@tjmaciei-mobl1> Message-ID: On 2 September 2017 at 04:42, Thiago Macieira wrote: > On Thursday, 31 August 2017 23:37:37 PDT Christian Gagneraud wrote: >> Hi there, >> >> Back on my 32 bits build of Qt >> >> I'm having a build failure on tst_qatomicinteger for qlonglong. >> >> Static failure: Q_STATIC_ASSERT(Q_ALIGNOF(QAtomicInteger) == >> Q_ALIGNOF(TypeInStruct)); >> >> Sounds very much like a 32/64 bits issue. >> >> Any idea on what is going on? > > It's a GCC bug. https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71660 Thanks for the info. I'm still quite puzzle thought. doc.qt.io says that for Qt 5.9, the supported platform contains: Linux 32/64-bit: GCC 4.8, GCC 4.9, GCC 5.3 According to your gcc bug report, the bug was introduced in gcc 5.1 but has never been fixed (?) So if Qt supports gcc 5.3, then it supports gcc with this bug? (K)Ubuntu-17.04 ships with gcc 6.3 and Qt-5.7, surely their 32 bits OS version doesn't ship with a broken QAtomicInteger... Any clue on how I can get around this bug? Thanks, Chris > > -- > Thiago Macieira - thiago.macieira (AT) intel.com > Software Architect - Intel Open Source Technology Center > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From thiago.macieira at intel.com Sat Sep 2 02:15:48 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Fri, 01 Sep 2017 17:15:48 -0700 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: References: <4003809.5ffsG9a4Zf@tjmaciei-mobl1> Message-ID: <2017793.EJzfhdgRK7@tjmaciei-mobl1> On Friday, 1 September 2017 16:42:08 PDT Christian Gagneraud wrote: > doc.qt.io says that for Qt 5.9, the supported platform contains: > Linux 32/64-bit: GCC 4.8, GCC 4.9, GCC 5.3 > > According to your gcc bug report, the bug was introduced in gcc 5.1 > but has never been fixed (?) Correct. We just ignore the failure. There's nothing we can do about it. > So if Qt supports gcc 5.3, then it supports gcc with this bug? Yes. > (K)Ubuntu-17.04 ships with gcc 6.3 and Qt-5.7, surely their 32 bits OS > version doesn't ship with a broken QAtomicInteger... > > Any clue on how I can get around this bug? You cannot. Just ignore this failure. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From chgans at gmail.com Sat Sep 2 09:02:58 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Sat, 2 Sep 2017 19:02:58 +1200 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: <2017793.EJzfhdgRK7@tjmaciei-mobl1> References: <4003809.5ffsG9a4Zf@tjmaciei-mobl1> <2017793.EJzfhdgRK7@tjmaciei-mobl1> Message-ID: On 2 September 2017 at 12:15, Thiago Macieira wrote: > On Friday, 1 September 2017 16:42:08 PDT Christian Gagneraud wrote: >> doc.qt.io says that for Qt 5.9, the supported platform contains: >> Linux 32/64-bit: GCC 4.8, GCC 4.9, GCC 5.3 >> >> According to your gcc bug report, the bug was introduced in gcc 5.1 >> but has never been fixed (?) > > Correct. We just ignore the failure. There's nothing we can do about it. Really? How do you cope in CI? You don't build for linux-32 anymore? > >> So if Qt supports gcc 5.3, then it supports gcc with this bug? > > Yes. > >> (K)Ubuntu-17.04 ships with gcc 6.3 and Qt-5.7, surely their 32 bits OS >> version doesn't ship with a broken QAtomicInteger... >> >> Any clue on how I can get around this bug? > > You cannot. Just ignore this failure. Scary, i then cannot run the tests about Qt's atomic integer solutions! Does QAtomicInteger works safely with Qt 5.6 on a linux 32 bits box with gcc > 5.1? Should i use QAtomicInteger on 64bits and QAtomicInteger on 32 bits? If you or I cannot build and so cannot run QAtomicInteger's test suites, then how do you or how do I know it works as adverstised? Thanks, Chris > > -- > Thiago Macieira - thiago.macieira (AT) intel.com > Software Architect - Intel Open Source Technology Center > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From thiago.macieira at intel.com Sat Sep 2 09:56:37 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Sat, 02 Sep 2017 00:56:37 -0700 Subject: [Interest] tst_qatomicinteger build failure (5.9 i386/x86_64) In-Reply-To: References: <2017793.EJzfhdgRK7@tjmaciei-mobl1> Message-ID: <1775527.lTRE45a0un@tjmaciei-mobl1> On Saturday, 2 September 2017 00:02:58 PDT Christian Gagneraud wrote: > > Correct. We just ignore the failure. There's nothing we can do about it. > > Really? How do you cope in CI? You don't build for linux-32 anymore? That looks like it's the case. See https://testresults.qt.io/coin/integration/qt/qtbase/tasks/1504282591 for an example of a failed build hat shows all the architectures tested by the CI. There's a second round of slower CI that QA runs every now and then. It may contain 32-bit builds, but if so the failure from this test is just ignored. > > You cannot. Just ignore this failure. > > Scary, i then cannot run the tests about Qt's atomic integer solutions! Run for the other ones (8-, 16- and 32-bit). > Does QAtomicInteger works safely with Qt 5.6 on a linux 32 > bits box with gcc > 5.1? Define "safely". If you mean that it generates proper atomic code to access 64-bit quantities? It does. Disassembly of the code produced by GCC and Clang reveals it's correct, and if you work around the failure, the rest of the test passes. If you mean you can replace an existing qunit64 in a structure with a QAtomicInteger and keep binary compatibility, there's a 50-50% chance that it won't. This is the test that is failing. > Should i use QAtomicInteger on 64bits and > QAtomicInteger on 32 bits? Yes. There are even two helper typedefs for you: quintptr and qregisteruint. The former is the size of a pointer, the same size as qssize_t, qptrdiff, size_t and ptrdiff_t. The latter is the size of the machine's word, which is usually the same as the pointer, except for ILP32 on 64-bit machines. > If you or I cannot build and so cannot run QAtomicInteger's test > suites, then how do you or how do I know it works as adverstised? Again, define "works as advertised". See above for the two possibilities. I did test manually that the operations on a 64-bit quantity work. And I can clearly confirm that the alignment broke. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From rjvbertin at gmail.com Sat Sep 2 10:26:34 2017 From: rjvbertin at gmail.com (=?ISO-8859-1?Q?Ren=E9_J=2EV=2E?= Bertin) Date: Sat, 02 Sep 2017 10:26:34 +0200 Subject: [Interest] Using startTimer from a const member function? Message-ID: <15212682.a83FDCjNVs@bola> Hi, I'm modifying a style (QtCurve) to not keep an animation timer running at all times but only when required. This means starting it from the Style::drawControl() method, which is a const function. It was trivial to move the few Style member variables into a private class accessed via d-ptr, but the timer itself is based on QObject::startTimer() and evidently I cannot call that function from a const function. For now I've worked around this by adding a q-ptr to the private subclass, and a `int startParentTimer(int d) { q->startTimer(d); }` function but that feels like a hack. (BTW, having to do this is why I didn't use mutual for the member variables.) The other devs on the project don't seem to raise any eyebrowse over this but I'd still like to know if there isnt a more elegant way to achieve this that does not involve refactoring the whole code to use a dedicated animation class (or have the private subclass inherit QObject too)? Thanks, René From Ch.Ehrlicher at gmx.de Sun Sep 3 10:21:26 2017 From: Ch.Ehrlicher at gmx.de (Christian Ehrlicher) Date: Sun, 3 Sep 2017 10:21:26 +0200 Subject: [Interest] QFileSystemWatcher and kate Message-ID: Hi, I stumbled across a problem with QFileSystemWatcher on Linux. It looks like the inotify implementation does not honor a file replace action as expected. I'm adding a file with addPath() to watch for changes. But after I edit the file with kate (or any other editor which saves the file in a temp file and then renaming it to correct name), the QFileSystemWatcher does no longer watch for modifications... Looking in the bug tracker I also found a bug which describes the problem: https://bugreports.qt.io/browse/QTBUG-19350?jql=text%20~%20%22qfilesystemwatcher%22 So it looks like this bug exists since inotify support was added but the bug report was closed because it was not reproducible (maybe not tested with inotify). Therefore my question - how to reopen this bug or should I file a new one? Thx, Christian From MRussell at neptec.com Sun Sep 3 18:30:08 2017 From: MRussell at neptec.com (Russell, Matthew) Date: Sun, 3 Sep 2017 16:30:08 +0000 Subject: [Interest] QQuickWindow::sceneGraphInvalidated not invoked Message-ID: <8D32ABA6D779C645A4CD11339363E5FF0D3DBC@EMAIL.corp.neptec.com> Hi, Is there anything that could prevent QQuickWindow::sceneGraphInvalidated from being invoked? I've integrated the "Scene Graph - Rendering FBOs in a thread" ( http://doc.qt.io/qt-5/qtquick-scenegraph-textureinthread-example.html ) example into my code.. It's hard to post code samples, but essentially I cannot find any meaningful difference between how that example connects to sceneGraphInvalidated and how I do. That said, when I run the example in QtCreator (using Qt 5.9.1), everything works (sceneGraphInvalidated is triggered, shutdown gets called, etc), but when I run my code, the signal is never sent, and I end up (on exit) with a seg fault. Just to be clear, I'm triggering an exit by clicking the window's X button. When I examine the backtrace in GDB, it starts with QQuickView's destructor, confirming that the sceneGraphInvalidated step seemed to just be skipped. My connection line: connect(window(), &QQuickWindow::sceneGraphInvalidated, m_renderThread, &RenderThread::shutDown, Qt::QueuedConnection); I've checked that window() actually has a value, everything seems good. I receive windowChange() events without any issue. I've made sure my parents are setup properly (and tried without them setup like the example), I seem to be running on the proper threads (i.e. I printed out the threads in the example, and my code jumps around threads in the same manner.) I've modified main.qml in the example to be pretty much the same as I have in my code. Nothing I do makes any difference in this respect. Is there anything I could accidentally be doing that could somehow be blocking my signal? Thanks. From MRussell at neptec.com Sun Sep 3 20:40:07 2017 From: MRussell at neptec.com (Russell, Matthew) Date: Sun, 3 Sep 2017 18:40:07 +0000 Subject: [Interest] QQuickWindow::sceneGraphInvalidated not invoked Message-ID: <8D32ABA6D779C645A4CD11339363E5FF0D3DF0@EMAIL.corp.neptec.com> (not able to reply to my previous message, hopefully subject matching will work.) I re-read the description of the order of the signals in the comments on this bug report : https://bugreports.qt.io/browse/QTBUG-42620 My problem turned out to be that I had my renderers owned by some QObjects in my scene graph. So when my QQuickView was being deleted, those objects were trying to destroy their renderers; all this before QQuickWindow emitted the sceneGraphInvalidated signal. To fix this, I simply made the renderers shared pointers, and my threaded renderer takes shared ownership of them. From thiago.macieira at intel.com Mon Sep 4 02:17:01 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Sun, 03 Sep 2017 21:17:01 -0300 Subject: [Interest] QFileSystemWatcher and kate In-Reply-To: References: Message-ID: <2542893.eSXQE5HDtT@tjmaciei-mobl1> On Sunday, 3 September 2017 05:21:26 -03 Christian Ehrlicher wrote: > Hi, > > I stumbled across a problem with QFileSystemWatcher on Linux. It looks > like the inotify implementation does not honor a file replace action as > expected. > I'm adding a file with addPath() to watch for changes. But after I edit > the file with kate (or any other editor which saves the file in a temp > file and then renaming it to correct name), the QFileSystemWatcher does > no longer watch for modifications... > Looking in the bug tracker I also found a bug which describes the > problem: > https://bugreports.qt.io/browse/QTBUG-19350?jql=text%20~%20%22qfilesystemwat > cher%22 So it looks like this bug exists since inotify support was added but > the bug report was closed because it was not reproducible (maybe not tested > with inotify). > Therefore my question - how to reopen this bug or should I file a new one? Strictly speaking, the behaviour you're seeing is the correct one: the file you were watching was not modified. It was deleted. The new contents do not belong to this file, but to another. If you were using tail -f to follow this file you wouldn't see anything -- you need -F. So to fix this issue (the strictly correct behaviour is not the expected one), we need to add some other mechanism, like tail's flag above. That's because we'll need more resources, as we need to check the directory for change. I have a vague recollection that if you do just that with QFSW, you'll get the notification you want. So simply start watching the directory that contains the file you're watching. I also wonder if inotify can notify you of a change in file metadata. That is, can it notify you of a change in st_nlink? -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From qt-project.pyqt at infinite-drive.com Mon Sep 4 10:22:46 2017 From: qt-project.pyqt at infinite-drive.com (Michael Hartung) Date: Mon, 4 Sep 2017 10:22:46 +0200 Subject: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt-Includes - Final slash switch surprise... Message-ID: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> Hi Folks, using VS-2015 compiling a project in "(Visual Studio 2013) - mode" together with "msvc2013_64" works nicely. As soon as I switch to native "Visual Studio 2015 - mode" together with "msvc2015_64"...: "C:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore/qglobal.h(45): fatal error C1083: Cannot open include file..." Checking projects "External Dependencies" e.g. for "qglobal.h" paths are correct: "c:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore\qglobal.h" Tried a lot and before ending up in a hysterical breakdown :-) I decided to publish. Which is striking is that the "Visual Studio 2013 - mode" is always the default after opening the pro file even from a "virgin" directory created with git. To be honest my visual usage/experience is limited to changing the Qt-Version, opening a pro file and triggering the build... and from poking around in the projects and Tools properies/options... All my sources are solely done with eclipse. Of course any helpful ideas are welcome... Thanks in advance! Michael From thiago.macieira at intel.com Mon Sep 4 14:02:10 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Mon, 04 Sep 2017 09:02:10 -0300 Subject: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt-Includes - Final slash switch surprise... In-Reply-To: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> References: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> Message-ID: <2708603.8ogfWJitF7@tjmaciei-mobl1> On Monday, 4 September 2017 05:22:46 -03 Michael Hartung wrote: > "C:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore/qglobal.h(45): fatal > error C1083: Cannot open include file..." The ... here is the most important information: the file it couldn't open. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From Ch.Ehrlicher at gmx.de Mon Sep 4 19:59:34 2017 From: Ch.Ehrlicher at gmx.de (Christian Ehrlicher) Date: Mon, 4 Sep 2017 19:59:34 +0200 Subject: [Interest] QFileSystemWatcher and kate In-Reply-To: <2542893.eSXQE5HDtT@tjmaciei-mobl1> References: <2542893.eSXQE5HDtT@tjmaciei-mobl1> Message-ID: Am 04.09.2017 um 02:17 schrieb Thiago Macieira: > On Sunday, 3 September 2017 05:21:26 -03 Christian Ehrlicher wrote: >> Hi, >> >> I stumbled across a problem with QFileSystemWatcher on Linux. It looks >> like the inotify implementation does not honor a file replace action as >> expected. >> I'm adding a file with addPath() to watch for changes. But after I edit >> the file with kate (or any other editor which saves the file in a temp >> file and then renaming it to correct name), the QFileSystemWatcher does >> no longer watch for modifications... >> Looking in the bug tracker I also found a bug which describes the >> problem: >> https://bugreports.qt.io/browse/QTBUG-19350?jql=text%20~%20%22qfilesystemwat >> cher%22 So it looks like this bug exists since inotify support was added but >> the bug report was closed because it was not reproducible (maybe not tested >> with inotify). >> Therefore my question - how to reopen this bug or should I file a new one? > Strictly speaking, the behaviour you're seeing is the correct one: the file you > were watching was not modified. It was deleted. The new contents do not belong > to this file, but to another. If you were using tail -f to follow this file you > wouldn't see anything -- you need -F. > > So to fix this issue (the strictly correct behaviour is not the expected one), > we need to add some other mechanism, like tail's flag above. That's because > we'll need more resources, as we need to check the directory for change. I did some tests. The windows implementation works as I (as a user of QFileSystemWatcher) would have expected it. A copy/rename of a file does not trigger an unwatch of the file. Inotify by itself does not provide such a behavior but maybe checking for IN_DELETE_SELF/IN_IGNORED and re-adding the file could be an option? At least the different behavior of the windows and inotify backend should be documented somewhere... Christian From Ch.Ehrlicher at gmx.de Mon Sep 4 20:00:20 2017 From: Ch.Ehrlicher at gmx.de (Christian Ehrlicher) Date: Mon, 4 Sep 2017 20:00:20 +0200 Subject: [Interest] QFileSystemWatcher and kate In-Reply-To: References: <2542893.eSXQE5HDtT@tjmaciei-mobl1> Message-ID: <89f3cea6-66a6-80e6-1131-637338f44410@gmx.de> Am 04.09.2017 um 19:59 schrieb Christian Ehrlicher: > Am 04.09.2017 um 02:17 schrieb Thiago Macieira: >> On Sunday, 3 September 2017 05:21:26 -03 Christian Ehrlicher wrote: >>> Hi, >>> >>> I stumbled across a problem with QFileSystemWatcher on Linux. It looks >>> like the inotify implementation does not honor a file replace action as >>> expected. >>> I'm adding a file with addPath() to watch for changes. But after I edit >>> the file with kate (or any other editor which saves the file in a temp >>> file and then renaming it to correct name), the QFileSystemWatcher does >>> no longer watch for modifications... >>> Looking in the bug tracker I also found a bug which describes the >>> problem: >>> https://bugreports.qt.io/browse/QTBUG-19350?jql=text%20~%20%22qfilesystemwat >>> >>> cher%22 So it looks like this bug exists since inotify support was >>> added but >>> the bug report was closed because it was not reproducible (maybe not >>> tested >>> with inotify). >>> Therefore my question - how to reopen this bug or should I file a >>> new one? >> Strictly speaking, the behaviour you're seeing is the correct one: >> the file you >> were watching was not modified. It was deleted. The new contents do >> not belong >> to this file, but to another. If you were using tail -f to follow >> this file you >> wouldn't see anything -- you need -F. >> >> So to fix this issue (the strictly correct behaviour is not the >> expected one), >> we need to add some other mechanism, like tail's flag above. That's >> because >> we'll need more resources, as we need to check the directory for change. > I did some tests. The windows implementation works as I (as a user of > QFileSystemWatcher) would have expected it. A copy/rename of a file > does not trigger an unwatch of the file. > Inotify by itself does not provide such a behavior but maybe checking > for IN_DELETE_SELF/IN_IGNORED and re-adding the file could be an option? > At least the different behavior of the windows and inotify backend > should be documented somewhere... Ah, and btw: even vi is using the temporary file 'trick' to save a file. Christian From moldovan.catalin at gmail.com Mon Sep 4 22:27:21 2017 From: moldovan.catalin at gmail.com (Catalin Moldovan) Date: Mon, 4 Sep 2017 23:27:21 +0300 Subject: [Interest] [Qt3D] Building Qt3D without Assimp on Rpi natively Message-ID: Hi, I'm having a hard time getting Qt3D built and installed natively with no Assimp on Raspberry Pi 3 with Raspbian Stretch. If I configure Qt5.9.2 with "-no-assimp" option Qt3D is built without Assimp or so it seems but during installation "sudo make install" I get an error when it reaches Qt3D "Project ERROR: Module does not define version". Alternatively if I try to build Qt3D separately after installing Qt 5.9.2, using qmake, it always sets Assimp build to yes, configured like this: qmake ../qt5/qt3d/qt3d.pro Configure summary: Qt 3D: Assimp ................................. yes System Assimp .......................... no Output Qt3D Job traces ................. no Output Qt3D GL traces .................. no Qt 3D GeometryLoaders: Autodesk FBX ........................... no It doesn't automatically detect already installed Assimp which is version 3.3.1 accorgind to apt-cache policy libassimp-dev libassimp-dev: Installed: 3.3.1~dfsg-4 Candidate: 3.3.1~dfsg-4 Is there a way of disabling Assimp from being built when building only Qt3D with qmake? Thanks, Catalin -------------- next part -------------- An HTML attachment was scrubbed... URL: From googleersatz at oliverniebuhr.de Tue Sep 5 08:34:15 2017 From: googleersatz at oliverniebuhr.de (Oliver Niebuhr) Date: Tue, 5 Sep 2017 08:34:15 +0200 Subject: [Interest] Linux Deployment Question Message-ID: <909cd57c-fc34-989e-3edc-d9c9d2d40573@oliverniebuhr.de> Good morning. I think I have found now all necessary Files to deploy my App from Antergos to other Distros. But in a clean Debian VM (GNU/Linux 9 (stretch) 64 bit, GNOME 3.22.2, Gallium 0.4 on llvmpipe (llvm 3.9, 128 bits)) i still get an Error when I try to start the Application. In my Pro File I added this Line (Tip from the tripleboot.org Blog): linux:QMAKE_LFLAGS += -Wl,-rpath,"'\$$ORIGIN'" So the App looks for the Qt5 Libs in the Directory where it is called from. Like expected. But no matter if I run the Binary directly or if I use this Script from the Qt Deployment Wiki: #!/bin/sh appname=`basename $0 | sed s,\.sh$,,` dirname=`dirname $0` tmp="${dirname#?}" if [ "${dirname%$tmp}" != "/" ]; then dirname=$PWD/$dirname fi LD_LIBRARY_PATH=$dirname export LD_LIBRARY_PATH $dirname/$appname "$@" I get this Error: /home/test/AKKomPolTool/./AKKomPolTool: Error: The Qt5 Main Libs are in the same Directory where the Binary is - so far so good. And all the Plugins are under /directory with binary/plugins/ I am not sure what I am still missing. Does a Debian User need to install some additional Qt5 Stuff first to make the App work? Any hint is much appreciated! This is the List of Files I am currently deploying (not sure if they are all necessary [well of course i need the GCC Files]): QT5 LIBs... /qt-buildx64/lib/libQt5Core.so.5 /qt-buildx64/lib/libQt5DBus.so.5 /qt-buildx64/lib/libQt5Gui.so.5 /qt-buildx64/lib/libQt5Network.so.5 /qt-buildx64/lib/libQt5PrintSupport.so.5 /qt-buildx64/lib/libQt5Sql.so.5 /qt-buildx64/lib/libQt5Widgets.so.5 /qt-buildx64/lib/libQt5XcbQpa.so.5 /qt-buildx64/lib/libQt5Xml.so.5 Bearer Plugins... /qt-buildx64/plugins/bearer/libqconnmanbearer.soplugins/bearer/ /qt-buildx64/plugins/bearer/libqgenericbearer.soplugins/bearer/ /qt-buildx64/plugins/bearer/libqnmbearer.soplugins/bearer/ Imageformats Plugins... /qt-buildx64/plugins/imageformats/libqgif.so /qt-buildx64/plugins/imageformats/libqicns.so /qt-buildx64/plugins/imageformats/libqico.so /qt-buildx64/plugins/imageformats/libqjp2.so /qt-buildx64/plugins/imageformats/libqjpeg.so /qt-buildx64/plugins/imageformats/libqmng.so /qt-buildx64/plugins/imageformats/libqtga.so /qt-buildx64/plugins/imageformats/libqtiff.so /qt-buildx64/plugins/imageformats/libqwbmp.so /qt-buildx64/plugins/imageformats/libqwebp.so Platforminputcontexts Plugins... /qt-buildx64/plugins/platforminputcontexts/libcomposeplatforminputcontextplugin.so /qt-buildx64/plugins/platforminputcontexts/libibusplatforminputcontextplugin.so Platforms Plugins... /binaries/qt-buildx64/plugins/platforms/libqxcb.so Platformthemes Plugins... /qt-buildx64/plugins/platformthemes/libqgtk3.so Printsupport Plugins... /qt-buildx64/plugins/printsupport/libcupsprintersupport.so Sqldrivers Plugins... /qt-buildx64/plugins/sqldrivers/libqsqlite.so Xcbglintegrations Plugins... /qt-buildx64/plugins/xcbglintegrations/libqxcb-glx-integration.so #FIX Check if we must really deploy all that nonsense!! 3RDPARTY AND SYSTEM LIBs... /usr/lib64/libbz2.so.1.0 /usr/lib64/libc.so.6 /usr/lib64/libdbus-1.so.3 /usr/lib64/libdl.so.2 /usr/lib64/libdouble-conversion.so.1 /usr/lib64/libexpat.so.1 /usr/lib64/libfontconfig.so.1 /usr/lib64/libfreetype.so.6 /usr/lib64/libgcc_s.so.1 /usr/lib64/libgcrypt.so.20 /usr/lib64/libGL.so.1 /usr/lib64/libGLdispatch.so.0 /usr/lib64/libglib-2.0.so.0 /usr/lib64/libGLX.so.0 /usr/lib64/libgpg-error.so.0 /usr/lib64/libgraphite2.so.3 /usr/lib64/libgthread-2.0.so.0 /usr/lib64/libharfbuzz.so.0 /usr/lib64/libICE.so.6 /usr/lib64/liblz4.so.1 /usr/lib64/liblzma.so.5 /usr/lib64/libm.so.6 /usr/lib64/libpcre.so.1 /usr/lib64/libpcre2-16.so.0 /usr/lib64/libpng16.so.16 /usr/lib64/libpthread.so.0 /usr/lib64/librt.so.1 /usr/lib64/libSM.so.6 /usr/lib64/libstdc++.so.6 /usr/lib64/libsystemd.so.0 /usr/lib64/libuuid.so.1 /usr/lib64/libX11.so.6 /usr/lib64/libX11-xcb.so.1 /usr/lib64/libXau.so.6 /usr/lib64/libxcb.so.1 /usr/lib64/libxcb-icccm.so.4 /usr/lib64/libxcb-image.so.0 /usr/lib64/libxcb-keysyms.so.1 /usr/lib64/libxcb-randr.so.0 /usr/lib64/libxcb-render.so.0 /usr/lib64/libxcb-render-util.so.0 /usr/lib64/libxcb-shape.so.0 /usr/lib64/libxcb-shm.so.0 /usr/lib64/libxcb-sync.so.1 /usr/lib64/libxcb-util.so.1 /usr/lib64/libxcb-xfixes.so.0 /usr/lib64/libxcb-xinerama.so.0 /usr/lib64/libxcb-xkb.so.1 /usr/lib64/libXdmcp.so.6 /usr/lib64/libXext.so.6 /usr/lib64/libXi.so.6 /usr/lib64/libxkbcommon.so.0 /usr/lib64/libxkbcommon-x11.so.0 /usr/lib64/libXrender.so.1 /usr/lib64/libz.so.1 OpenSSL 1.1.0 LIBs... /home/oliver/developing/binaries/opensslx64release/lib/libssl.so.1.1 /home/oliver/developing/binaries/opensslx64release/lib/libcrypto.so.1.1 ICU 59.1 LIBs... /home/oliver/developing/binaries/icux64release/lib/libicui18n.so.59 /home/oliver/developing/binaries/icux64release/lib/libicuuc.so.59 /home/oliver/developing/binaries/icux64release/lib/libicudata.so.59 Thank You very much! Oliver From julienbordes at hotmail.fr Tue Sep 5 10:48:57 2017 From: julienbordes at hotmail.fr (Julien Bordes) Date: Tue, 5 Sep 2017 08:48:57 +0000 Subject: [Interest] News about Qt on QNX ? In-Reply-To: <313011503938895@web12j.yandex.ru> References: , <313011503938895@web12j.yandex.ru> Message-ID: > > Hello All, > > > > I am currently developing a QtQuick/QML based application with Qt 5.9.1 and QNX 6.6 on a iMX6 platform with a Linux workstation and I have some difficulties to get fresh informations about Qt on QNX and I don't know where to look anymore. > > (Thiago Macieira) > > Qt on QNX is still very well supported, by Blackberry themselves. It's one of > the platforms tested in the Qt continuous integration system. I have no doubt about the work of Blackberry but nevertheless there is not much fresh news availables about modules compatibility neither on the Qt website and wiki nor on QNX website, even our local license sellers weren't able to provide us thoses informations. > I have no clue about iMX6 though. According ot the (wiki https://wiki.qt.io/QNX_Platform#Reference_boards) a Freescale i.MX6Q Sabre Lite, board is use for development and testing. > > I first searched in the wiki but it was updated for the last time in April 2015 (http://wiki.qt.io/Qt_Status_on_QNX) and there is no news on the Qt documentation (http://doc.qt.io/qt-5/platform-notes-qnx.html) and in Supported platforms (http://doc.qt.io/qt-5/supported-platforms.html). > > > > The most recent informations I found are in this presentation from September 2016 at QtCon by James McDonnel https://conf.qtcon.org/system/attachments/154/original/Qt_on_QNX_-_2016_-_003.pdf?1474298472) and in the Qt mailing list archives (http://lists.qt-project.org/pipermail/development/2017-March/029347.html). > > > > > > I also found out in april 2017 in the Development mailing list (http://lists.qt-project.org/pipermail/development/2017-April/029597.html) and in the Maintainers list (https://wiki.qt.io/Maintainers) than James McDonnel has been nominated as the official maintainer for Qt on QNX. > > > > Based on those previous statements I have some questions, regarding Qt on QNX. > > > > - Roadmap and Modules support > > > > Does a roadmap about Qt on QNX exist ? If yes where can we find it ? > > (Thiago Macieira) > > In the same place as the roadmap for all the other platforms. > > We don't make roadmaps. I thought you did :) http://blog.qt.io/blog/2016/02/22/qt-roadmap-for-2016/ http://blog.qt.io/blog/2017/02/22/qt-roadmap-for-2017/ There is also informations in the wiki in "Release" and "Known issues" about current and future versions of Qt, but there is not often mention of QNX. > > It seems Qt 5.9 is supported by both QNX 6.6 and QNX 7.0, what will it be for later version of QT (5.10, 5.1x, and later 6.x) ? > > (Thiago Macieira) > > Most likely we'll need to support 6.6 for some time. The fact that it uses GCC > 4.7 was a problem, as I wanted to drop it from our supported list for 5.10. > > > > > It there a way to know the list of modules supported for recent versions of Qt ? I also found there is a Qt Bug about this planned to be fixed for Qt 5.10 (https://bugreports.qt.io/browse/QTBUG-59976). > > (Thiago Macieira) > > All of the ones we develop. That excludes qtwebkit and qtwebengine, since they > are upstream projects. The question was more QNX oriented as today not all modules are compatible with QNX 6.6 (e.g. Virtual keyboard). > > - Webengine > > > > Since Qt5.6 QtWebkit is no more officially supported by Qt. However you can compile it anyway but it works only with QWidgets and not with QtQuick/QML because QtDeclarative is not supported, from what I understand. > > (Thiago Macieira) > > And there's QtWebkit 5.212 if you need an update. When I checkout QtWebkit version to 5.212 I cannot compile Qt 5.9.1 for QNX 6.6 with gcc 4.7.3. The WebKit build was disabled for the following reasons: * cross-compilation of QtWebKit with qmake is not supported yet * QNX is not supported. * Using gcc version 4.7, but at least gcc version 4.9 is required to build QtWebKit. But as far as I understand it is expected as QNX is not supported anymore by the new QtWebkit (http://lists.qt-project.org/pipermail/development/2016-June/026156.html) Beside this I can compile QWebkit "bf172ae - Fix compilation with ICU 59 (HEAD, origin/5.9.1)", even if there is a small adjustment to be able to do so. I filled a bug report about it one year ago and I updated it this week (https://bugreports.qt.io/browse/QTBUG-54307). > (Konstantin Tokarev) > > It's probably not hard to provide QML API of QtWebKit on QNX. > > The only requirement is one sufficiently motivated developer with access to compiler toolchain and QNX device, and knowledge of QNX native IPC. Feel free to ask me any questions on implementation on IRC and mailing lists I would be happy to help but my QNX's IPC skills right now are very limited. Beside that I would like to know what is the position of Blackberry/QNX about web page display in the future with Qt and especially QtQuick/QML ? > > Nevetheless there is no solution to display HTML page with QtQuick/QML on QNX and QtWebengine (QtWebkit replacement) doesn't work on QNX because it is tied to the rendering engine Blink from chromium project. > > > > In the presentation from September 2016 there is a mention of it and a question "QtWebEngine from BLINK for QNX?". > > > > I would like to know if there is a port in progress ? If yes what is the current status of this port and will it be supported on both QNX 6.6 and QNX 7.0 ? > > > > > > - Virtual Keyboard > > > > Is there a plan to make the QtVirtualKeyboard module compatible with QNX ? > > > > There is already a bug (https://bugreports.qt.io/browse/QTBUG-54576) which has been reported by a silver user, but there is no gerrit open. > > > > I personally tried the patch for virtual keyboard and it worked at least up to Qt 5.8 but it seems to not work anymore with Qt 5.9.1. > > > > - Qt Creator and remote debugging > > > > It is currently impossible to remote debug custom QML plugin used through QML loader because even if you add "set sysroot" or "set solib-search-path" in "Options > Debug > Gdb > Additionnal startup commands" the symbols are not loaded and the breakpoints do not work, at least with Qt Creator 4.2. > > > > It seem already has been described in a old bug which has never been opened (https://bugreports.qt.io/browse/QTCREATORBUG-12980) > > > > Is there another method use by Qt developers to debug remote custom QML application with plugins on QNX ? > > > > I tried with the latest Qt Creator version v4.4-rc1, but for now I cannot start the debugging on QNX (https://bugreports.qt.io/browse/QTCREATORBUG-18804). > > > > > > - Development > > > > Those last ones are organizational questions about how Qt and Qt Creator are developed for QNX, because it is not really clear for us. > > > > We understand there are QNX employees working on Qt, but are they also employees of Qt working on QNX port? Maybe it depends on which module/part ? > > (Thiago Macieira) > > They work for Blackberry. > > > From what we understood QtCreator is only maintained by Qt even for the QNX part, is that right ? > > (Thiago Macieira) > > Why does it matter? It is to have a better understanding of the of Qt and QtCreator developpement organisation. Actually we also ask our local QNX and Qt sellers about this question and they didn't provide us clear answer. > > It would be very helpful if James McDonnel or someone else working on Qt for QNX would be able to answer some of those questions. > > > > Thank you in advance. Regards. Julien Bordes. -------------- next part -------------- An HTML attachment was scrubbed... URL: From tuukka.turunen at qt.io Tue Sep 5 11:35:39 2017 From: tuukka.turunen at qt.io (Tuukka Turunen) Date: Tue, 5 Sep 2017 09:35:39 +0000 Subject: [Interest] News about Qt on QNX ? In-Reply-To: References: <313011503938895@web12j.yandex.ru> Message-ID: <99968A52-4EF1-46A9-8EBA-51C698E3A897@qt.io> Hi Julien, Qt 5.9 LTS (and later versions) support QNX 7.0 (http://doc.qt.io/qt-5/supported-platforms.html). You are right that the HTML5 engine offering on QNX is somewhat complicated as Qt WebEngine does not work on it due to lacking Chromium port for QNX. According to QNX this is something they are working on, so perhaps this something to be addressed later (http://blogs.blackberry.com/2016/12/blackberry-qnx-partners-with-obigo-to-build-a-better-browser-for-your-car/). It is possible to run Qt WebKit, but this needs to be used with care as it is also not out of the box supported. If the needs are related to a commercial project, there are multiple service partners as well as consulting teams of The Qt Company to help reach a solution for the HTML5 needs also on QNX. Yours, Tuukka From: Interest on behalf of Julien Bordes Date: Tuesday, 5 September 2017 at 11.48 To: "interest at qt-project.org" Subject: Re: [Interest] News about Qt on QNX ? > > Hello All, > > > > I am currently developing a QtQuick/QML based application with Qt 5.9.1 and QNX 6.6 on a iMX6 platform with a Linux workstation and I have some difficulties to get fresh informations about Qt on QNX and I don't know where to look anymore. > > (Thiago Macieira) > > Qt on QNX is still very well supported, by Blackberry themselves. It's one of > the platforms tested in the Qt continuous integration system. I have no doubt about the work of Blackberry but nevertheless there is not much fresh news availables about modules compatibility neither on the Qt website and wiki nor on QNX website, even our local license sellers weren't able to provide us thoses informations. > I have no clue about iMX6 though. According ot the (wiki https://wiki.qt.io/QNX_Platform#Reference_boards) a Freescale i.MX6Q Sabre Lite, board is use for development and testing. > > I first searched in the wiki but it was updated for the last time in April 2015 (http://wiki.qt.io/Qt_Status_on_QNX) and there is no news on the Qt documentation (http://doc.qt.io/qt-5/platform-notes-qnx.html) and in Supported platforms (http://doc.qt.io/qt-5/supported-platforms.html). > > > > The most recent informations I found are in this presentation from September 2016 at QtCon by James McDonnel https://conf.qtcon.org/system/attachments/154/original/Qt_on_QNX_-_2016_-_003.pdf?1474298472) and in the Qt mailing list archives (http://lists.qt-project.org/pipermail/development/2017-March/029347.html). > > > > > > I also found out in april 2017 in the Development mailing list (http://lists.qt-project.org/pipermail/development/2017-April/029597.html) and in the Maintainers list (https://wiki.qt.io/Maintainers) than James McDonnel has been nominated as the official maintainer for Qt on QNX. > > > > Based on those previous statements I have some questions, regarding Qt on QNX. > > > > - Roadmap and Modules support > > > > Does a roadmap about Qt on QNX exist ? If yes where can we find it ? > > (Thiago Macieira) > > In the same place as the roadmap for all the other platforms. > > We don't make roadmaps. I thought you did :) http://blog.qt.io/blog/2016/02/22/qt-roadmap-for-2016/ http://blog.qt.io/blog/2017/02/22/qt-roadmap-for-2017/ There is also informations in the wiki in "Release" and "Known issues" about current and future versions of Qt, but there is not often mention of QNX. > > It seems Qt 5.9 is supported by both QNX 6.6 and QNX 7.0, what will it be for later version of QT (5.10, 5.1x, and later 6.x) ? > > (Thiago Macieira) > > Most likely we'll need to support 6.6 for some time. The fact that it uses GCC > 4.7 was a problem, as I wanted to drop it from our supported list for 5.10. > > > > > It there a way to know the list of modules supported for recent versions of Qt ? I also found there is a Qt Bug about this planned to be fixed for Qt 5.10 (https://bugreports.qt.io/browse/QTBUG-59976). > > (Thiago Macieira) > > All of the ones we develop. That excludes qtwebkit and qtwebengine, since they > are upstream projects. The question was more QNX oriented as today not all modules are compatible with QNX 6.6 (e.g. Virtual keyboard). > > - Webengine > > > > Since Qt5.6 QtWebkit is no more officially supported by Qt. However you can compile it anyway but it works only with QWidgets and not with QtQuick/QML because QtDeclarative is not supported, from what I understand. > > (Thiago Macieira) > > And there's QtWebkit 5.212 if you need an update. When I checkout QtWebkit version to 5.212 I cannot compile Qt 5.9.1 for QNX 6.6 with gcc 4.7.3. The WebKit build was disabled for the following reasons: * cross-compilation of QtWebKit with qmake is not supported yet * QNX is not supported. * Using gcc version 4.7, but at least gcc version 4.9 is required to build QtWebKit. But as far as I understand it is expected as QNX is not supported anymore by the new QtWebkit (http://lists.qt-project.org/pipermail/development/2016-June/026156.html) Beside this I can compile QWebkit "bf172ae - Fix compilation with ICU 59 (HEAD, origin/5.9.1)", even if there is a small adjustment to be able to do so. I filled a bug report about it one year ago and I updated it this week (https://bugreports.qt.io/browse/QTBUG-54307). > (Konstantin Tokarev) > > It's probably not hard to provide QML API of QtWebKit on QNX. > > The only requirement is one sufficiently motivated developer with access to compiler toolchain and QNX device, and knowledge of QNX native IPC. Feel free to ask me any questions on implementation on IRC and mailing lists I would be happy to help but my QNX's IPC skills right now are very limited. Beside that I would like to know what is the position of Blackberry/QNX about web page display in the future with Qt and especially QtQuick/QML ? > > Nevetheless there is no solution to display HTML page with QtQuick/QML on QNX and QtWebengine (QtWebkit replacement) doesn't work on QNX because it is tied to the rendering engine Blink from chromium project. > > > > In the presentation from September 2016 there is a mention of it and a question "QtWebEngine from BLINK for QNX?". > > > > I would like to know if there is a port in progress ? If yes what is the current status of this port and will it be supported on both QNX 6.6 and QNX 7.0 ? > > > > > > - Virtual Keyboard > > > > Is there a plan to make the QtVirtualKeyboard module compatible with QNX ? > > > > There is already a bug (https://bugreports.qt.io/browse/QTBUG-54576) which has been reported by a silver user, but there is no gerrit open. > > > > I personally tried the patch for virtual keyboard and it worked at least up to Qt 5.8 but it seems to not work anymore with Qt 5.9.1. > > > > - Qt Creator and remote debugging > > > > It is currently impossible to remote debug custom QML plugin used through QML loader because even if you add "set sysroot" or "set solib-search-path" in "Options > Debug > Gdb > Additionnal startup commands" the symbols are not loaded and the breakpoints do not work, at least with Qt Creator 4.2. > > > > It seem already has been described in a old bug which has never been opened (https://bugreports.qt.io/browse/QTCREATORBUG-12980) > > > > Is there another method use by Qt developers to debug remote custom QML application with plugins on QNX ? > > > > I tried with the latest Qt Creator version v4.4-rc1, but for now I cannot start the debugging on QNX (https://bugreports.qt.io/browse/QTCREATORBUG-18804). > > > > > > - Development > > > > Those last ones are organizational questions about how Qt and Qt Creator are developed for QNX, because it is not really clear for us. > > > > We understand there are QNX employees working on Qt, but are they also employees of Qt working on QNX port? Maybe it depends on which module/part ? > > (Thiago Macieira) > > They work for Blackberry. > > > From what we understood QtCreator is only maintained by Qt even for the QNX part, is that right ? > > (Thiago Macieira) > > Why does it matter? It is to have a better understanding of the of Qt and QtCreator developpement organisation. Actually we also ask our local QNX and Qt sellers about this question and they didn't provide us clear answer. > > It would be very helpful if James McDonnel or someone else working on Qt for QNX would be able to answer some of those questions. > > > > Thank you in advance. Regards. Julien Bordes. -------------- next part -------------- An HTML attachment was scrubbed... URL: From lorn.potter at gmail.com Tue Sep 5 11:56:40 2017 From: lorn.potter at gmail.com (Lorn Potter) Date: Tue, 5 Sep 2017 19:56:40 +1000 Subject: [Interest] News about Qt on QNX ? In-Reply-To: <3601395.1kyBBvWXnZ@tjmaciei-mobl1> References: <3601395.1kyBBvWXnZ@tjmaciei-mobl1> Message-ID: <1d9f12c5-79fd-b264-d726-9dfe0617dcff@gmail.com> On 29/8/17 2:38 am, Thiago Macieira wrote: > Qt on QNX is still very well supported, by Blackberry themselves. It's one of > the platforms tested in the Qt continuous integration system. Do they still? The BB backend for QtSensors was completely removed in 5.6. -- Lorn 'ljp' Potter Software Engineer, llornkcor technologies Qt Sensors / connectivity From Jake.Petroules at qt.io Tue Sep 5 11:58:31 2017 From: Jake.Petroules at qt.io (Jake Petroules) Date: Tue, 5 Sep 2017 09:58:31 +0000 Subject: [Interest] News about Qt on QNX ? In-Reply-To: <1d9f12c5-79fd-b264-d726-9dfe0617dcff@gmail.com> References: <3601395.1kyBBvWXnZ@tjmaciei-mobl1> <1d9f12c5-79fd-b264-d726-9dfe0617dcff@gmail.com> Message-ID: <49F2F055-67F5-41C2-88A6-3688A0E0754A@qt.io> > On Sep 5, 2017, at 2:56 AM, Lorn Potter wrote: > > On 29/8/17 2:38 am, Thiago Macieira wrote: >> Qt on QNX is still very well supported, by Blackberry themselves. It's one of >> the platforms tested in the Qt continuous integration system. > > Do they still? The BB backend for QtSensors was completely removed in 5.6. BB10 is a superset of QNX. Presumably the BB backend used APIs specific to BB10 that are not available in base QNX? > > -- > Lorn 'ljp' Potter > Software Engineer, llornkcor technologies > Qt Sensors / connectivity > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Jake Petroules - jake.petroules at qt.io The Qt Company - Silicon Valley Qbs build tool evangelist - qbs.io From annulen at yandex.ru Tue Sep 5 12:31:10 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Tue, 05 Sep 2017 13:31:10 +0300 Subject: [Interest] News about Qt on QNX ? In-Reply-To: References: , <313011503938895@web12j.yandex.ru> Message-ID: <82871504607470@web7g.yandex.ru> 05.09.2017, 11:49, "Julien Bordes" : >> > - Webengine >> > >> > Since Qt5.6 QtWebkit is no more officially supported by Qt. However you can compile it anyway but it works only with QWidgets and not with QtQuick/QML because QtDeclarative is not supported, from what I understand. >> >> (Thiago Macieira) >> >> And there's QtWebkit 5.212 if you need an update. > > When I checkout QtWebkit version to 5.212 I cannot compile Qt 5.9.1 for QNX 6.6 with gcc 4.7.3. > > The WebKit build was disabled for the following reasons: >     * cross-compilation of QtWebKit with qmake is not supported yet >     * QNX is not supported. >     * Using gcc version 4.7, but at least gcc version 4.9 is required to build QtWebKit. > > But as far as I understand it is expected as QNX is not supported anymore by the new QtWebkit (http://lists.qt-project.org/pipermail/development/2016-June/026156.html) Building for QNX is disabled by default because nobody was interested in trying this so far. IIRC it's not trivial to obtain QNX toolchain so I haven't tried myself either. However, I can guide you if you are willing to try. > > Beside this I can compile QWebkit "bf172ae - Fix compilation with ICU 59 (HEAD, origin/5.9.1)", even if  there is a small adjustment to be able to do so. > I filled a bug report about it one year ago and I updated it this week (https://bugreports.qt.io/browse/QTBUG-54307). > >> (Konstantin Tokarev) >> >> It's probably not hard to provide QML API of QtWebKit on QNX. >> >> The only requirement is one sufficiently motivated developer with access to compiler toolchain and QNX device, and knowledge of QNX native IPC. Feel free to ask me any questions on implementation on IRC and mailing lists > > I would be happy to help but my QNX's IPC skills right now are very limited. It might be possible to do without QNX IPC if there is support for unix domain sockets in QNX. This may work slower but won't require new code to be written. Also, Brett Stottlemyer have implemented QNX IPC backend for QtRemoteObjects so this code can be used as reference. http://code.qt.io/cgit/qt/qtremoteobjects.git/tree/src/remoteobjects Note that widgets API should work just fine without any IPC hassles. > > Beside that I would like to know what is the position of Blackberry/QNX about web page display in the future with Qt and especially QtQuick/QML ? > >> > Nevetheless there is no solution to display HTML page with QtQuick/QML on QNX and QtWebengine (QtWebkit replacement) doesn't work on QNX because it is tied to the rendering engine Blink from chromium project. >> > >> > In the presentation from September 2016 there is a mention of it and a question "QtWebEngine from BLINK for QNX?". >> > >> > I would like to know if there is a port in progress ? If yes what is the current status of this port and will it be supported on both QNX 6.6 and QNX 7.0 ? >> > >> > >> > - Virtual Keyboard >> > >> > Is there a plan to make the QtVirtualKeyboard module compatible with QNX ? >> > >> > There is already a bug (https://bugreports.qt.io/browse/QTBUG-54576) which has been reported by a silver user, but there is no gerrit open. >> > >> > I personally tried the patch for virtual keyboard and it worked at least up to Qt 5.8 but it seems to not work anymore with Qt 5.9.1. >> > >> > - Qt Creator and remote debugging >> > >> > It is currently impossible to remote debug custom QML plugin used through QML loader because even if you add "set sysroot" or "set solib-search-path" in "Options > Debug > Gdb > Additionnal startup commands" the symbols are not loaded and the breakpoints do not work, at least with Qt Creator 4.2. >> > >> > It seem already has been described in a old bug which has never been opened (https://bugreports.qt.io/browse/QTCREATORBUG-12980) >> > >> > Is there another method use by Qt developers to debug remote custom QML application with plugins on QNX ? >> > >> > I tried with the latest Qt Creator version v4.4-rc1, but for now I cannot start the debugging on QNX (https://bugreports.qt.io/browse/QTCREATORBUG-18804). >> > >> > >> > - Development >> > >> > Those last ones are organizational questions about how Qt and Qt Creator are developed for QNX, because it is not really clear for us. >> > >> > We understand there are QNX employees working on Qt, but are they also employees of Qt working on QNX port? Maybe it depends on which module/part ? >> >> (Thiago Macieira) >> >> They work for Blackberry. >> >> > From what we understood QtCreator is only maintained by Qt even for the QNX part, is that right ? >> >> (Thiago Macieira) >> >> Why does it matter? > > It is to have a better understanding of the of Qt and QtCreator developpement organisation. > Actually we also ask our local QNX and Qt sellers about this question and they didn't provide us clear answer. > >> > It would be very helpful if James McDonnel or someone else working on Qt for QNX would be able to answer some of those questions. >> > >> > Thank you in advance. > > Regards. > > Julien Bordes. > , > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest --  Regards, Konstantin From annulen at yandex.ru Tue Sep 5 12:38:52 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Tue, 05 Sep 2017 13:38:52 +0300 Subject: [Interest] News about Qt on QNX ? In-Reply-To: <82871504607470@web7g.yandex.ru> References: , <313011503938895@web12j.yandex.ru> <82871504607470@web7g.yandex.ru> Message-ID: <48091504607932@web58g.yandex.ru> 05.09.2017, 13:31, "Konstantin Tokarev" : > 05.09.2017, 11:49, "Julien Bordes" : >>>  > - Webengine >>>  > >>>  > Since Qt5.6 QtWebkit is no more officially supported by Qt. However you can compile it anyway but it works only with QWidgets and not with QtQuick/QML because QtDeclarative is not supported, from what I understand. >>> >>>  (Thiago Macieira) >>> >>>  And there's QtWebkit 5.212 if you need an update. >> >>  When I checkout QtWebkit version to 5.212 I cannot compile Qt 5.9.1 for QNX 6.6 with gcc 4.7.3. >> >>  The WebKit build was disabled for the following reasons: >>      * cross-compilation of QtWebKit with qmake is not supported yet >>      * QNX is not supported. >>      * Using gcc version 4.7, but at least gcc version 4.9 is required to build QtWebKit. >> >>  But as far as I understand it is expected as QNX is not supported anymore by the new QtWebkit (http://lists.qt-project.org/pipermail/development/2016-June/026156.html) > > Building for QNX is disabled by default because nobody was interested in trying this so far. IIRC it's not trivial to obtain QNX toolchain so I haven't tried myself either. > > However, I can guide you if you are willing to try. But note that gcc 4.7 is not supported, you need to obtain at least 4.9 (or 4.8, if you disable some features) > >>  Beside this I can compile QWebkit "bf172ae - Fix compilation with ICU 59 (HEAD, origin/5.9.1)", even if  there is a small adjustment to be able to do so. >>  I filled a bug report about it one year ago and I updated it this week (https://bugreports.qt.io/browse/QTBUG-54307). >> >>>  (Konstantin Tokarev) >>> >>>  It's probably not hard to provide QML API of QtWebKit on QNX. >>> >>>  The only requirement is one sufficiently motivated developer with access to compiler toolchain and QNX device, and knowledge of QNX native IPC. Feel free to ask me any questions on implementation on IRC and mailing lists >> >>  I would be happy to help but my QNX's IPC skills right now are very limited. > > It might be possible to do without QNX IPC if there is support for unix domain sockets in QNX. This may work slower but won't require new code to be written. > > Also, Brett Stottlemyer have implemented QNX IPC backend for QtRemoteObjects so this code can be used as reference. > > http://code.qt.io/cgit/qt/qtremoteobjects.git/tree/src/remoteobjects > > Note that widgets API should work just fine without any IPC hassles. > >>  Beside that I would like to know what is the position of Blackberry/QNX about web page display in the future with Qt and especially QtQuick/QML ? >> >>>  > Nevetheless there is no solution to display HTML page with QtQuick/QML on QNX and QtWebengine (QtWebkit replacement) doesn't work on QNX because it is tied to the rendering engine Blink from chromium project. >>>  > >>>  > In the presentation from September 2016 there is a mention of it and a question "QtWebEngine from BLINK for QNX?". >>>  > >>>  > I would like to know if there is a port in progress ? If yes what is the current status of this port and will it be supported on both QNX 6.6 and QNX 7.0 ? >>>  > >>>  > >>>  > - Virtual Keyboard >>>  > >>>  > Is there a plan to make the QtVirtualKeyboard module compatible with QNX ? >>>  > >>>  > There is already a bug (https://bugreports.qt.io/browse/QTBUG-54576) which has been reported by a silver user, but there is no gerrit open. >>>  > >>>  > I personally tried the patch for virtual keyboard and it worked at least up to Qt 5.8 but it seems to not work anymore with Qt 5.9.1. >>>  > >>>  > - Qt Creator and remote debugging >>>  > >>>  > It is currently impossible to remote debug custom QML plugin used through QML loader because even if you add "set sysroot" or "set solib-search-path" in "Options > Debug > Gdb > Additionnal startup commands" the symbols are not loaded and the breakpoints do not work, at least with Qt Creator 4.2. >>>  > >>>  > It seem already has been described in a old bug which has never been opened (https://bugreports.qt.io/browse/QTCREATORBUG-12980) >>>  > >>>  > Is there another method use by Qt developers to debug remote custom QML application with plugins on QNX ? >>>  > >>>  > I tried with the latest Qt Creator version v4.4-rc1, but for now I cannot start the debugging on QNX (https://bugreports.qt.io/browse/QTCREATORBUG-18804). >>>  > >>>  > >>>  > - Development >>>  > >>>  > Those last ones are organizational questions about how Qt and Qt Creator are developed for QNX, because it is not really clear for us. >>>  > >>>  > We understand there are QNX employees working on Qt, but are they also employees of Qt working on QNX port? Maybe it depends on which module/part ? >>> >>>  (Thiago Macieira) >>> >>>  They work for Blackberry. >>> >>>  > From what we understood QtCreator is only maintained by Qt even for the QNX part, is that right ? >>> >>>  (Thiago Macieira) >>> >>>  Why does it matter? >> >>  It is to have a better understanding of the of Qt and QtCreator developpement organisation. >>  Actually we also ask our local QNX and Qt sellers about this question and they didn't provide us clear answer. >> >>>  > It would be very helpful if James McDonnel or someone else working on Qt for QNX would be able to answer some of those questions. >>>  > >>>  > Thank you in advance. >> >>  Regards. >> >>  Julien Bordes. >>  , >> >>  _______________________________________________ >>  Interest mailing list >>  Interest at qt-project.org >>  http://lists.qt-project.org/mailman/listinfo/interest > > -- > Regards, > Konstantin > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Regards, Konstantin From thiago.macieira at intel.com Tue Sep 5 15:45:02 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Tue, 05 Sep 2017 10:45:02 -0300 Subject: [Interest] News about Qt on QNX ? In-Reply-To: <1d9f12c5-79fd-b264-d726-9dfe0617dcff@gmail.com> References: <3601395.1kyBBvWXnZ@tjmaciei-mobl1> <1d9f12c5-79fd-b264-d726-9dfe0617dcff@gmail.com> Message-ID: <3934892.C8yQ7lk9Af@tjmaciei-mobl1> On Tuesday, 5 September 2017 06:56:40 -03 Lorn Potter wrote: > On 29/8/17 2:38 am, Thiago Macieira wrote: > > Qt on QNX is still very well supported, by Blackberry themselves. It's one > > of the platforms tested in the Qt continuous integration system. > > Do they still? The BB backend for QtSensors was completely removed in 5.6. BB10 support was removed. QNX support is still there. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From rmanni at gmail.com Tue Sep 5 16:39:48 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Tue, 5 Sep 2017 16:39:48 +0200 Subject: [Interest] state of qttranslations Message-ID: Hi, I am trying to use qttranslations in my application to make common widgets (like open dialog) have their translations. I am compiling it by myself, standard way: qmake && make I am trying to load the full translations qt_.qm Some of them works nearly fine like "hu", but most dont eg "en", "fr", "de" etc. Those which do not work are very small qm files. Checking their ts files, the difference seems to be that they are "catalogs" of smaller ones. Running lrelease on them results in a message that 0 was translated: This is a catalog and does not work: $ lrelease qt_de.ts Updating 'qt_de.qm'... Generated 0 translation(s) (0 finished and 0 unfinished) This is not a catalog and does not work: $ lrelease qt_hu.ts Updating 'qt_hu.qm'... Generated 2131 translation(s) (2113 finished and 18 unfinished) Ignored 100 untranslated source text(s) I observe the same with qm files shipped on archlinux. Does anyone know what do I misunderstand? How do one handle qt's translations nowadays? Thanks for any help, Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From allen.winter at kdab.com Wed Sep 6 00:03:09 2017 From: allen.winter at kdab.com (Allen Winter) Date: Tue, 05 Sep 2017 18:03:09 -0400 Subject: [Interest] [ANNOUNCE] GammaRay 2.8.1 Message-ID: <2878286.VXQKALvdV1@zazzy> Announcing GammaRay 2.8.1 Highlights of this Minor Release: =========================== * Fix compilation with interestingly packaged qmllint on Ubuntu. * Fix error handling in LLDB injector when LLDB scripting support is not available. * Fix tree view corruption when searching. * Fix linking with backward-cpp in BFD mode using a static libbdf. * Fix assert when launching a Windows target using a Qt debug build. * Fix crash during attaching on Windows. * Fix build with MSVC2010 and Qt4.8. * Fix attach dialog not seeing 32bit processes on 64bit Windows systems. * Add support for Qt 5.9.2. The source code can be found on GitHub at: https://github.com/KDAB/GammaRay Tarballs and zipballs for v2.8.1 are available from: https://github.com/KDAB/GammaRay/releases For Linux users: * we provide pre-built (against Qt5) packages for some popular distributions at: https://software.opensuse.org/package/gammaray * for those of you still using Qt4 we also provide the Qt4 probe from: https://software.opensuse.org/package/gammaray-qt4 For OS users: * we provide homebrew recipes for a Qt5 version and the Qt4 probe at https://github.com/KDAB/homebrew-tap For Windows users: * you'll need to build from source as we do not provide installers yet GammaRay is a tool to poke around in a Qt-application and also to manipulate the application to some extent. GammaRay uses various DLL injection techniques to hook into an application at runtime and provide access to a lot of interesting information. For more information about GammaRay, please visit http://www.kdab.com/gammaray and https://github.com/KDAB/GammaRay/wiki -- Allen Winter | allen.winter at kdab.com | Senior Software Engineer KDAB (USA) LLC, a KDAB Group company Tel. USA +1-866-777-KDAB(5322) ext3 KDAB - The Qt, C++ and OpenGL Experts -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4661 bytes Desc: not available URL: From abinjoseph1 at hotmail.com Wed Sep 6 08:34:46 2017 From: abinjoseph1 at hotmail.com (Abin Joseph) Date: Wed, 6 Sep 2017 06:34:46 +0000 Subject: [Interest] Unable to do Valgrind Memory Analyzer on i.MX6Q target In-Reply-To: <986519C39291F34C886B1EBB543F29635650C689@MBCHN01.chn.nestgroup.net> References: <986519C39291F34C886B1EBB543F29635650C689@MBCHN01.chn.nestgroup.net> Message-ID: Hi, I am trying to do valgrind memory analysis with Qt Creator on iMX6Q board. I am able to run the application from Qt creator and also able to run application from terminal using command /opt/HelloWorld -platform wayland I have valgrind in my target board, and if I run valgrind -v /opt/HelloWorld -platform wayland, it gives me whole logs which are so difficult to trace out the exact errors. But still application launches. But if I click Analyze->Valgrind Memory Analyzer on Qt Creator, it gives me error and the application does not launch. If I do not give any run time arguments I get the error: Analyzing memory of /opt/HelloWorld QML debugging is enabled. Only use this in a safe environment. This application failed to start because it could not find or load the Qt platform plugin "xcb" in "". Available platform plugins are: eglfs, linuxfb, minimal, minimalegl, offscreen, wayland-egl, wayland. Reinstalling the application may fix this problem. Analyzing finished. ** Process killed by signal ** If I do not give any run time arguments -platform wayland, I get the error: Analyzing memory of /opt/HelloWorld QML debugging is enabled. Only use this in a safe environment. Failed to create display (No such file or directory) Analyzing finished. ** Unknown error ** Is there something I am missing to run valgrind memory analysis on iMx6Q board? With Regards, Abin -------------- next part -------------- An HTML attachment was scrubbed... URL: From rmanni at gmail.com Wed Sep 6 09:57:34 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Wed, 6 Sep 2017 09:57:34 +0200 Subject: [Interest] state of qttranslations In-Reply-To: References: Message-ID: Hi, solving my own mystery, the thing which I did not understand is that catalog like qm files are referring other qm files which has the translations themselves, so those also need to be present to be able to load the catalogs. Problem solved, sorry for the noise :) Br, Robert On Tue, Sep 5, 2017 at 4:39 PM, Manner Róbert wrote: > Hi, > > I am trying to use qttranslations in my application to make common widgets > (like open dialog) have their translations. > I am compiling it by myself, standard way: qmake && make > I am trying to load the full translations qt_.qm > > Some of them works nearly fine like "hu", but most dont eg "en", "fr", > "de" etc. Those which do not work are very small qm files. Checking their > ts files, the difference seems to be that they are "catalogs" of smaller > ones. Running lrelease on them results in a message that 0 was translated: > > This is a catalog and does not work: > > $ lrelease qt_de.ts > Updating 'qt_de.qm'... > Generated 0 translation(s) (0 finished and 0 unfinished) > > This is not a catalog and does not work: > > $ lrelease qt_hu.ts > Updating 'qt_hu.qm'... > Generated 2131 translation(s) (2113 finished and 18 unfinished) > Ignored 100 untranslated source text(s) > > I observe the same with qm files shipped on archlinux. > > Does anyone know what do I misunderstand? How do one handle qt's > translations nowadays? > > Thanks for any help, > > Robert > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rmanni at gmail.com Wed Sep 6 10:29:47 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Wed, 6 Sep 2017 10:29:47 +0200 Subject: [Interest] Linux Deployment Question In-Reply-To: <909cd57c-fc34-989e-3edc-d9c9d2d40573@oliverniebuhr.de> References: <909cd57c-fc34-989e-3edc-d9c9d2d40573@oliverniebuhr.de> Message-ID: Hi, not sure if it is the same problem, but for us shipping the libxcb* libX11* did not work across linux distros. If you are not shipping those, then basically you are not able to ship such as libc, because then you will not be able to load these libraries from the system. Which worked is, that you compile what you ship with an old libc (eg on an old distro like centos6) so a newer libc can be used from the system. libxcb* libX* libc.so* libstdc++.so* libm.so* libpthread.so* librt.so* libGL.so* libglapi.so* libdl.so* libgcc_s.so* libGL.so* libexpat.so* libdrm.so* libselinux.so* This is our current skiplist which we do not ship ^ Br, Robert On Tue, Sep 5, 2017 at 8:34 AM, Oliver Niebuhr < googleersatz at oliverniebuhr.de> wrote: > Good morning. > > I think I have found now all necessary Files to deploy my App from > Antergos to other Distros. But in a clean Debian VM (GNU/Linux 9 > (stretch) 64 bit, GNOME 3.22.2, Gallium 0.4 on llvmpipe (llvm 3.9, 128 > bits)) i still get an Error when I try to start the Application. In my > Pro File I added this Line (Tip from the tripleboot.org Blog): > > linux:QMAKE_LFLAGS += -Wl,-rpath,"'\$$ORIGIN'" > > So the App looks for the Qt5 Libs in the Directory where it is called > from. Like expected. > > But no matter if I run the Binary directly or if I use this Script from > the Qt Deployment Wiki: > > #!/bin/sh > appname=`basename $0 | sed s,\.sh$,,` > > dirname=`dirname $0` > tmp="${dirname#?}" > > if [ "${dirname%$tmp}" != "/" ]; then > dirname=$PWD/$dirname > fi > LD_LIBRARY_PATH=$dirname > export LD_LIBRARY_PATH > $dirname/$appname "$@" > > I get this Error: > /home/test/AKKomPolTool/./AKKomPolTool: > Error: > > The Qt5 Main Libs are in the same Directory where the Binary is - so far > so good. And all the Plugins are under > /directory with binary/plugins/ > > I am not sure what I am still missing. Does a Debian User need to > install some additional Qt5 Stuff first to make the App work? Any hint > is much appreciated! > > This is the List of Files I am currently deploying (not sure if they are > all necessary [well of course i need the GCC Files]): > > QT5 LIBs... > /qt-buildx64/lib/libQt5Core.so.5 > /qt-buildx64/lib/libQt5DBus.so.5 > /qt-buildx64/lib/libQt5Gui.so.5 > /qt-buildx64/lib/libQt5Network.so.5 > /qt-buildx64/lib/libQt5PrintSupport.so.5 > /qt-buildx64/lib/libQt5Sql.so.5 > /qt-buildx64/lib/libQt5Widgets.so.5 > /qt-buildx64/lib/libQt5XcbQpa.so.5 > /qt-buildx64/lib/libQt5Xml.so.5 > > Bearer Plugins... > /qt-buildx64/plugins/bearer/libqconnmanbearer.soplugins/bearer/ > /qt-buildx64/plugins/bearer/libqgenericbearer.soplugins/bearer/ > /qt-buildx64/plugins/bearer/libqnmbearer.soplugins/bearer/ > > Imageformats Plugins... > /qt-buildx64/plugins/imageformats/libqgif.so > /qt-buildx64/plugins/imageformats/libqicns.so > /qt-buildx64/plugins/imageformats/libqico.so > /qt-buildx64/plugins/imageformats/libqjp2.so > /qt-buildx64/plugins/imageformats/libqjpeg.so > /qt-buildx64/plugins/imageformats/libqmng.so > /qt-buildx64/plugins/imageformats/libqtga.so > /qt-buildx64/plugins/imageformats/libqtiff.so > /qt-buildx64/plugins/imageformats/libqwbmp.so > /qt-buildx64/plugins/imageformats/libqwebp.so > > Platforminputcontexts Plugins... > /qt-buildx64/plugins/platforminputcontexts/libcomposeplatforminputcontext > plugin.so > /qt-buildx64/plugins/platforminputcontexts/libibusplatforminputcontextplu > gin.so > > Platforms Plugins... > /binaries/qt-buildx64/plugins/platforms/libqxcb.so > > Platformthemes Plugins... > /qt-buildx64/plugins/platformthemes/libqgtk3.so > > Printsupport Plugins... > /qt-buildx64/plugins/printsupport/libcupsprintersupport.so > > Sqldrivers Plugins... > /qt-buildx64/plugins/sqldrivers/libqsqlite.so > > Xcbglintegrations Plugins... > /qt-buildx64/plugins/xcbglintegrations/libqxcb-glx-integration.so > > #FIX Check if we must really deploy all that nonsense!! > 3RDPARTY AND SYSTEM LIBs... > /usr/lib64/libbz2.so.1.0 > /usr/lib64/libc.so.6 > /usr/lib64/libdbus-1.so.3 > /usr/lib64/libdl.so.2 > /usr/lib64/libdouble-conversion.so.1 > /usr/lib64/libexpat.so.1 > /usr/lib64/libfontconfig.so.1 > /usr/lib64/libfreetype.so.6 > /usr/lib64/libgcc_s.so.1 > /usr/lib64/libgcrypt.so.20 > /usr/lib64/libGL.so.1 > /usr/lib64/libGLdispatch.so.0 > /usr/lib64/libglib-2.0.so.0 > /usr/lib64/libGLX.so.0 > /usr/lib64/libgpg-error.so.0 > /usr/lib64/libgraphite2.so.3 > /usr/lib64/libgthread-2.0.so.0 > /usr/lib64/libharfbuzz.so.0 > /usr/lib64/libICE.so.6 > /usr/lib64/liblz4.so.1 > /usr/lib64/liblzma.so.5 > /usr/lib64/libm.so.6 > /usr/lib64/libpcre.so.1 > /usr/lib64/libpcre2-16.so.0 > /usr/lib64/libpng16.so.16 > /usr/lib64/libpthread.so.0 > /usr/lib64/librt.so.1 > /usr/lib64/libSM.so.6 > /usr/lib64/libstdc++.so.6 > /usr/lib64/libsystemd.so.0 > /usr/lib64/libuuid.so.1 > /usr/lib64/libX11.so.6 > /usr/lib64/libX11-xcb.so.1 > /usr/lib64/libXau.so.6 > /usr/lib64/libxcb.so.1 > /usr/lib64/libxcb-icccm.so.4 > /usr/lib64/libxcb-image.so.0 > /usr/lib64/libxcb-keysyms.so.1 > /usr/lib64/libxcb-randr.so.0 > /usr/lib64/libxcb-render.so.0 > /usr/lib64/libxcb-render-util.so.0 > /usr/lib64/libxcb-shape.so.0 > /usr/lib64/libxcb-shm.so.0 > /usr/lib64/libxcb-sync.so.1 > /usr/lib64/libxcb-util.so.1 > /usr/lib64/libxcb-xfixes.so.0 > /usr/lib64/libxcb-xinerama.so.0 > /usr/lib64/libxcb-xkb.so.1 > /usr/lib64/libXdmcp.so.6 > /usr/lib64/libXext.so.6 > /usr/lib64/libXi.so.6 > /usr/lib64/libxkbcommon.so.0 > /usr/lib64/libxkbcommon-x11.so.0 > /usr/lib64/libXrender.so.1 > /usr/lib64/libz.so.1 > > OpenSSL 1.1.0 LIBs... > /home/oliver/developing/binaries/opensslx64release/lib/libssl.so.1.1 > /home/oliver/developing/binaries/opensslx64release/lib/libcrypto.so.1.1 > > ICU 59.1 LIBs... > /home/oliver/developing/binaries/icux64release/lib/libicui18n.so.59 > /home/oliver/developing/binaries/icux64release/lib/libicuuc.so.59 > /home/oliver/developing/binaries/icux64release/lib/libicudata.so.59 > > Thank You very much! > Oliver > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kfunk at kde.org Wed Sep 6 12:13:10 2017 From: kfunk at kde.org (Kevin Funk) Date: Wed, 06 Sep 2017 12:13:10 +0200 Subject: [Interest] Linux Deployment Question In-Reply-To: <909cd57c-fc34-989e-3edc-d9c9d2d40573@oliverniebuhr.de> References: <909cd57c-fc34-989e-3edc-d9c9d2d40573@oliverniebuhr.de> Message-ID: <6811690.cJz4dLOTDi@kerberos> On Tuesday, 5 September 2017 08:34:15 CEST Oliver Niebuhr wrote: > Good morning. > > I think I have found now all necessary Files to deploy my App from > Antergos to other Distros. You might also want to have a look at the (relatively new) Linux deployment tool: https://github.com/probonopd/linuxdeployqt Maybe it makes deployment easier for you. Regards, Kevin > But in a clean Debian VM (GNU/Linux 9 > (stretch) 64 bit, GNOME 3.22.2, Gallium 0.4 on llvmpipe (llvm 3.9, 128 > bits)) i still get an Error when I try to start the Application. In my > Pro File I added this Line (Tip from the tripleboot.org Blog): > > linux:QMAKE_LFLAGS += -Wl,-rpath,"'\$$ORIGIN'" > > So the App looks for the Qt5 Libs in the Directory where it is called > from. Like expected. > > But no matter if I run the Binary directly or if I use this Script from > the Qt Deployment Wiki: > > #!/bin/sh > appname=`basename $0 | sed s,\.sh$,,` > > dirname=`dirname $0` > tmp="${dirname#?}" > > if [ "${dirname%$tmp}" != "/" ]; then > dirname=$PWD/$dirname > fi > LD_LIBRARY_PATH=$dirname > export LD_LIBRARY_PATH > $dirname/$appname "$@" > > I get this Error: > /home/test/AKKomPolTool/./AKKomPolTool: > Error: > > The Qt5 Main Libs are in the same Directory where the Binary is - so far > so good. And all the Plugins are under > /directory with binary/plugins/ > > I am not sure what I am still missing. Does a Debian User need to > install some additional Qt5 Stuff first to make the App work? Any hint > is much appreciated! > > This is the List of Files I am currently deploying (not sure if they are > all necessary [well of course i need the GCC Files]): > > QT5 LIBs... > /qt-buildx64/lib/libQt5Core.so.5 > /qt-buildx64/lib/libQt5DBus.so.5 > /qt-buildx64/lib/libQt5Gui.so.5 > /qt-buildx64/lib/libQt5Network.so.5 > /qt-buildx64/lib/libQt5PrintSupport.so.5 > /qt-buildx64/lib/libQt5Sql.so.5 > /qt-buildx64/lib/libQt5Widgets.so.5 > /qt-buildx64/lib/libQt5XcbQpa.so.5 > /qt-buildx64/lib/libQt5Xml.so.5 > > Bearer Plugins... > /qt-buildx64/plugins/bearer/libqconnmanbearer.soplugins/bearer/ > /qt-buildx64/plugins/bearer/libqgenericbearer.soplugins/bearer/ > /qt-buildx64/plugins/bearer/libqnmbearer.soplugins/bearer/ > > Imageformats Plugins... > /qt-buildx64/plugins/imageformats/libqgif.so > /qt-buildx64/plugins/imageformats/libqicns.so > /qt-buildx64/plugins/imageformats/libqico.so > /qt-buildx64/plugins/imageformats/libqjp2.so > /qt-buildx64/plugins/imageformats/libqjpeg.so > /qt-buildx64/plugins/imageformats/libqmng.so > /qt-buildx64/plugins/imageformats/libqtga.so > /qt-buildx64/plugins/imageformats/libqtiff.so > /qt-buildx64/plugins/imageformats/libqwbmp.so > /qt-buildx64/plugins/imageformats/libqwebp.so > > Platforminputcontexts Plugins... > /qt-buildx64/plugins/platforminputcontexts/libcomposeplatforminputcontextplu > gin.so > /qt-buildx64/plugins/platforminputcontexts/libibusplatforminputcontextplugi > n.so > > Platforms Plugins... > /binaries/qt-buildx64/plugins/platforms/libqxcb.so > > Platformthemes Plugins... > /qt-buildx64/plugins/platformthemes/libqgtk3.so > > Printsupport Plugins... > /qt-buildx64/plugins/printsupport/libcupsprintersupport.so > > Sqldrivers Plugins... > /qt-buildx64/plugins/sqldrivers/libqsqlite.so > > Xcbglintegrations Plugins... > /qt-buildx64/plugins/xcbglintegrations/libqxcb-glx-integration.so > > #FIX Check if we must really deploy all that nonsense!! > 3RDPARTY AND SYSTEM LIBs... > /usr/lib64/libbz2.so.1.0 > /usr/lib64/libc.so.6 > /usr/lib64/libdbus-1.so.3 > /usr/lib64/libdl.so.2 > /usr/lib64/libdouble-conversion.so.1 > /usr/lib64/libexpat.so.1 > /usr/lib64/libfontconfig.so.1 > /usr/lib64/libfreetype.so.6 > /usr/lib64/libgcc_s.so.1 > /usr/lib64/libgcrypt.so.20 > /usr/lib64/libGL.so.1 > /usr/lib64/libGLdispatch.so.0 > /usr/lib64/libglib-2.0.so.0 > /usr/lib64/libGLX.so.0 > /usr/lib64/libgpg-error.so.0 > /usr/lib64/libgraphite2.so.3 > /usr/lib64/libgthread-2.0.so.0 > /usr/lib64/libharfbuzz.so.0 > /usr/lib64/libICE.so.6 > /usr/lib64/liblz4.so.1 > /usr/lib64/liblzma.so.5 > /usr/lib64/libm.so.6 > /usr/lib64/libpcre.so.1 > /usr/lib64/libpcre2-16.so.0 > /usr/lib64/libpng16.so.16 > /usr/lib64/libpthread.so.0 > /usr/lib64/librt.so.1 > /usr/lib64/libSM.so.6 > /usr/lib64/libstdc++.so.6 > /usr/lib64/libsystemd.so.0 > /usr/lib64/libuuid.so.1 > /usr/lib64/libX11.so.6 > /usr/lib64/libX11-xcb.so.1 > /usr/lib64/libXau.so.6 > /usr/lib64/libxcb.so.1 > /usr/lib64/libxcb-icccm.so.4 > /usr/lib64/libxcb-image.so.0 > /usr/lib64/libxcb-keysyms.so.1 > /usr/lib64/libxcb-randr.so.0 > /usr/lib64/libxcb-render.so.0 > /usr/lib64/libxcb-render-util.so.0 > /usr/lib64/libxcb-shape.so.0 > /usr/lib64/libxcb-shm.so.0 > /usr/lib64/libxcb-sync.so.1 > /usr/lib64/libxcb-util.so.1 > /usr/lib64/libxcb-xfixes.so.0 > /usr/lib64/libxcb-xinerama.so.0 > /usr/lib64/libxcb-xkb.so.1 > /usr/lib64/libXdmcp.so.6 > /usr/lib64/libXext.so.6 > /usr/lib64/libXi.so.6 > /usr/lib64/libxkbcommon.so.0 > /usr/lib64/libxkbcommon-x11.so.0 > /usr/lib64/libXrender.so.1 > /usr/lib64/libz.so.1 > > OpenSSL 1.1.0 LIBs... > /home/oliver/developing/binaries/opensslx64release/lib/libssl.so.1.1 > /home/oliver/developing/binaries/opensslx64release/lib/libcrypto.so.1.1 > > ICU 59.1 LIBs... > /home/oliver/developing/binaries/icux64release/lib/libicui18n.so.59 > /home/oliver/developing/binaries/icux64release/lib/libicuuc.so.59 > /home/oliver/developing/binaries/icux64release/lib/libicudata.so.59 > > Thank You very much! > Oliver > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Kevin Funk | kfunk at kde.org | http://kfunk.org -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 163 bytes Desc: This is a digitally signed message part. URL: From qt-project.pyqt at infinite-drive.com Wed Sep 6 14:57:41 2017 From: qt-project.pyqt at infinite-drive.com (Michael Hartung) Date: Wed, 6 Sep 2017 14:57:41 +0200 Subject: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt-Includes - Final slash switch surprise... In-Reply-To: <2708603.8ogfWJitF7@tjmaciei-mobl1> References: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> <2708603.8ogfWJitF7@tjmaciei-mobl1> Message-ID: Hi Thiago, Thanks for your answer. Unfortunately its content does not make any sense. Or was it simply caused by an obsession to answer just any qt-interest email? And by that keeping others away from thinking about it. In fact nobody needs that! Michael Am 04.09.2017 um 14:02 schrieb Thiago Macieira: > On Monday, 4 September 2017 05:22:46 -03 Michael Hartung wrote: >> "C:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore/qglobal.h(45): fatal >> error C1083: Cannot open include file..." > > The ... here is the most important information: the file it couldn't open. > From Kai.Koehne at qt.io Wed Sep 6 15:39:45 2017 From: Kai.Koehne at qt.io (Kai Koehne) Date: Wed, 6 Sep 2017 13:39:45 +0000 Subject: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt-Includes - Final slash switch surprise... In-Reply-To: References: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> <2708603.8ogfWJitF7@tjmaciei-mobl1> Message-ID: > -----Original Message----- > From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > On Behalf Of Michael Hartung > Sent: Wednesday, September 6, 2017 2:58 PM > To: interest at qt-project.org > Subject: Re: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt- > Includes - Final slash switch surprise... > > Hi Thiago, > > Thanks for your answer. Unfortunately its content does not make any sense. > Or was it simply caused by an obsession to answer just any qt-interest email? > > And by that keeping others away from thinking about it. > > In fact nobody needs that! Thiago is completely right: You showed an error that the compiler cannot find an include file, but did leave out _which_ file it cannot find. You seem to think that it couldn't find "qglobal.h", but that's not the case - it is an #include in line 45 of qglobal.h that the compiler cannot resolve. Anyhow, peeking into the Qt 5.9 qglobal.h shows that it's most likely # include Which just happens to be the first system include. So it seems your MSVC environment is botched. Check the content of %INCLUDE%. For me it lists e.g. C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.10.25017\include , which is also where type_traits is found. Kai > Michael > > > Am 04.09.2017 um 14:02 schrieb Thiago Macieira: > > On Monday, 4 September 2017 05:22:46 -03 Michael Hartung wrote: > >> "C:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore/qglobal.h(45): fatal > >> error C1083: Cannot open include file..." > > > > The ... here is the most important information: the file it couldn't open. > > > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From jeanmichael.celerier at gmail.com Wed Sep 6 23:25:54 2017 From: jeanmichael.celerier at gmail.com (=?UTF-8?Q?Jean=2DMicha=C3=ABl_Celerier?=) Date: Wed, 6 Sep 2017 23:25:54 +0200 Subject: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt-Includes - Final slash switch surprise... In-Reply-To: References: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> <2708603.8ogfWJitF7@tjmaciei-mobl1> Message-ID: 'The ... here' refers to the ellipsis at the end of your error message. Best, ------- Jean-Michaël Celerier http://www.jcelerier.name On Wed, Sep 6, 2017 at 2:57 PM, Michael Hartung wrote: > Hi Thiago, > > Thanks for your answer. Unfortunately its content does not make any sense. > Or was it simply caused by an obsession to answer just any qt-interest > email? > > And by that keeping others away from thinking about it. > > In fact nobody needs that! > > Michael > > > Am 04.09.2017 um 14:02 schrieb Thiago Macieira: > >> On Monday, 4 September 2017 05:22:46 -03 Michael Hartung wrote: >> >>> "C:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore/qglobal.h(45): fatal >>> error C1083: Cannot open include file..." >>> >> >> The ... here is the most important information: the file it couldn't open. >> >> > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From qt-project.pyqt at infinite-drive.com Thu Sep 7 11:16:18 2017 From: qt-project.pyqt at infinite-drive.com (Michael Hartung) Date: Thu, 7 Sep 2017 11:16:18 +0200 Subject: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt-Includes - Final slash switch surprise... In-Reply-To: References: <0ae25bf2-e9f6-6104-fcd1-1d526acb3b50@infinite-drive.com> <2708603.8ogfWJitF7@tjmaciei-mobl1> Message-ID: <9d26f6f1-90eb-bbee-1bcf-b2016bc6c044@infinite-drive.com> Hi Kai, thanks for your more detailed answer which definitely better fits to my needs. My bad! Was Thiago definitely right! Michael Am 06.09.2017 um 15:39 schrieb Kai Koehne: > > >> -----Original Message----- >> From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] >> On Behalf Of Michael Hartung >> Sent: Wednesday, September 6, 2017 2:58 PM >> To: interest at qt-project.org >> Subject: Re: [Interest] Visual Studio 2015 / Qt 5.9.1 Application build - Qt- >> Includes - Final slash switch surprise... >> >> Hi Thiago, >> >> Thanks for your answer. Unfortunately its content does not make any sense. >> Or was it simply caused by an obsession to answer just any qt-interest email? >> >> And by that keeping others away from thinking about it. >> >> In fact nobody needs that! > > Thiago is completely right: You showed an error that the compiler cannot find an include file, but did leave out _which_ file it cannot find. You seem to think that it couldn't find "qglobal.h", but that's not the case - it is an #include in line 45 of qglobal.h that the compiler cannot resolve. > > Anyhow, peeking into the Qt 5.9 qglobal.h shows that it's most likely > > # include > > Which just happens to be the first system include. > > So it seems your MSVC environment is botched. Check the content of %INCLUDE%. For me it lists e.g. C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.10.25017\include , which is also where type_traits is found. > > Kai > >> Michael >> >> >> Am 04.09.2017 um 14:02 schrieb Thiago Macieira: >>> On Monday, 4 September 2017 05:22:46 -03 Michael Hartung wrote: >>>> "C:\Qt\Qt5.9.1\5.9.1\msvc2015_64\include\QtCore/qglobal.h(45): fatal >>>> error C1083: Cannot open include file..." >>> >>> The ... here is the most important information: the file it couldn't open. >>> >> >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > From coroberti at gmail.com Thu Sep 7 19:10:46 2017 From: coroberti at gmail.com (coroberti .) Date: Thu, 7 Sep 2017 20:10:46 +0300 Subject: [Interest] Qt for Android - The New Selection Handles Are Breaking Apps Message-ID: Hi, Qt port for Android was developed by the great and efforts of Bogdan, Eskil and many others where I mem it was Qt-5.2 and 5.3 where it has started to be usable and Qt-apps were developed. It was steadily improving by Qt-5.4, 5.5, 5.6 and 5.7 with many new features, options and native elements added. Unfortunately, my widget apps have been broken by the good intentions feature: Selection Handles - which appeared first in Qt-5.8, and it is still broken in the LTS intended Qt-5.9. Unlike most other features, it comes not on demand, but as a part of any text area, and breaks the existing Qt-apps. Needless to say that expectation of any developer could be an option to exclude such an intrusive feature. However, this is not the case - there's no any API to get rid from it. Android Text Editing Widgets are totally broken: https://bugreports.qt.io/browse/QTBUG-62426 and Selection Handles with QML have many issues opened with many workarounds proposed, like: https://appbus.wordpress.com/2017/03/25/text-selection-handles-qt-5-8/ and issues: https://bugreports.qt.io/browse/QTBUG-58700 and some more. Rock&Roll is dead, but we and QWidgets are not, not yet. Android apps are coming to Chromebook Desktop and look there superior when based on QWidgets. With other people joining me as seen in the tickets above, I'd like to ask the maintainers to consider giving priority to: 1. Option to exclude Selection Handles from the appearance; 2. Fixing the issues with Selection Handles; 3. Optional, making an option to customize their appearance. Thank you in advance! Kind regards, Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From smurphy at walbro.com Thu Sep 7 22:44:09 2017 From: smurphy at walbro.com (Murphy, Sean) Date: Thu, 7 Sep 2017 20:44:09 +0000 Subject: [Interest] QtQuick Menu position on Android Message-ID: <954F3AC9B636FB4F933D586E76E307EC6DF43318@USWCCEXC04> I'm extremely new to both QML development and Android development, so I may just be missing something obviously, but I'm struggling with a Menu's location, but only on the Android platform. I've got a settings button down in the lower right corner of my app. When clicked, it should popup() a Menu. On both the Windows and MacOS builds, whenever the button is clicked the menu pops up right near the button. But for some reason on Android, the menu pops up near the middle of the window. As shown in the code below, the menu is populated via a model from the C++ side, if that matters. I'm using Menu.popup(), which according to the documentation http://doc.qt.io/qt-5/qml-qtquick-controls-menu.html#popup-method, it supposedly " Opens this menu under the mouse cursor. It can block on some platforms, so test it accordingly." On Mac & Windows, that statement seems to hold true, but not on Android. Am I doing something wrong? Relevant QML: import QtQuick 2.2 import QtQuick.Window 2.1 import QtQuick.Controls 1.4 import QtQuick.Controls.Styles 1.4 import QtQuick.Extras 1.4 import QtQuick.Dialogs 1.2 import QtQuick.Layouts 1.3 import QtQml 2.2 Window { id: root visible: true width: 600 height: 1024 Menu { id: settingsMenu Instantiator { model: sModel MenuItem { text: model.display onTriggered: datafeed.slotSetSelectedPort(text) } onObjectAdded: { settingsMenu.insertItem(index, object) //console.log("QML: object added: " + object.text + ", index= " + index); } onObjectRemoved: { settingsMenu.removeItem(object) //console.log("QML: object removed: " + object.text); } } } Button { id: btSettings width: 40 height: 40 style: ButtonStyle { background: Rectangle { implicitWidth: 40 implicitHeight: 40 border.width: control.activeFocus ? 2 : 1 border.color: "#888888" radius: 4 gradient: Gradient { GradientStop { position: 0 color: control.pressed ? "#040404" : "#080808" } GradientStop { position: 1 color: control.pressed ? "#040404" : "#0a0a0a" } } } } x: parent.width - 60 y: parent.height - 60 onClicked: settingsMenu.popup() enabled: !datafeed.connected opacity: enabled ? 1.0 : 0.3 iconSource: "qrc:/images/settings_white.png" } } From sierdzio at gmail.com Fri Sep 8 09:14:07 2017 From: sierdzio at gmail.com (Tomasz Siekierda) Date: Fri, 8 Sep 2017 09:14:07 +0200 Subject: [Interest] Milo Code Database - kickstart your projects easily Message-ID: Hi Qt! We've just released Milo Code Database: a set of small, useful helper classes for Qt, easy to include and use - plus a scaffolding for new projects we call the New Project Template. At Milo, we use them in many of our projects to speed up development, and we figured they may be useful to the community as well. MCDB comes in 2 forms: 1. Installer - which helps to set up a new project in just a few seconds (with scaffolding for Qt app, documentation, CI integration, versioning, config, logger, and more). It can even set up a git repo and do the initial commit for you :-) 2. Set of independent git repos, easy to clone / download / use as submodule. So, what are the goodies? * New project template - template for all new Qt projects (QML and Widgets). Contains a scaffolding for Qt app with documentation, tests, resources etc. * MScripts - scripts for integration with GitLab CI, Mattermost, versioning scripts for all platforms (including git SHA), and more * MConfig - a small class which helps storing user app settings and configuration * MLog - small but powerful log class, with full categorized logging support * MRestAPI - handy set of base classes for talking with REST web servers * MBarcodeScanner - small helper class that allows to scan barcodes and QR codes using QZXing library * MCharts - implements some simple yet meaningful and eye friendly charts to display data * MSentry - remote logger support (Sentry/ Raven). Easy to turn on and off, just with a config switch * MCrypto - AES module, containing simple and convenient API, based on OpenSSL and it has a fallback if OSSL is not available License? MIT. We are actually open to other licenses, too, if somebody thinks that MIT is not proper in this case. Where can I get it? MCDB installers (Linux and Windows): https://seafile.milosolutions.com/d/2c50614e1e/ MCDB docs are available online: https://docs.milosolutions.com/milo-code-db/main/ And all submodules are listed and documented here: https://docs.milosolutions.com/milo-code-db/main/subprojects.html Each module's source code is on GitHub: https://github.com/milosolutions Contributing We're open to any comments, critique, pull requests etc. We can also help if any piece of the documentation is unclear. It's all being constantly updated and we're happy to get some external input. Feel free to reach me at sierdzio at gmail.com or more officially at ppm-qt at milosolutions.com Cheerio, sierdzio From thomas at silentwings.no Fri Sep 8 11:37:51 2017 From: thomas at silentwings.no (Thomas Sevaldrud) Date: Fri, 8 Sep 2017 11:37:51 +0200 Subject: [Interest] Access to QT_OPENGL_BUGLIST by QWindowsOpenGLTester methods Message-ID: Hi, I know that this question was up earlier this year on the list, but it never got any answer that I could find, I'm sorry if I missed it... Anyway, I need to access the OpenGL blacklist info, specifically I want to see if ANGLE has been disabled before I try to create an ANGLE GL Context In my application I am using the QCoreApplication attributes to control which type of GL context is created, for example when I need an ANGLE context: QCoreApplication::setAttribute(Qt::AA_UseOpenGLES, true); QCoreApplication::setAttribute(Qt::AA_UseDesktopOpenGL, false); QCoreApplication::setAttribute(Qt::AA_UseSoftwareOpenGL, false); These attributes seem to override the settings in the blacklist Now, if ANGLE has specifically been disabled through the blacklist file, I want to go straight for software opengl rendering instead. If I had access to the static methods in QWindowsOpenGLTester I could check this before setting the attributes. Is there a supported way to use this class from my project, or is there another way to achieve the same? Cheers -------------- next part -------------- An HTML attachment was scrubbed... URL: From h4nn35.work at gmail.com Fri Sep 8 13:33:52 2017 From: h4nn35.work at gmail.com (Johannes Pointner) Date: Fri, 8 Sep 2017 13:33:52 +0200 Subject: [Interest] QtWebEngine Linux navigator object Message-ID: Hi, I'm trying to get the info from the browser if the device supports multitouch. There is a navigator object maxTouchPoints (https://developer.mozilla.org/en-US/docs/Web/API/Navigator/maxTouchPoints) which would do the trick. This works with Chromium in Linux and with Chrome and QtWebEngine in Windows but not with the QtWebEngine in Linux (ArchLinux). Is this supported by the QtWebEngine under Linux? Is there something I missed (config build flag maybe)? Thanks, Hannes From alexandru.croitor at qt.io Fri Sep 8 13:55:11 2017 From: alexandru.croitor at qt.io (Alexandru Croitor) Date: Fri, 8 Sep 2017 11:55:11 +0000 Subject: [Interest] QtWebEngine Linux navigator object In-Reply-To: References: Message-ID: Hi, The relevant code that finds the maxTouchPoints value is int qtwebengine/src/3rdparty/chromium/ui/base/touch/touch_device_linux.cc inside the MaxTouchPoints() function. That function iterates over the available touchscreens (not touchpads though). Thus my initial guess is that the touchscreen is not detected for some reason. I would try to find where the list of touchscreen is populated and work from there. > On 8. Sep 2017, at 13:33, Johannes Pointner wrote: > > Hi, > > I'm trying to get the info from the browser if the device supports multitouch. > There is a navigator object maxTouchPoints > (https://developer.mozilla.org/en-US/docs/Web/API/Navigator/maxTouchPoints) > which would do the trick. > This works with Chromium in Linux and with Chrome and QtWebEngine in > Windows but not with the QtWebEngine in Linux (ArchLinux). > > Is this supported by the QtWebEngine under Linux? > Is there something I missed (config build flag maybe)? > > Thanks, > Hannes > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From chgans at gmail.com Sun Sep 10 03:26:47 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Sun, 10 Sep 2017 13:26:47 +1200 Subject: [Interest] Automatically build qtwebengine when building Qt5 Message-ID: Hi there, When i build qt5-32bits (5.6 branch) on my 64 bits linux workstation, qtwebengine doesn't get built/installed automatically. Are there any options i have to pass to init-repository and or configure? init-repository fetches qtwebengine source tree, but configure/make doesn't build/install it. I'm currently building qtwebengine "manually", and the only "trick" I had to do was a "export PKG_CONFIG_PATH=/usr/lib/i386-linux-gnu/pkgconfig/" as it wouldn't pick libpci-dev. Current build status: 10161/12525, hope to get fully built soon. Thanks, Chris From coroberti at gmail.com Sun Sep 10 10:51:42 2017 From: coroberti at gmail.com (coroberti .) Date: Sun, 10 Sep 2017 11:51:42 +0300 Subject: [Interest] Qt for Anroid - How to add dependencies to build.gradle? Message-ID: Hi, Is there a way to add to build.gradle some extra dependencies to the 3-rd libs besides manual editing of the file? Some trick in Qt project pro file to add? If find myself manually adding something like: compile 'com.android.support:support-v4:25.+' compile 'com.google.http-client:google-http-client-gson:1.22.0' Thanks in advance! Kind regards, Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From coroberti at gmail.com Sun Sep 10 18:24:47 2017 From: coroberti at gmail.com (coroberti .) Date: Sun, 10 Sep 2017 19:24:47 +0300 Subject: [Interest] Qt for Anroid - How to add dependencies to build.gradle? In-Reply-To: <8eb23659-63cf-e332-4ac8-8efa5a263ba4@gmx.de> References: <8eb23659-63cf-e332-4ac8-8efa5a263ba4@gmx.de> Message-ID: Yes, it works! Thank you very much! Kind regards, Robert On Sun, Sep 10, 2017 at 3:43 PM, noname wrote: > hi, > > you can add a custom directory to your pro-file of which all contained > files (and subdirectories) will be merged into the generated "android" > folder. the generated one will be replaced with the provided "custom" ones > if present, i think. > should be "ANDROID_PACKAGE_SOURCE_DIR" in the pro file. more info in the > docs: http://doc.qt.io/qt-5/deployment-android.html# > android-specific-qmake-variables. > that way i'm also adding custom dependencies using a custom "build.gradle" > (full file, i.e. i copied the entire generated build.gradle to my > "customized_android" directory, then added my custom lines). > > > kind regards > > > On 10.09.2017 10:51, coroberti . wrote: > > Hi, > Is there a way to add to build.gradle some extra dependencies to the 3-rd > libs > besides manual editing of the file? > > Some trick in Qt project pro file to add? > > If find myself manually adding something like: > > compile 'com.android.support:support-v4:25.+' > compile 'com.google.http-client:google-http-client-gson:1.22.0' > > Thanks in advance! > > Kind regards, > Robert > > > _______________________________________________ > Interest mailing listInterest at qt-project.orghttp://lists.qt-project.org/mailman/listinfo/interest > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Kai.Koehne at qt.io Mon Sep 11 11:22:22 2017 From: Kai.Koehne at qt.io (Kai Koehne) Date: Mon, 11 Sep 2017 09:22:22 +0000 Subject: [Interest] Automatically build qtwebengine when building Qt5 In-Reply-To: References: Message-ID: > -----Original Message----- > From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > On Behalf Of Christian Gagneraud > Sent: Sunday, September 10, 2017 3:27 AM > To: interest at qt-project.org > Subject: [Interest] Automatically build qtwebengine when building Qt5 > > Hi there, > > When i build qt5-32bits (5.6 branch) on my 64 bits linux workstation, > qtwebengine doesn't get built/installed automatically. > > Are there any options i have to pass to init-repository and or configure? If you didn't pass anything special to init-repository and/or configure, qtwebengine should be built automatically. What can happen though is that - Another module fails to build, and make didn't even reach qtwebengine. Inspect your build log closely, sometimes this can be quite hidden. - The qmake/configure logic in qtwebengine failed a precondition, and qtwebengine got skipped. Again, this should show up in the log. > init-repository fetches qtwebengine source tree, but configure/make > doesn't build/install it. > > I'm currently building qtwebengine "manually", and the only "trick" I had to > do was a "export PKG_CONFIG_PATH=/usr/lib/i386-linux-gnu/pkgconfig/" as > it wouldn't pick libpci-dev. Make sure that you set this environment variable before running top-level configure/qmake then. Regards Kai From daniel.hein at adbsafegate.com Mon Sep 11 15:24:49 2017 From: daniel.hein at adbsafegate.com (Daniel Hein) Date: Mon, 11 Sep 2017 15:24:49 +0200 Subject: [Interest] QJSEngine/QJSValue setProperty and deleteProperty Question Message-ID: Hi, I have a question concerning the exact semantics of QJSValue::setProperty and QJSValue::deleteProperty functions. Specifically, what happens if I do something akin to this example: *) I evaluate a script using the QJSEngine::evaluate() function. As a consequence I now have a set of properties in the global context of the script engine corresponding to the script defined properties, such as objects and functions *) I copy? one of these functions into a container object living in the global context using container.setProperty("functionName", function) *) I delete the function object in the global context global.deleteProperty("functionName") My question is, does deleting the property "functionName" in the global context also invalidate the copy in the container? According to the QJSValue documentation: > Note that a QJSValue for which isObject() is true only carries a > reference to an actual object; copying the QJSValue will only copy the > object reference, not the object itself. If you want to clone an > object (i.e. copy an object's properties to another object), you can > do so with the help of a for-in statement in script code, or > QJSValueIterator in C++. copying a QJSValue results in a flat copy, if the QJSValue wraps an object. In JavaScript functions are objects. So, I am unsure, if the global.deleteProperty("functionName") call just deletes this particular object reference, or the object itself. Thanks for any help, Daniel From thiago.macieira at intel.com Mon Sep 11 14:36:50 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Mon, 11 Sep 2017 07:36:50 -0500 Subject: [Interest] Automatically build qtwebengine when building Qt5 In-Reply-To: References: Message-ID: <3168278.WCYcnW6rfm@tjmaciei-mobl1> On Monday, 11 September 2017 04:22:22 CDT Kai Koehne wrote: > - Another module fails to build, and make didn't even reach qtwebengine. > Inspect your build log closely, sometimes this can be quite hidden. This happens more often than you'd realise. I've seen people as this very same question on IRC and then I ask them to "echo $?" after make. They suddenly realise the build ended with an error. Anyway, here's what I usually do when building Qt: make -k || make -j1 That will build as far as it can before giving up, then if there was an error, it will stop on the first error. Fix the error, then rinse and repeat until success. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From h4nn35.work at gmail.com Wed Sep 13 08:41:49 2017 From: h4nn35.work at gmail.com (Johannes Pointner) Date: Wed, 13 Sep 2017 08:41:49 +0200 Subject: [Interest] QtWebEngine Linux navigator object In-Reply-To: References: Message-ID: Hi Alexandru, thanks for the hint. I added a few debug messages and saw that the chromium code doesn't know that there is a touch device available(touch_device_linux.cc:IsTouchDevicePresent() ) -> so maxTouchPoints returns obviously zero. Also the InputDeviceManager has no keyboard or mouse device. So I suspect that under Linux Qt gets the information about the input devices and the chromium part never gets the events about the input devices? Could this be the case? Also I think I know why this works on Windows, because they don't use there the InputDeviceManager to get this information instead they call system functions. On Fri, Sep 8, 2017 at 1:55 PM, Alexandru Croitor wrote: > Hi, > > The relevant code that finds the maxTouchPoints value is int qtwebengine/src/3rdparty/chromium/ui/base/touch/touch_device_linux.cc inside the MaxTouchPoints() function. > > That function iterates over the available touchscreens (not touchpads though). Thus my initial guess is that the touchscreen is not detected for some reason. I would try to find where the list of touchscreen is populated and work from there. > > >> On 8. Sep 2017, at 13:33, Johannes Pointner wrote: >> >> Hi, >> >> I'm trying to get the info from the browser if the device supports multitouch. >> There is a navigator object maxTouchPoints >> (https://developer.mozilla.org/en-US/docs/Web/API/Navigator/maxTouchPoints) >> which would do the trick. >> This works with Chromium in Linux and with Chrome and QtWebEngine in >> Windows but not with the QtWebEngine in Linux (ArchLinux). >> >> Is this supported by the QtWebEngine under Linux? >> Is there something I missed (config build flag maybe)? >> >> Thanks, >> Hannes >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > From elvstone at gmail.com Wed Sep 13 09:26:34 2017 From: elvstone at gmail.com (Elvis Stansvik) Date: Wed, 13 Sep 2017 09:26:34 +0200 Subject: [Interest] What's the purpose of Q_DECLARE_INTERFACE? Message-ID: I'm reading about extending applications with Qt plugins. What is the purpose of the Q_DECLARE_INTERFACE/Q_INTERFACES mechanism? The docs says: "This macro associates the given Identifier (a string literal) to the interface class called ClassName." [1] and "Use the Q_DECLARE_INTERFACE() macro to tell Qt's meta-object system about the interface." [2] The Plugins wiki page also has some more info [3]: "Q_DECLARE_INTERFACE is a macro that defines helper functions that make qobject_cast(hammer) return a Tool pointer. qobject_cast is just a template function and you can think of Q_DECLARE_INTERFACE providing a template specialization for the interface. The macro itself just expands as a call to qt_metacast that moc generated above. So, Q_DECLARE_INTEFACE defines qobject_cast(object) which is a template specialization that translates to object->qt_metacall("in.forwardbias.tool/1.0");" (and also some info about the Q_INTERFACES macro: "When moc runs on the hammer.h code, it inspects Q_INTERFACES. It generates code for a function called qt_metacall - void *Hammer::qt_metacast(const char *iname). This goal of this 'casting' function is to return a pointer to an interface depending on iname. moc also verifies whether the interface names that you have put inside Q_INTERFACES have indeed been declared. It can do this by inspecting the header files and looking for a Q_DECLARE_INTERFACE.") So from that I gather that the purpose of the mechanism is to make qobject_cast(..) work, despite SomeInterface not being a QObject? But none of the docs explain why this is needed. What is this featured used for within Qt? Or is it provided for me as an application developer? If so, when would I use it? Sorry if the question is dumb, but I think all the docs explain what the macros do, not why you would want that or why it is needed. And a related question: Can I use Q_DECLARE_INTERFACE on a class that is not actually an interface, a class that does not have 100% pure virtual member functions? E.g. if I want some base functionality in my plugin "interface" class? (I realize of course that my plugins would then have to link against the lib that provides this "interface", so it would be a tighter coupling than just a shared header, but nonetheless). Thanks in advance, Elvis [1] http://doc.qt.io/qt-5/qtplugin.html#Q_DECLARE_INTERFACE [2] http://doc.qt.io/qt-5/plugins-howto.html#the-low-level-api-extending-qt-applications [3] https://wiki.qt.io/Plugins From alexandru.croitor at qt.io Wed Sep 13 10:31:31 2017 From: alexandru.croitor at qt.io (Alexandru Croitor) Date: Wed, 13 Sep 2017 08:31:31 +0000 Subject: [Interest] QtWebEngine Linux navigator object In-Reply-To: References: Message-ID: > On 13. Sep 2017, at 08:41, Johannes Pointner wrote: > > Hi Alexandru, > > thanks for the hint. I added a few debug messages and saw that the > chromium code doesn't know that there is a touch device > available(touch_device_linux.cc:IsTouchDevicePresent() ) -> so > maxTouchPoints returns obviously zero. Also the InputDeviceManager has > no keyboard or mouse device. > > So I suspect that under Linux Qt gets the information about the input > devices and the chromium part never gets the events about the input > devices? Could this be the case? Afaik we don't pass information about touch devices from Qt to Chromium. Chromium detects them by itself. Obviously the detection code will be different between Qt and Chromium, thus different results. It is possible that we are also not calling some code that Chromium expects us to call to properly detect touch devices. It's hard to say. You'll have to dive into the code with a debugger and figure that out. > Also I think I know why this works on Windows, because they don't use > there the InputDeviceManager to get this information instead they call > system functions. > > > > On Fri, Sep 8, 2017 at 1:55 PM, Alexandru Croitor > wrote: >> Hi, >> >> The relevant code that finds the maxTouchPoints value is int qtwebengine/src/3rdparty/chromium/ui/base/touch/touch_device_linux.cc inside the MaxTouchPoints() function. >> >> That function iterates over the available touchscreens (not touchpads though). Thus my initial guess is that the touchscreen is not detected for some reason. I would try to find where the list of touchscreen is populated and work from there. >> >> >>> On 8. Sep 2017, at 13:33, Johannes Pointner wrote: >>> >>> Hi, >>> >>> I'm trying to get the info from the browser if the device supports multitouch. >>> There is a navigator object maxTouchPoints >>> (https://developer.mozilla.org/en-US/docs/Web/API/Navigator/maxTouchPoints) >>> which would do the trick. >>> This works with Chromium in Linux and with Chrome and QtWebEngine in >>> Windows but not with the QtWebEngine in Linux (ArchLinux). >>> >>> Is this supported by the QtWebEngine under Linux? >>> Is there something I missed (config build flag maybe)? >>> >>> Thanks, >>> Hannes >>> _______________________________________________ >>> Interest mailing list >>> Interest at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/interest >> From elvstone at gmail.com Wed Sep 13 10:32:34 2017 From: elvstone at gmail.com (Elvis Stansvik) Date: Wed, 13 Sep 2017 10:32:34 +0200 Subject: [Interest] What's the purpose of Q_DECLARE_INTERFACE? In-Reply-To: References: Message-ID: (Adding back the interest@ list to recipients) 2017-09-13 10:15 GMT+02:00 Christian Gagneraud : > > > On 13/09/2017 7:26 pm, "Elvis Stansvik" wrote: > > I'm reading about extending applications with Qt plugins. > > What is the purpose of the Q_DECLARE_INTERFACE/Q_INTERFACES mechanism? > > The docs says: > > "This macro associates the given Identifier (a string literal) to > the interface class called ClassName." [1] > > and > > "Use the Q_DECLARE_INTERFACE() macro to tell Qt's meta-object > system about the interface." [2] > > The Plugins wiki page also has some more info [3]: > > "Q_DECLARE_INTERFACE is a macro that defines helper functions that > make qobject_cast(hammer) return a Tool pointer. qobject_cast > is just a template function and you can think of Q_DECLARE_INTERFACE > providing a template specialization for the interface. The macro > itself just expands as a call to qt_metacast that moc generated above. > So, Q_DECLARE_INTEFACE defines qobject_cast(object) which is a > template specialization that translates to > object->qt_metacall("in.forwardbias.tool/1.0");" > > (and also some info about the Q_INTERFACES macro: "When moc runs on > the hammer.h code, it inspects Q_INTERFACES. It generates code for a > function called qt_metacall - void *Hammer::qt_metacast(const char > *iname). This goal of this 'casting' function is to return a pointer > to an interface depending on iname. moc also verifies whether the > interface names that you have put inside Q_INTERFACES have indeed been > declared. It can do this by inspecting the header files and looking > for a Q_DECLARE_INTERFACE.") > > So from that I gather that the purpose of the mechanism is to make > qobject_cast(..) work, despite SomeInterface not being > a QObject? > > But none of the docs explain why this is needed. What is this featured > used for within Qt? Or is it provided for me as an application > developer? If so, when would I use it? > > Sorry if the question is dumb, but I think all the docs explain what > the macros do, not why you would want that or why it is needed. > > > I think there are 2 things here: > > - the macro alows you to tag a QObject as implementing an interface with a > symbolic name of your own (not a mangled c++ symbol) and a version of your > own (which c++ doesn't provide at all). > - qobject_cast is (way) faster than dynamic_cast > > The versioning and the symbolic naming gives host application way more > freedom than dynamic_cast alone (fancy rename an interface class name each > time you change it in a binary incompatible way?) > > Basically, with this mechanism, the linker doesn't interfer with how you > want to cast your QObject. You are in full control to decide if it is safe > or not to cast qobects. > > My 2 cents, I'm not an expert, but like you I have interest in all of these, > please anyone, correct me if I'm wrong. Thanks a lot Chris. That makes a lot of sense. I had not thought about the versioning/greater control aspect. With versioning, I guess you mean baking a version number of your choosing into the IID string? Elvis > > Chris > > > And a related question: Can I use Q_DECLARE_INTERFACE on a class that > is not actually an interface, a class that does not have 100% pure > virtual member functions? E.g. if I want some base functionality in my > plugin "interface" class? (I realize of course that my plugins would > then have to link against the lib that provides this "interface", so > it would be a tighter coupling than just a shared header, but > nonetheless). > > Thanks in advance, > Elvis > > [1] http://doc.qt.io/qt-5/qtplugin.html#Q_DECLARE_INTERFACE > [2] > http://doc.qt.io/qt-5/plugins-howto.html#the-low-level-api-extending-qt-applications > [3] https://wiki.qt.io/Plugins > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > > From chgans at gmail.com Wed Sep 13 10:39:24 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Wed, 13 Sep 2017 20:39:24 +1200 Subject: [Interest] What's the purpose of Q_DECLARE_INTERFACE? In-Reply-To: References: Message-ID: On 13/09/2017 8:32 pm, "Elvis Stansvik" wrote: (Adding back the interest@ list to recipients) Thanks for doing that, I never reply in private without good reason. But I used my phone to answer your message, so I guess something went wrong... Chris 2017-09-13 10:15 GMT+02:00 Christian Gagneraud : > > > On 13/09/2017 7:26 pm, "Elvis Stansvik" wrote: > > I'm reading about extending applications with Qt plugins. > > What is the purpose of the Q_DECLARE_INTERFACE/Q_INTERFACES mechanism? > > The docs says: > > "This macro associates the given Identifier (a string literal) to > the interface class called ClassName." [1] > > and > > "Use the Q_DECLARE_INTERFACE() macro to tell Qt's meta-object > system about the interface." [2] > > The Plugins wiki page also has some more info [3]: > > "Q_DECLARE_INTERFACE is a macro that defines helper functions that > make qobject_cast(hammer) return a Tool pointer. qobject_cast > is just a template function and you can think of Q_DECLARE_INTERFACE > providing a template specialization for the interface. The macro > itself just expands as a call to qt_metacast that moc generated above. > So, Q_DECLARE_INTEFACE defines qobject_cast(object) which is a > template specialization that translates to > object->qt_metacall("in.forwardbias.tool/1.0");" > > (and also some info about the Q_INTERFACES macro: "When moc runs on > the hammer.h code, it inspects Q_INTERFACES. It generates code for a > function called qt_metacall - void *Hammer::qt_metacast(const char > *iname). This goal of this 'casting' function is to return a pointer > to an interface depending on iname. moc also verifies whether the > interface names that you have put inside Q_INTERFACES have indeed been > declared. It can do this by inspecting the header files and looking > for a Q_DECLARE_INTERFACE.") > > So from that I gather that the purpose of the mechanism is to make > qobject_cast(..) work, despite SomeInterface not being > a QObject? > > But none of the docs explain why this is needed. What is this featured > used for within Qt? Or is it provided for me as an application > developer? If so, when would I use it? > > Sorry if the question is dumb, but I think all the docs explain what > the macros do, not why you would want that or why it is needed. > > > I think there are 2 things here: > > - the macro alows you to tag a QObject as implementing an interface with a > symbolic name of your own (not a mangled c++ symbol) and a version of your > own (which c++ doesn't provide at all). > - qobject_cast is (way) faster than dynamic_cast > > The versioning and the symbolic naming gives host application way more > freedom than dynamic_cast alone (fancy rename an interface class name each > time you change it in a binary incompatible way?) > > Basically, with this mechanism, the linker doesn't interfer with how you > want to cast your QObject. You are in full control to decide if it is safe > or not to cast qobects. > > My 2 cents, I'm not an expert, but like you I have interest in all of these, > please anyone, correct me if I'm wrong. Thanks a lot Chris. That makes a lot of sense. I had not thought about the versioning/greater control aspect. With versioning, I guess you mean baking a version number of your choosing into the IID string? Elvis > > Chris > > > And a related question: Can I use Q_DECLARE_INTERFACE on a class that > is not actually an interface, a class that does not have 100% pure > virtual member functions? E.g. if I want some base functionality in my > plugin "interface" class? (I realize of course that my plugins would > then have to link against the lib that provides this "interface", so > it would be a tighter coupling than just a shared header, but > nonetheless). > > Thanks in advance, > Elvis > > [1] http://doc.qt.io/qt-5/qtplugin.html#Q_DECLARE_INTERFACE > [2] > http://doc.qt.io/qt-5/plugins-howto.html#the-low-level-api- extending-qt-applications > [3] https://wiki.qt.io/Plugins > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From chgans at gmail.com Wed Sep 13 10:48:05 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Wed, 13 Sep 2017 20:48:05 +1200 Subject: [Interest] What's the purpose of Q_DECLARE_INTERFACE? In-Reply-To: References: Message-ID: On 13/09/2017 8:32 pm, "Elvis Stansvik" wrote: (Adding back the interest@ list to recipients) 2017-09-13 10:15 GMT+02:00 Christian Gagneraud : > > > On 13/09/2017 7:26 pm, "Elvis Stansvik" wrote: > > I'm reading about extending applications with Qt plugins. > > What is the purpose of the Q_DECLARE_INTERFACE/Q_INTERFACES mechanism? > > The docs says: > > "This macro associates the given Identifier (a string literal) to > the interface class called ClassName." [1] > > and > > "Use the Q_DECLARE_INTERFACE() macro to tell Qt's meta-object > system about the interface." [2] > > The Plugins wiki page also has some more info [3]: > > "Q_DECLARE_INTERFACE is a macro that defines helper functions that > make qobject_cast(hammer) return a Tool pointer. qobject_cast > is just a template function and you can think of Q_DECLARE_INTERFACE > providing a template specialization for the interface. The macro > itself just expands as a call to qt_metacast that moc generated above. > So, Q_DECLARE_INTEFACE defines qobject_cast(object) which is a > template specialization that translates to > object->qt_metacall("in.forwardbias.tool/1.0");" > > (and also some info about the Q_INTERFACES macro: "When moc runs on > the hammer.h code, it inspects Q_INTERFACES. It generates code for a > function called qt_metacall - void *Hammer::qt_metacast(const char > *iname). This goal of this 'casting' function is to return a pointer > to an interface depending on iname. moc also verifies whether the > interface names that you have put inside Q_INTERFACES have indeed been > declared. It can do this by inspecting the header files and looking > for a Q_DECLARE_INTERFACE.") > > So from that I gather that the purpose of the mechanism is to make > qobject_cast(..) work, despite SomeInterface not being > a QObject? > > But none of the docs explain why this is needed. What is this featured > used for within Qt? Or is it provided for me as an application > developer? If so, when would I use it? > > Sorry if the question is dumb, but I think all the docs explain what > the macros do, not why you would want that or why it is needed. > > > I think there are 2 things here: > > - the macro alows you to tag a QObject as implementing an interface with a > symbolic name of your own (not a mangled c++ symbol) and a version of your > own (which c++ doesn't provide at all). > - qobject_cast is (way) faster than dynamic_cast > > The versioning and the symbolic naming gives host application way more > freedom than dynamic_cast alone (fancy rename an interface class name each > time you change it in a binary incompatible way?) > > Basically, with this mechanism, the linker doesn't interfer with how you > want to cast your QObject. You are in full control to decide if it is safe > or not to cast qobects. > > My 2 cents, I'm not an expert, but like you I have interest in all of these, > please anyone, correct me if I'm wrong. Thanks a lot Chris. That makes a lot of sense. I had not thought about the versioning/greater control aspect. With versioning, I guess you mean baking a version number of your choosing into the IID string? I think some Qt examples show it, yes you can have an interface named com.company/project/interface/1.2 if you want. Up to you! Welcome freedom! ;) Chris Elvis > > Chris > > > And a related question: Can I use Q_DECLARE_INTERFACE on a class that > is not actually an interface, a class that does not have 100% pure > virtual member functions? E.g. if I want some base functionality in my > plugin "interface" class? (I realize of course that my plugins would > then have to link against the lib that provides this "interface", so > it would be a tighter coupling than just a shared header, but > nonetheless). > > Thanks in advance, > Elvis > > [1] http://doc.qt.io/qt-5/qtplugin.html#Q_DECLARE_INTERFACE > [2] > http://doc.qt.io/qt-5/plugins-howto.html#the-low-level-api- extending-qt-applications > [3] https://wiki.qt.io/Plugins > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From coroberti at gmail.com Wed Sep 13 12:58:13 2017 From: coroberti at gmail.com (coroberti .) Date: Wed, 13 Sep 2017 13:58:13 +0300 Subject: [Interest] Qt for Android - Signed Debug Packages by QtCreator Message-ID: Hi, androiddeployqt with flag --sign passed by QtCreator makes release package and never debug package. Sometimes, there is a need to bug Qt Android software from QtCreator using a debug package that could work only if it is signed, i.e. when using Google's API where authentication uses signature of the app. Is where a way to workaround the current settings and make debugging from QtCreator by using singed debug app without brutal patching of androiddeployqt? Thank you in advance! Kind regards, Robert From jhihn at gmx.com Wed Sep 13 17:55:44 2017 From: jhihn at gmx.com (Jason H) Date: Wed, 13 Sep 2017 17:55:44 +0200 Subject: [Interest] [Android-development] Qt for Android - Signed Debug Packages by QtCreator In-Reply-To: References: Message-ID: I filed a bug about this. Vote for it. However the result for me was a debug package, just incorrectly named. > Sent: Wednesday, September 13, 2017 at 6:58 AM > From: "coroberti ." > To: "interest at qt-project.org" , android-development > Subject: [Android-development] Qt for Android - Signed Debug Packages by QtCreator > > Hi, > androiddeployqt with flag --sign passed by QtCreator makes > release package and never debug package. > > Sometimes, there is a need to bug Qt Android software from QtCreator > using a debug package that could work only if it is signed, i.e. when > using Google's API > where authentication uses signature of the app. > > Is where a way to workaround the current settings and make debugging > from QtCreator by > using singed debug app without brutal patching of androiddeployqt? > > Thank you in advance! > > Kind regards, > Robert > _______________________________________________ > Android-development mailing list > Android-development at qt-project.org > http://lists.qt-project.org/mailman/listinfo/android-development > From coroberti at gmail.com Wed Sep 13 18:29:33 2017 From: coroberti at gmail.com (coroberti .) Date: Wed, 13 Sep 2017 19:29:33 +0300 Subject: [Interest] [Android-development] Qt for Android - Signed Debug Packages by QtCreator In-Reply-To: References: Message-ID: Dear Jason, What is the bug report number or link to vote? Thanks! Kind regards, Robert On Wed, Sep 13, 2017 at 6:55 PM, Jason H wrote: > I filed a bug about this. Vote for it. > > However the result for me was a debug package, just incorrectly named. > > >> Sent: Wednesday, September 13, 2017 at 6:58 AM >> From: "coroberti ." >> To: "interest at qt-project.org" , android-development >> Subject: [Android-development] Qt for Android - Signed Debug Packages by QtCreator >> >> Hi, >> androiddeployqt with flag --sign passed by QtCreator makes >> release package and never debug package. >> >> Sometimes, there is a need to bug Qt Android software from QtCreator >> using a debug package that could work only if it is signed, i.e. when >> using Google's API >> where authentication uses signature of the app. >> >> Is where a way to workaround the current settings and make debugging >> from QtCreator by >> using singed debug app without brutal patching of androiddeployqt? >> >> Thank you in advance! >> >> Kind regards, >> Robert >> _______________________________________________ >> Android-development mailing list >> Android-development at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/android-development >> From jhihn at gmx.com Wed Sep 13 23:39:48 2017 From: jhihn at gmx.com (Jason H) Date: Wed, 13 Sep 2017 23:39:48 +0200 Subject: [Interest] [Android-development] Qt for Android - Signed Debug Packages by QtCreator In-Reply-To: References: Message-ID: https://bugreports.qt.io/browse/QTBUG-58799 > Sent: Wednesday, September 13, 2017 at 12:29 PM > From: "coroberti ." > To: "Jason H" > Cc: "interest at qt-project.org" , android-development > Subject: Re: [Android-development] Qt for Android - Signed Debug Packages by QtCreator > > Dear Jason, > What is the bug report number or link to vote? > > Thanks! > Kind regards, > Robert > > > On Wed, Sep 13, 2017 at 6:55 PM, Jason H wrote: > > I filed a bug about this. Vote for it. > > > > However the result for me was a debug package, just incorrectly named. > > > > > >> Sent: Wednesday, September 13, 2017 at 6:58 AM > >> From: "coroberti ." > >> To: "interest at qt-project.org" , android-development > >> Subject: [Android-development] Qt for Android - Signed Debug Packages by QtCreator > >> > >> Hi, > >> androiddeployqt with flag --sign passed by QtCreator makes > >> release package and never debug package. > >> > >> Sometimes, there is a need to bug Qt Android software from QtCreator > >> using a debug package that could work only if it is signed, i.e. when > >> using Google's API > >> where authentication uses signature of the app. > >> > >> Is where a way to workaround the current settings and make debugging > >> from QtCreator by > >> using singed debug app without brutal patching of androiddeployqt? > >> > >> Thank you in advance! > >> > >> Kind regards, > >> Robert > >> _______________________________________________ > >> Android-development mailing list > >> Android-development at qt-project.org > >> http://lists.qt-project.org/mailman/listinfo/android-development > >> > From chgans at gmail.com Thu Sep 14 03:51:52 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Thu, 14 Sep 2017 13:51:52 +1200 Subject: [Interest] Layouting children on Resize event vs LayoutRequest event Message-ID: Hi there, I am observing a fundamental behaviour difference when moving/resizing children widgets in resizeEvent vs doing the same on a LayoutRequest event. Basically, if i move the children on Resize event, Qt will triggers events in 2 bunches (traced with QApplication ::notify() overload), the first bunch contains move/resize events for the parent widget (along with other widgets), the second bunch will contains the recently moved/resized children events. I don't want this behaviour. If i move/resize children on LayoutRequest event, all widgets will get their resize/move events in a single bunch. I want this behaviour. And my question is: Is this behaviour guarenteed by Qt? Or could this behaviour change at any moment? The documentation [1] doesn't say anything about that. Could/should the documentation be updated to reflect this difference? I think that re-layouting children on LayoutRequest is actually more efficient. as it gives Qt an opportunity to optimise out move/resize/paint events. Thanks, Chris [1] http://doc.qt.io/qt-5/layout.html#manual-layout From coroberti at gmail.com Thu Sep 14 06:28:19 2017 From: coroberti at gmail.com (coroberti .) Date: Thu, 14 Sep 2017 07:28:19 +0300 Subject: [Interest] [Android-development] Qt for Android - Signed Debug Packages by QtCreator In-Reply-To: References: Message-ID: Yes, voted. This is not only the name being wrong. You can look at the code of androiddeplyqt to see what happens and what is the difference between debug package and release. When --sign is passed by QtCreator, this signals to androiddeplyqt to make the release package, and this is where the wrong name comes. Release means that gdb server not to be deployed and the package cannot be used for placing and break-point and making any debugging. Thanks. Kind regards, Robert On Thu, Sep 14, 2017 at 12:39 AM, Jason H wrote: > https://bugreports.qt.io/browse/QTBUG-58799 > >> Sent: Wednesday, September 13, 2017 at 12:29 PM >> From: "coroberti ." >> To: "Jason H" >> Cc: "interest at qt-project.org" , android-development >> Subject: Re: [Android-development] Qt for Android - Signed Debug Packages by QtCreator >> >> Dear Jason, >> What is the bug report number or link to vote? >> >> Thanks! >> Kind regards, >> Robert >> >> >> On Wed, Sep 13, 2017 at 6:55 PM, Jason H wrote: >> > I filed a bug about this. Vote for it. >> > >> > However the result for me was a debug package, just incorrectly named. >> > >> > >> >> Sent: Wednesday, September 13, 2017 at 6:58 AM >> >> From: "coroberti ." >> >> To: "interest at qt-project.org" , android-development >> >> Subject: [Android-development] Qt for Android - Signed Debug Packages by QtCreator >> >> >> >> Hi, >> >> androiddeployqt with flag --sign passed by QtCreator makes >> >> release package and never debug package. >> >> >> >> Sometimes, there is a need to bug Qt Android software from QtCreator >> >> using a debug package that could work only if it is signed, i.e. when >> >> using Google's API >> >> where authentication uses signature of the app. >> >> >> >> Is where a way to workaround the current settings and make debugging >> >> from QtCreator by >> >> using singed debug app without brutal patching of androiddeployqt? >> >> >> >> Thank you in advance! >> >> >> >> Kind regards, >> >> Robert >> >> _______________________________________________ >> >> Android-development mailing list >> >> Android-development at qt-project.org >> >> http://lists.qt-project.org/mailman/listinfo/android-development >> >> >> From Sudhir.Sharma at mindtree.com Thu Sep 14 08:34:20 2017 From: Sudhir.Sharma at mindtree.com (Sudhir Sharma) Date: Thu, 14 Sep 2017 06:34:20 +0000 Subject: [Interest] Issue with toLatin1() Message-ID: Hi, I am getting following error while executing my application on QT 5.9.1 with VS2015 (win64). Any idea what could be issue? "The procedure entry point tolatin1 could not be located in qt5xml.dll" Regards, Sudhir ________________________________ http://www.mindtree.com/email/disclaimer.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From safa1996alfulaij at gmail.com Wed Sep 13 14:56:27 2017 From: safa1996alfulaij at gmail.com (Safa Alfulaij) Date: Wed, 13 Sep 2017 15:56:27 +0300 Subject: [Interest] Requesting a reviewer for updated Arabic translations Message-ID: Hello lists. I've updated the current “ancient” Arabic translations for Qt 5 (9 files till now), but I know none who can review them. Can anyone please review the updated translations [1]? Thank, Safa [1]: https://codereview.qt-project.org/#/c/205299 P.S.: I'm not subscribed, please CC me. -------------- next part -------------- An HTML attachment was scrubbed... URL: From tony at rightsoft.com.au Thu Sep 14 11:53:45 2017 From: tony at rightsoft.com.au (Tony Rietwyk) Date: Thu, 14 Sep 2017 19:53:45 +1000 Subject: [Interest] Issue with toLatin1() In-Reply-To: References: Message-ID: <1b9fcde3-cb72-97cc-f26c-fb131f99bdfd@rightsoft.com.au> Hi Sudhir, Looks suspicious - I would expect that routine to be in qt5core. Have you tried a full rebuild of your project?  Since going to VS2015, I have to do this at least once a day, as the debug information is often stale when the program runs. Regards, Tony On 14/09/2017 4:34 PM, Sudhir Sharma wrote: > > Hi, > > I am getting following error while executing my application on QT > 5.9.1 with VS2015 (win64). > > Any idea what could be issue? > > “The procedure entry point tolatin1 could not be located in qt5xml.dll” > > Regards, > > Sudhir > > > ------------------------------------------------------------------------ > > http://www.mindtree.com/email/disclaimer.html > > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From jschneiderman at gog.com Thu Sep 14 15:34:20 2017 From: jschneiderman at gog.com (John Schneiderman) Date: Thu, 14 Sep 2017 15:34:20 +0200 Subject: [Interest] Building Qt 5.9.1 Expanded Path Too Long Message-ID: <20170914133421.F30EB68264@mail.gog.com> I’m trying to build Qt 5.9.1 using Microsoft Visual Studio 2015.3. After the configuration command I run the nmake command and eventually I get the error listed below. I’ve tried even shrinking the build and source directory paths down to just: QtB and Qt, but that didn’t work either. Is there something else I can try? qwaitcondition_win.cpp NMAKE : fatal error U1095: expanded command line 'echo compiling C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qarraydata.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qbitarray.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qbytearray.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qbytearraylist.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qbytearraymatcher.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qcollator.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qcryptographichash.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qdatetime.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qdatetimeparser.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qeasingcurve.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qfreelist.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qhash.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qline.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qlinkedlist.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qlist.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qlocale.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qlocale_tools.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qpoint.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qmap.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qmargins.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qmessageauthenticationcode.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qcontiguouscache.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qrect.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qregexp.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qrefcount.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qringbuffer.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qshareddata.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qsharedpointer.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qsimd.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qsize.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qstring.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qstringbuilder.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qstringlist.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qtextboundaryfinder.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qtimeline.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qunicodetools.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qvsnprintf.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qversionnumber.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qlocale_win.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qcollator_win.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qtimezone.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qtimezoneprivate.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qtimezoneprivate_win.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qregularexpression.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qcommandlineoption.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qcommandlineparser.cpp C:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib\tools\qharfbuzz.cpp && cl -c -FIqt_pch.h -Yuqt_pch.h -Fp.pch\release\Qt5Core_pch.pch -nologo -Zc:wchar_t -FS -Zc:rvalueCast -Zc:inline -Zc:strictStrings -Zc:throwingNew -O2 -MD -utf-8 -EHsc -MP -GL -W3 -w34100 -w34189 -w44996 -w44456 -w44457 -w44458 -wd4577 -wd4467 -DUNICODE -DWIN32 -DQT_NAMESPACE=QtSdk -DQT_NO_USING_NAMESPACE -DQT_NO_FOREACH -DQT_NO_NARROWING_CONVERSIONS_IN_CONNECT -DQT_BUILD_CORE_LIB -DQT_BUILDING_QT -D_CRT_SECURE_NO_WARNINGS -D_USE_MATH_DEFINES -DQT_ASCII_CAST_WARNINGS -DQT_MOC_COMPAT -DQT_USE_QSTRINGBUILDER -DQT_DEPRECATED_WARNINGS -DQT_DISABLE_DEPRECATED_BEFORE=0x040800 -DQT_NO_DEBUG -DPCRE2_CODE_UNIT_WIDTH=16 -DPCRE2_STATIC -DNDEBUG -D_WINDLL -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\corelib -I. -Iglobal -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\harfbuzz\src -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\md5 -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\md4 -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\sha3 -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\double-conversion\include -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\double-conversion\include\double-conversion -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\include -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\include\QtCore -I..\..\include -I..\..\include\QtCore -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\include\QtCore\5.9.1 -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\include\QtCore\5.9.1\QtCore -I..\..\include\QtCore\5.9.1 -I..\..\include\QtCore\5.9.1\QtCore -Itmp -I.moc\release -ID:\Sources\gdb\32\zlib\include -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\src\3rdparty\pcre2\src -IC:\qt-everywhere-opensource-src-5.9.1\qtbase\mkspecs\win32-msvc -Fo.obj\release\ @C:\Users\JSCHNE~1\AppData\Local\Temp\nmA28C.tmp' too long Stop. NMAKE : fatal error U1077: '"C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\nmake.exe"' : return code '0x2' Stop. NMAKE : fatal error U1077: 'cd' : return code '0x2' Stop. NMAKE : fatal error U1077: 'cd' : return code '0x2' Stop. NMAKE : fatal error U1077: 'cd' : return code '0x2' Stop. Thanks, John Schneiderman Senior Software Engineer Galaxy Desktop Team GOG.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From qtgardener at gmail.com Thu Sep 14 15:34:39 2017 From: qtgardener at gmail.com (Qt Gardener) Date: Thu, 14 Sep 2017 19:04:39 +0530 Subject: [Interest] Issue with toLatin1() In-Reply-To: References: Message-ID: Hi Sudhir This looks weird to me. Does the program stop working because of this ? I suspect that you may compiling & linking against one 5.9.x. While executing it may be working Qt libraries of another version. Dheerendra Purohit *PthinkS – *“We are the Qt/QML Developers” http://www.pthinks.com On Thu, Sep 14, 2017 at 12:04 PM, Sudhir Sharma wrote: > Hi, > > > > I am getting following error while executing my application on QT 5.9.1 > with VS2015 (win64). > > Any idea what could be issue? > > > > “The procedure entry point tolatin1 could not be located in qt5xml.dll” > > > > Regards, > > Sudhir > > ------------------------------ > > http://www.mindtree.com/email/disclaimer.html > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From thiago.macieira at intel.com Thu Sep 14 16:53:43 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Thu, 14 Sep 2017 07:53:43 -0700 Subject: [Interest] Building Qt 5.9.1 Expanded Path Too Long In-Reply-To: <20170914133421.F30EB68264@mail.gog.com> References: <20170914133421.F30EB68264@mail.gog.com> Message-ID: <215075353.AJIFpH2WUL@tjmaciei-mobl1> On Thursday, 14 September 2017 06:34:20 PDT John Schneiderman wrote: > I’m trying to build Qt 5.9.1 using Microsoft Visual Studio 2015.3. After the > configuration command I run the nmake command and eventually I get the > error listed below. I’ve tried even shrinking the build and source > directory paths down to just: QtB and Qt, but that didn’t work either. Is > there something else I can try? Drop the -silent argument to configure. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From thiago.macieira at intel.com Thu Sep 14 16:52:45 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Thu, 14 Sep 2017 07:52:45 -0700 Subject: [Interest] Issue with toLatin1() In-Reply-To: References: Message-ID: <6955423.IR1d4zpJfl@tjmaciei-mobl1> On Wednesday, 13 September 2017 23:34:20 PDT Sudhir Sharma wrote: > Hi, > > I am getting following error while executing my application on QT 5.9.1 with > VS2015 (win64). Any idea what could be issue? > > "The procedure entry point tolatin1 could not be located in qt5xml.dll" First of all, C++ is case sensitive, so please be sure you did a proper copy & paste. There's no method called "tolatin1" anywhere in Qt. If you meant "toLatin1", then it exists as a member of QString, which lives in Qt5Core.dll. In any case, I'd expect to see a class name somewhere too. So there's something wrong with your error message. Paste the *correct* error message here so we can investigate. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From MRussell at neptec.com Thu Sep 14 19:39:56 2017 From: MRussell at neptec.com (Russell, Matthew) Date: Thu, 14 Sep 2017 17:39:56 +0000 Subject: [Interest] Memory Error when in background Message-ID: <8D32ABA6D779C645A4CD11339363E5FF0D45BF@EMAIL.corp.neptec.com> I've encountered the weirdness issue with my Qt 5.9.1 app. When launched, it seems to be fine until I check the memory usage with Task Manager, at which point, the memory footprint slowly increases from ~100 MiB to a few gigs and my computer shuts it down or crashes. I do not see this behaviour in Linux (Ubuntu: 14.04 or 16.06.) My app is built with 5.9.1, and implements OpenGL following the "Scene Graph - Rendering FBOs in a thread" ( http://doc.qt.io/qt-5/qtquick-scenegraph-textureinthread-example.html ) example. To date I have not been able to narrow down the cause. VDL (Visual Leak Detector in Visual Studio) reports no leaks. I've been trying to leverage the debugging and profiling tools in QtCreator but haven't had any luck so far (but that's probably due to my own inexperience with QtCreator.. I'm more accustom to CMake, ninja, and vim +YCM/rtags rather than any IDE.) In short, I can run the app for multiple days, then open Task Manager to watch the memory, and see it start to rise until the app crashes. Has anyone seen this before, or have any hints to what it may be? Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From flamaros.xavier at gmail.com Thu Sep 14 19:43:45 2017 From: flamaros.xavier at gmail.com (Xavier Bigand) Date: Thu, 14 Sep 2017 19:43:45 +0200 Subject: [Interest] Using ccache with QtCreator under Windows Message-ID: Hello, I am trying to use ccache, that I got here : https://code.google.com/archive/p/ccache-win32/ This binary seems working fine, and I just put it in the Qt\Tools\mingw530_32\bin folder near g++ and other binaries. In .pro file of our project I simply add the line : QMAKE_CXX = ccache g++ The command line is correctly generated for the compilation, but it fails: C:/sdks/Qt/Tools/mingw530_32/bin/mingw32-make -f Makefile.Debug mingw32-make[1]: Entering directory 'C:/projects/HomeDesign3D/Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' ccache g++ -c -pipe -fno-keep-inline-dllexport -isystem ../../Dependencies/boost_1_52_0 -isystem ../../Dependencies/glew/include -isystem ../../Dependencies/libpng -isystem ../../Dependencies/glsl-optimizer/src/glsl -isystem ../../Dependencies/curl-7.30.0/builds/libcurl-vc10-x86-release-dll-ipv6-sspi-spnego-winssl/include -isystem ../../Dependencies/glsl-optimizer/src/glsl -isystem ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies -isystem ../../Dependencies/glew/include -isystem ../../Dependencies/iGLU -isystem ../../Dependencies/protobuf-2.6.1/src -isystem ../../Dependencies/boost -isystem ../../Dependencies/freetype-2.5.0.1/include -isystem ../../Dependencies/freetype-gl -isystem ../../Dependencies/lzham/include -isystem ../../Dependencies/quazip-0.7.1 -isystem ../../Dependencies/assimp/include -isystem ../../Dependencies/DropboxQt/src -isystem ../../Dependencies/QtXlsxWriter/src/xlsx -isystem ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies/boost -I ../../Source -isystem ../../Source/GUI_QML/qtsingleapplication -Werror=return-type -g -DDEBUG -D_DEBUG -std=gnu++11 -frtti -Wall -Wextra -Wconversion -Wno-unused-variable -Wno-unused-parameter -Wno-reorder -fexceptions -mthreads -DUNICODE -DH3D_QMAKE -DH3D_USE_WEBVIEW -DBOOST_GEOMETRY_OVERLAY_NO_THROW -DH3D_DEVELOPER_MODE -DH3D_WORLDWIDE_VERSION -DH3D_GOLD_VERSION -DH3D_NO_STORE -DH3D_QT -DUSE_ASSIMP -DXLSX_NO_LIB -D_USE_MATH_DEFINES -DGLEW_STATIC -DENABLED_ASSERT -DENGINE3D_STATIC_LIBRARY -DARCHITECTURE_STATIC_LIBRARY -DPLATFORM_STATIC_LIBRARY -DCORE_STATIC_LIBRARY -DQUAZIP_STATIC -DQT_QML_DEBUG -DQT_QUICK_LIB -DQT_SVG_LIB -DQT_OPENGL_LIB -DQT_PRINTSUPPORT_LIB -DQT_WIDGETS_LIB -DQT_MULTIMEDIA_LIB -DQT_GUI_LIB -DQT_QML_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_CORE_LIB -DQT_NEEDS_QMAIN -I..\Qt -I. -I..\..\Source -I..\..\Source\GUI_QML\qtquick2applicationviewer -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQuick -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtSvg -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtOpenGL -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtPrintSupport -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtWidgets -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtMultimedia -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtGui -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtANGLE -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQml -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtNetwork -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1 -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1\QtCore -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtXml -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore -Idebug -I..\..\..\..\sdks\Qt\5.7\mingw53_32\mkspecs\win32-g++ -o debug\Application.o ..\..\Source\Core\Application.cpp Makefile.Debug:5501: recipe for target 'debug/Application.o' failed mingw32-make[1]: Leaving directory 'C:/projects/HomeDesign3D/Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' mingw32-make[1]: *** [debug/Application.o] Error 1 Makefile:34: recipe for target 'debug' failed mingw32-make: *** [debug] Error 2 19:26:57: The process "C:\sdks\Qt\Tools\mingw530_32\bin\mingw32-make.exe" exited with code 2. Error while building/deploying project HomeDesign3D (kit: Desktop Qt 5.7.1 MinGW 32bit) When executing step "Make" Is there a way to get the complete output of the ccache execution? Because here I just can't understand what goes wrong. Does someone do a similar thing or have an idea of what make the command line failed? -- Xavier -------------- next part -------------- An HTML attachment was scrubbed... URL: From annulen at yandex.ru Thu Sep 14 19:46:55 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Thu, 14 Sep 2017 20:46:55 +0300 Subject: [Interest] Using ccache with QtCreator under Windows In-Reply-To: References: Message-ID: <314331505411215@web27j.yandex.ru> 14.09.2017, 20:44, "Xavier Bigand" : > Hello, > > I am trying to use ccache, that I got here : https://code.google.com/archive/p/ccache-win32/ > This binary seems working fine, and I just put it in the Qt\Tools\mingw530_32\bin folder near g++ and other binaries. > > In .pro file of our project I simply add the line : QMAKE_CXX = ccache g++ No, QMAKE_CXX must refer to compiler executable only. Use symlink name "g++" and pointing to ccache as your compiler. > The command line is correctly generated for the compilation, but it fails: > > C:/sdks/Qt/Tools/mingw530_32/bin/mingw32-make -f Makefile.Debug > > mingw32-make[1]: Entering directory 'C:/projects/HomeDesign3D/Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' > > ccache g++ -c -pipe -fno-keep-inline-dllexport -isystem ../../Dependencies/boost_1_52_0 -isystem ../../Dependencies/glew/include -isystem ../../Dependencies/libpng -isystem ../../Dependencies/glsl-optimizer/src/glsl -isystem ../../Dependencies/curl-7.30.0/builds/libcurl-vc10-x86-release-dll-ipv6-sspi-spnego-winssl/include -isystem ../../Dependencies/glsl-optimizer/src/glsl -isystem ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies -isystem ../../Dependencies/glew/include -isystem ../../Dependencies/iGLU -isystem ../../Dependencies/protobuf-2.6.1/src -isystem ../../Dependencies/boost -isystem ../../Dependencies/freetype-2.5.0.1/include -isystem ../../Dependencies/freetype-gl -isystem ../../Dependencies/lzham/include -isystem ../../Dependencies/quazip-0.7.1 -isystem ../../Dependencies/assimp/include -isystem ../../Dependencies/DropboxQt/src -isystem ../../Dependencies/QtXlsxWriter/src/xlsx -isystem ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies/boost -I ../../Source -isystem ../../Source/GUI_QML/qtsingleapplication -Werror=return-type -g -DDEBUG -D_DEBUG -std=gnu++11 -frtti -Wall -Wextra -Wconversion -Wno-unused-variable -Wno-unused-parameter -Wno-reorder -fexceptions -mthreads -DUNICODE -DH3D_QMAKE -DH3D_USE_WEBVIEW -DBOOST_GEOMETRY_OVERLAY_NO_THROW -DH3D_DEVELOPER_MODE -DH3D_WORLDWIDE_VERSION -DH3D_GOLD_VERSION -DH3D_NO_STORE -DH3D_QT -DUSE_ASSIMP -DXLSX_NO_LIB -D_USE_MATH_DEFINES -DGLEW_STATIC -DENABLED_ASSERT -DENGINE3D_STATIC_LIBRARY -DARCHITECTURE_STATIC_LIBRARY -DPLATFORM_STATIC_LIBRARY -DCORE_STATIC_LIBRARY -DQUAZIP_STATIC -DQT_QML_DEBUG -DQT_QUICK_LIB -DQT_SVG_LIB -DQT_OPENGL_LIB -DQT_PRINTSUPPORT_LIB -DQT_WIDGETS_LIB -DQT_MULTIMEDIA_LIB -DQT_GUI_LIB -DQT_QML_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_CORE_LIB -DQT_NEEDS_QMAIN -I..\Qt -I. -I..\..\Source -I..\..\Source\GUI_QML\qtquick2applicationviewer -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQuick -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtSvg -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtOpenGL -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtPrintSupport -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtWidgets -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtMultimedia -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtGui -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtANGLE -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQml -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtNetwork -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1 -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1\QtCore -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtXml -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore -Idebug -I..\..\..\..\sdks\Qt\5.7\mingw53_32\mkspecs\win32-g++ -o debug\Application.o ..\..\Source\Core\Application.cpp > > Makefile.Debug:5501: recipe for target 'debug/Application.o' failed > > mingw32-make[1]: Leaving directory 'C:/projects/HomeDesign3D/Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' > > mingw32-make[1]: *** [debug/Application.o] Error 1 > > Makefile:34: recipe for target 'debug' failed > > mingw32-make: *** [debug] Error 2 > > 19:26:57: The process "C:\sdks\Qt\Tools\mingw530_32\bin\mingw32-make.exe" exited with code 2. > > Error while building/deploying project HomeDesign3D (kit: Desktop Qt 5.7.1 MinGW 32bit) > > When executing step "Make" > > Is there a way to get the complete output of the ccache execution? Because here I just can't understand what goes wrong. > > Does someone do a similar thing or have an idea of what make the command line failed? > > -- > Xavier > , > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest --  Regards, Konstantin From roland at logikalsolutions.com Thu Sep 14 23:17:22 2017 From: roland at logikalsolutions.com (Roland Hughes) Date: Thu, 14 Sep 2017 16:17:22 -0500 Subject: [Interest] Interest Digest, Vol 72, Issue 5 In-Reply-To: References: Message-ID: <0d75c34e-9076-0010-5df3-9251f4003bfd@logikalsolutions.com> Follow up, I think it is bad build instructions. ==== checking for PostgreSQL... /home/roland/piBuild/pitools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-g++ -c -pipe -march=armv7-a -marm -mthumb-interwork -mfpu=neon-vfpv4 -mtune=cortex-a7 -mabi=aapcs-linux -mfloat-abi=hard --sysroot=/home/roland/piBuild/sysroot -O2 -Wall -W -fPIC -I. -I/usr/include/postgresql -I../../../mkspecs/devices/linux-rasp-pi2-g++ -o psql.o psql.cpp /home/roland/piBuild/pitools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-g++ -Wl,-rpath-link,/home/roland/piBuild/sysroot/opt/vc/lib -Wl,-rpath-link,/home/roland/piBuild/sysroot/usr/lib/arm-linux-gnueabihf -Wl,-rpath-link,/home/roland/piBuild/sysroot/lib/arm-linux-gnueabihf -mfloat-abi=hard -Wl,-O1 -o psql psql.o --sysroot=/home/roland/piBuild/sysroot -L/usr/lib/x86_64-linux-gnu -lpq */usr/lib/x86_64-linux-gnu/libpq.so: file not recognized: File format not recognized* collect2: error: ld returned 1 exit status Makefile:92: recipe for target 'psql' failed make: *** [psql] Error 1 PostgreSQL disabled. PostgreSQL support cannot be enabled due to functionality tests! Turn on verbose messaging (-v) to /home/roland/piBuild/qt-everywhere-opensource-src-5.7.1/qtbase/configure to see the final report. If you believe this message is in error you may use the continue switch (-continue) to /home/roland/piBuild/qt-everywhere-opensource-src-5.7.1/qtbase/configure to continue. roland at roland-desktop:~/piBuild/qt-everywhere-opensource-src-5.7.1$ cd / *roland at roland-desktop:/$ sudo find -iname libpq.so****[sudo] password for roland: ****./usr/lib/x86_64-linux-gnu/libpq.so****./home/roland/piBuild/sysroot/usr/lib/arm-linux-gnueabihf/libpq.so*** ==== Following stuff from here: https://www.ics.com/blog/configuring-qt-creator-raspberry-pi which is just a tweak of this: https://wiki.qt.io/RaspberryPi2EGLFS Obviously the library path isn't going through the Raspbian sysroot tree first. -- Roland Hughes, President Logikal Solutions (630)-205-1593 http://www.theminimumyouneedtoknow.com http://www.infiniteexposure.net http://www.johnsmith-book.com http://www.logikalblog.com http://www.interestingauthors.com/blog http://lesedi.us/ http://onedollarcontentstore.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From fromqt at tungware.se Fri Sep 15 01:18:32 2017 From: fromqt at tungware.se (Henry Skoglund) Date: Fri, 15 Sep 2017 01:18:32 +0200 Subject: [Interest] Issue with toLatin1() In-Reply-To: <6955423.IR1d4zpJfl@tjmaciei-mobl1> References: <6955423.IR1d4zpJfl@tjmaciei-mobl1> Message-ID: <3550509e-d112-0ac7-abfc-e7a6feba6aff@tungware.se> On 2017-09-14 16:52, Thiago Macieira wrote: > On Wednesday, 13 September 2017 23:34:20 PDT Sudhir Sharma wrote: >> Hi, >> >> I am getting following error while executing my application on QT 5.9.1 with >> VS2015 (win64). Any idea what could be issue? >> >> "The procedure entry point tolatin1 could not be located in qt5xml.dll" > > First of all, C++ is case sensitive, so please be sure you did a proper copy & > paste. There's no method called "tolatin1" anywhere in Qt. If you meant > "toLatin1", then it exists as a member of QString, which lives in Qt5Core.dll. > In any case, I'd expect to see a class name somewhere too. > > So there's something wrong with your error message. Paste the *correct* error > message here so we can investigate. > Just a note, perhaps this is on Windows 10, in that case Microsoft has decided to "improve" that ages-old error message "The procedure entry point xxx could not be located in yyy" (it's been the same in Windows 2,3,7,NT,ME,Vista believe me I've been there). Anyway on Windows 10, the error message now points out the importing/consuming file of the entry point/function, *not* the producing/exporting file housing that function. So the message could mean that the qt5xml.dll failed to import toLatin1() from Qt5Core.dll. Perhaps you have an old (not 5.9.1) Qt5Xml.dll? Rgrds Henry From thiago.macieira at intel.com Fri Sep 15 04:15:05 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Thu, 14 Sep 2017 19:15:05 -0700 Subject: [Interest] Issue with toLatin1() In-Reply-To: <3550509e-d112-0ac7-abfc-e7a6feba6aff@tungware.se> References: <6955423.IR1d4zpJfl@tjmaciei-mobl1> <3550509e-d112-0ac7-abfc-e7a6feba6aff@tungware.se> Message-ID: <1538741.0oX0VOd4Re@tjmaciei-mobl1> On Thursday, 14 September 2017 16:18:32 PDT Henry Skoglund wrote: > Anyway on Windows 10, the error message now points out the > importing/consuming file of the entry point/function, *not* the > producing/exporting file housing that function. So the message could > mean that the qt5xml.dll failed to import toLatin1() from Qt5Core.dll. > Perhaps you have an old (not 5.9.1) Qt5Xml.dll? It is most likely a mixup of Qt5Xml.dll and Qt5Core.dll from different versions, probably from different compilers. The last change affecting that function happened in Qt 5.3.0, when I added rvalue and lvalue ref overloads. But Microsoft didn't add support for them until VS 2015. So the most likely scenario is that you're mixing 2013 and 2015 builds. I'm guessing here. If the correct (full) name of the function had been pasted, I could know for sure which of the two is the old version. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From jschneiderman at gog.com Fri Sep 15 10:29:50 2017 From: jschneiderman at gog.com (John Schneiderman) Date: Fri, 15 Sep 2017 10:29:50 +0200 Subject: [Interest] Building Qt 5.9.1 Expanded Path Too Long In-Reply-To: <215075353.AJIFpH2WUL@tjmaciei-mobl1> References: <20170914133421.F30EB68264@mail.gog.com> <215075353.AJIFpH2WUL@tjmaciei-mobl1> Message-ID: <20170915082949.EC6DF68263@mail.gog.com> >> Drop the -silent argument to configure. Thanks, that fixed the issue. -------------- next part -------------- An HTML attachment was scrubbed... URL: From flamaros.xavier at gmail.com Fri Sep 15 11:07:56 2017 From: flamaros.xavier at gmail.com (Xavier Bigand) Date: Fri, 15 Sep 2017 11:07:56 +0200 Subject: [Interest] Using ccache with QtCreator under Windows In-Reply-To: <314331505411215@web27j.yandex.ru> References: <314331505411215@web27j.yandex.ru> Message-ID: I tried with a symlink as you suggested, but it still doesn't work. I also putted the path of mingw binaries in my PATH environment variable. 2017-09-14 19:46 GMT+02:00 Konstantin Tokarev : > > > 14.09.2017, 20:44, "Xavier Bigand" : > > Hello, > > > > I am trying to use ccache, that I got here : https://code.google.com/ > archive/p/ccache-win32/ > > This binary seems working fine, and I just put it in > the Qt\Tools\mingw530_32\bin folder near g++ and other binaries. > > > > In .pro file of our project I simply add the line : QMAKE_CXX = ccache > g++ > > No, QMAKE_CXX must refer to compiler executable only. > Use symlink name "g++" and pointing to ccache as your compiler. > > > The command line is correctly generated for the compilation, but it > fails: > > > > C:/sdks/Qt/Tools/mingw530_32/bin/mingw32-make -f Makefile.Debug > > > > mingw32-make[1]: Entering directory 'C:/projects/HomeDesign3D/ > Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' > > > > ccache g++ -c -pipe -fno-keep-inline-dllexport -isystem > ../../Dependencies/boost_1_52_0 -isystem ../../Dependencies/glew/include > -isystem ../../Dependencies/libpng -isystem ../../Dependencies/glsl-optimizer/src/glsl > -isystem ../../Dependencies/curl-7.30.0/builds/libcurl-vc10-x86- > release-dll-ipv6-sspi-spnego-winssl/include -isystem > ../../Dependencies/glsl-optimizer/src/glsl -isystem > ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies -isystem > ../../Dependencies/glew/include -isystem ../../Dependencies/iGLU -isystem > ../../Dependencies/protobuf-2.6.1/src -isystem ../../Dependencies/boost > -isystem ../../Dependencies/freetype-2.5.0.1/include -isystem > ../../Dependencies/freetype-gl -isystem ../../Dependencies/lzham/include > -isystem ../../Dependencies/quazip-0.7.1 -isystem > ../../Dependencies/assimp/include -isystem ../../Dependencies/DropboxQt/src > -isystem ../../Dependencies/QtXlsxWriter/src/xlsx -isystem > ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies/boost -I > ../../Source -isystem ../../Source/GUI_QML/qtsingleapplication > -Werror=return-type -g -DDEBUG -D_DEBUG -std=gnu++11 -frtti -Wall -Wextra > -Wconversion -Wno-unused-variable -Wno-unused-parameter -Wno-reorder > -fexceptions -mthreads -DUNICODE -DH3D_QMAKE -DH3D_USE_WEBVIEW > -DBOOST_GEOMETRY_OVERLAY_NO_THROW -DH3D_DEVELOPER_MODE > -DH3D_WORLDWIDE_VERSION -DH3D_GOLD_VERSION -DH3D_NO_STORE -DH3D_QT > -DUSE_ASSIMP -DXLSX_NO_LIB -D_USE_MATH_DEFINES -DGLEW_STATIC > -DENABLED_ASSERT -DENGINE3D_STATIC_LIBRARY -DARCHITECTURE_STATIC_LIBRARY > -DPLATFORM_STATIC_LIBRARY -DCORE_STATIC_LIBRARY -DQUAZIP_STATIC > -DQT_QML_DEBUG -DQT_QUICK_LIB -DQT_SVG_LIB -DQT_OPENGL_LIB > -DQT_PRINTSUPPORT_LIB -DQT_WIDGETS_LIB -DQT_MULTIMEDIA_LIB -DQT_GUI_LIB > -DQT_QML_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_CORE_LIB -DQT_NEEDS_QMAIN > -I..\Qt -I. -I..\..\Source -I..\..\Source\GUI_QML\qtquick2applicationviewer > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQuick > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtSvg > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtOpenGL > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtPrintSupport > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtWidgets > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtMultimedia > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtGui > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtANGLE > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQml > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtNetwork > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1 > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1\QtCore > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtXml > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore -Idebug > -I..\..\..\..\sdks\Qt\5.7\mingw53_32\mkspecs\win32-g++ -o > debug\Application.o ..\..\Source\Core\Application.cpp > > > > Makefile.Debug:5501: recipe for target 'debug/Application.o' failed > > > > mingw32-make[1]: Leaving directory 'C:/projects/HomeDesign3D/ > Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' > > > > mingw32-make[1]: *** [debug/Application.o] Error 1 > > > > Makefile:34: recipe for target 'debug' failed > > > > mingw32-make: *** [debug] Error 2 > > > > 19:26:57: The process "C:\sdks\Qt\Tools\mingw530_32\bin\mingw32-make.exe" > exited with code 2. > > > > Error while building/deploying project HomeDesign3D (kit: Desktop Qt > 5.7.1 MinGW 32bit) > > > > When executing step "Make" > > > > Is there a way to get the complete output of the ccache execution? > Because here I just can't understand what goes wrong. > > > > Does someone do a similar thing or have an idea of what make the command > line failed? > > > > -- > > Xavier > > , > > > > _______________________________________________ > > Interest mailing list > > Interest at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/interest > > > -- > Regards, > Konstantin -- Xavier -------------- next part -------------- An HTML attachment was scrubbed... URL: From annulen at yandex.ru Fri Sep 15 11:45:36 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Fri, 15 Sep 2017 12:45:36 +0300 Subject: [Interest] Using ccache with QtCreator under Windows In-Reply-To: References: <314331505411215@web27j.yandex.ru> Message-ID: <211051505468736@web4o.yandex.ru> 15.09.2017, 12:07, "Xavier Bigand" : > I tried with a symlink as you suggested, but it still doesn't work. > I also putted the path of mingw binaries in my PATH environment variable. Ah, that's on Windows. Then I'm not sure it will work, because Windows symlinks may behave differently > > 2017-09-14 19:46 GMT+02:00 Konstantin Tokarev : > >> 14.09.2017, 20:44, "Xavier Bigand" : >>> Hello, >>> >>> I am trying to use ccache, that I got here : https://code.google.com/archive/p/ccache-win32/ >>> This binary seems working fine, and I just put it in the Qt\Tools\mingw530_32\bin folder near g++ and other binaries. >>> >>> In .pro file of our project I simply add the line : QMAKE_CXX = ccache g++ >> >> No, QMAKE_CXX must refer to compiler executable only. >> Use symlink name "g++" and pointing to ccache as your compiler. >> >>> The command line is correctly generated for the compilation, but it fails: >>> >>> C:/sdks/Qt/Tools/mingw530_32/bin/mingw32-make -f Makefile.Debug >>> >>> mingw32-make[1]: Entering directory 'C:/projects/HomeDesign3D/Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' >>> >>> ccache g++ -c -pipe -fno-keep-inline-dllexport -isystem ../../Dependencies/boost_1_52_0 -isystem ../../Dependencies/glew/include -isystem ../../Dependencies/libpng -isystem ../../Dependencies/glsl-optimizer/src/glsl -isystem ../../Dependencies/curl-7.30.0/builds/libcurl-vc10-x86-release-dll-ipv6-sspi-spnego-winssl/include -isystem ../../Dependencies/glsl-optimizer/src/glsl -isystem ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies -isystem ../../Dependencies/glew/include -isystem ../../Dependencies/iGLU -isystem ../../Dependencies/protobuf-2.6.1/src -isystem ../../Dependencies/boost -isystem ../../Dependencies/freetype-2.5.0.1/include -isystem ../../Dependencies/freetype-gl -isystem ../../Dependencies/lzham/include -isystem ../../Dependencies/quazip-0.7.1 -isystem ../../Dependencies/assimp/include -isystem ../../Dependencies/DropboxQt/src -isystem ../../Dependencies/QtXlsxWriter/src/xlsx -isystem ../../Dependencies/zlib-1.2.7 -isystem ../../Dependencies/boost -I ../../Source -isystem ../../Source/GUI_QML/qtsingleapplication -Werror=return-type -g -DDEBUG -D_DEBUG -std=gnu++11 -frtti -Wall -Wextra -Wconversion -Wno-unused-variable -Wno-unused-parameter -Wno-reorder -fexceptions -mthreads -DUNICODE -DH3D_QMAKE -DH3D_USE_WEBVIEW -DBOOST_GEOMETRY_OVERLAY_NO_THROW -DH3D_DEVELOPER_MODE -DH3D_WORLDWIDE_VERSION -DH3D_GOLD_VERSION -DH3D_NO_STORE -DH3D_QT -DUSE_ASSIMP -DXLSX_NO_LIB -D_USE_MATH_DEFINES -DGLEW_STATIC -DENABLED_ASSERT -DENGINE3D_STATIC_LIBRARY -DARCHITECTURE_STATIC_LIBRARY -DPLATFORM_STATIC_LIBRARY -DCORE_STATIC_LIBRARY -DQUAZIP_STATIC -DQT_QML_DEBUG -DQT_QUICK_LIB -DQT_SVG_LIB -DQT_OPENGL_LIB -DQT_PRINTSUPPORT_LIB -DQT_WIDGETS_LIB -DQT_MULTIMEDIA_LIB -DQT_GUI_LIB -DQT_QML_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_CORE_LIB -DQT_NEEDS_QMAIN -I..\Qt -I. -I..\..\Source -I..\..\Source\GUI_QML\qtquick2applicationviewer -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQuick -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtSvg -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtOpenGL -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtPrintSupport -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtWidgets -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtMultimedia -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtGui -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtANGLE -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtQml -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtNetwork -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1 -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore\5.7.1\QtCore -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtXml -I..\..\..\..\sdks\Qt\5.7\mingw53_32\include\QtCore -Idebug -I..\..\..\..\sdks\Qt\5.7\mingw53_32\mkspecs\win32-g++ -o debug\Application.o ..\..\Source\Core\Application.cpp >>> >>> Makefile.Debug:5501: recipe for target 'debug/Application.o' failed >>> >>> mingw32-make[1]: Leaving directory 'C:/projects/HomeDesign3D/Projects/build-HomeDesign3D-Desktop_Qt_5_7_1_MinGW_32bit2-Debug' >>> >>> mingw32-make[1]: *** [debug/Application.o] Error 1 >>> >>> Makefile:34: recipe for target 'debug' failed >>> >>> mingw32-make: *** [debug] Error 2 >>> >>> 19:26:57: The process "C:\sdks\Qt\Tools\mingw530_32\bin\mingw32-make.exe" exited with code 2. >>> >>> Error while building/deploying project HomeDesign3D (kit: Desktop Qt 5.7.1 MinGW 32bit) >>> >>> When executing step "Make" >>> >>> Is there a way to get the complete output of the ccache execution? Because here I just can't understand what goes wrong. >>> >>> Does someone do a similar thing or have an idea of what make the command line failed? >>> >>> -- >>> Xavier >>> , >>> >>> _______________________________________________ >>> Interest mailing list >>> Interest at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/interest >> >> -- >> Regards, >> Konstantin > > -- > Xavier --  Regards, Konstantin From daniel at zonque.org Fri Sep 15 15:56:25 2017 From: daniel at zonque.org (Daniel Mack) Date: Fri, 15 Sep 2017 15:56:25 +0200 Subject: [Interest] Camera support for QtWebEngine In-Reply-To: <0a1ab5fa-5e31-ac93-f6da-76a282fb4f91@zonque.org> References: <0a1ab5fa-5e31-ac93-f6da-76a282fb4f91@zonque.org> Message-ID: Hi, In the meantime, I've updated to Qt 5.9.1, but that alone didn't fix the issue yet. To recap, the system I'm working with is a Qualcomm APQ8016 (Snapdragon 410), and I'm running an image built with Yocto. The goal is to have QtWebEngine expose the camera in the system so that they become available though regular HTML5 media objects. The camera works with the QtMultimedia example code, so I hoped that QtWebEngine would just work. I dug a little deeper now, and it turns out that the code that currently handles the cameras is located in src/3rdparty/chromium/media/capture/video/linux/v4l2_capture_delegate.cc, and the code there chokes on the fact that the cameras are exposed with V4L2_CAP_VIDEO_CAPTURE_MPLANE instead of V4L2_CAP_VIDEO_CAPTURE capabilities, and V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE as buffer type. There's also code in src/3rdparty/chromium/media/gpu/ which seems to be prepared for type of devices, but according to the build system, it can only be built for chromeos, not plain Linux. I was under the impression that QtWebKit would just piggy-back on QtMultiMedia for such things. Is that how things are supposed to work? If so, what build-time configuration must be made to make this happen? Thanks, Daniel On 05/24/2017 07:53 PM, Daniel Mack wrote: > Hi, > > I'm running QtWebKit 5.8 on Linux on aarch64 - the binaries are built > with Yocto/OpenEmbedded, using the meta-qt5 layer. > > Out-of-the-box, access to V4L2 camera devices isn't supported, and I > wonder where to start looking. The same application works fine on > x86-64, so it's really a platform-specific issue. > > Are there any configuration parameters I need to be aware of when > cross-compiling the sources? Does QtWebkit use QtMultimedia for such > purposes by default? Because running a simple QCamera based application > reports "Camera error "The camera service is missing" via the > QCamera::Error signal. I'm not very familiar with the internals of Qt, > so I can't make sense of that. > > > > Thanks for any pointer, > Daniel > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > From xbenlau at gmail.com Sun Sep 17 03:08:29 2017 From: xbenlau at gmail.com (Ben Lau) Date: Sun, 17 Sep 2017 09:08:29 +0800 Subject: [Interest] Milo Code Database - kickstart your projects easily In-Reply-To: References: Message-ID: Hi Tomasz, It is cool. It is very surprising to find that it has a template for a barcode scanner. Do you have the plan to make it works with Qt Creator directly? Such that it could create a project by File -> New File or Project -> Milo Code Database -> xxxxx On 8 September 2017 at 15:14, Tomasz Siekierda wrote: > Hi Qt! > > We've just released Milo Code Database: a set of small, useful helper > classes > for Qt, easy to include and use - plus a scaffolding for new projects we > call > the New Project Template. At Milo, we use them in many of our projects > to speed up development, and we figured they may be useful to the community > as well. > > MCDB comes in 2 forms: > > 1. Installer - which helps to set up a new project in just a few seconds > (with > scaffolding for Qt app, documentation, CI integration, versioning, config, > logger, and more). It can even set up a git repo and do the initial commit > for > you :-) > 2. Set of independent git repos, easy to clone / download / use as > submodule. > > So, what are the goodies? > > * New project template - template for all new Qt projects (QML and > Widgets). > Contains a scaffolding for Qt app with documentation, tests, resources etc. > * MScripts - scripts for integration with GitLab CI, Mattermost, versioning > scripts for all platforms (including git SHA), and more > * MConfig - a small class which helps storing user app settings and > configuration > * MLog - small but powerful log class, with full categorized logging > support > * MRestAPI - handy set of base classes for talking with REST web servers > * MBarcodeScanner - small helper class that allows to scan barcodes and QR > codes > using QZXing library > * MCharts - implements some simple yet meaningful and eye friendly charts > to > display data > * MSentry - remote logger support (Sentry/ Raven). Easy to turn on and > off, just > with a config switch > * MCrypto - AES module, containing simple and convenient API, based on > OpenSSL > and it has a fallback if OSSL is not available > > License? > > MIT. We are actually open to other licenses, too, if somebody thinks that > MIT > is not proper in this case. > > Where can I get it? > > MCDB installers (Linux and Windows): > https://seafile.milosolutions.com/d/2c50614e1e/ > MCDB docs are available online: > https://docs.milosolutions.com/milo-code-db/main/ > And all submodules are listed and documented here: > https://docs.milosolutions.com/milo-code-db/main/subprojects.html > Each module's source code is on GitHub: https://github.com/milosolutions > > Contributing > > We're open to any comments, critique, pull requests etc. We can also help > if > any piece of the documentation is unclear. It's all being constantly > updated and > we're happy to get some external input. > > Feel free to reach me at sierdzio at gmail.com or more officially at > ppm-qt at milosolutions.com > > Cheerio, > sierdzio > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From chgans at gmail.com Mon Sep 18 10:01:14 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Mon, 18 Sep 2017 20:01:14 +1200 Subject: [Interest] [Development] QRandomGenerator and boot times In-Reply-To: References: <1700113.SrpWEvuCgM@tjmaciei-mobl1> <106099e9-84be-6a11-5bb8-86758b411b80@qt.io> <19569896.oT66G5BakO@tjmaciei-mobl1> Message-ID: [adding back to the mailing list] On 18/09/2017 7:59 pm, "Christian Gagneraud" wrote: On 18/09/2017 7:34 pm, "Sami Nurmenniemi" wrote: On 15.09.2017 18:21, Thiago Macieira wrote: > On Friday, 15 September 2017 00:31:36 PDT Sami Nurmenniemi wrote: > >> I think we'll just have to accept blocking for the devices without >> hwrng. I don't know if we really support any such devices. If we do and >> boot time is essential for those, we'll have to figure out some way >> (probably saving entropy over reboot). >> > And that would be a non-Qt job. I wasn't worried about Qt on real devices > because I don't expect it to be run early enough to matter. On VMs, that's > another story, and if you don't trust your undercloud-provided /dev/hwrng, > why > are you using that cloud? (Though I'll say there's an Intel team working on > figuring out how a VM can get trust from the actual hardware, skipping the > hypervisor trust) > We have made some safety critical customer demos where fast boot time of Qt framework is essential. It was demoing boot time of 1.2s for a Qt application to start after powering on the device. I suppose that demo (and real safety critical use cases) is going to have problems with the QRandomGenerator even with hwrng in use. So basically, you have artificially "optimised" boot time by ignoring entropy management and now you're blaming QRandomGenerator. You can do one or the other, not both. Me too I can make a device boot in less than a second, but then don't be surprised if some features are missing/buggy.... This sound very much like a beginner mistake, where boot time takes over functionality. Every one can boot an unfunctional device in less than a second. This is nothing new. Respectfully yours, Chris > Anyway, I've read the entire Python thread to figure out what their > conclusion > was. Turns out, after spending hours reading everything, the bug ends > without > a conclusion. It must have been concluded, but it's not recorded in the bug > report! > > _______________________________________________ Development mailing list Development at qt-project.org http://lists.qt-project.org/mailman/listinfo/development -------------- next part -------------- An HTML attachment was scrubbed... URL: From sierdzio at gmail.com Mon Sep 18 14:34:40 2017 From: sierdzio at gmail.com (Tomasz Siekierda) Date: Mon, 18 Sep 2017 14:34:40 +0200 Subject: [Interest] Milo Code Database - kickstart your projects easily In-Reply-To: References: Message-ID: On 17 September 2017 at 03:08, Ben Lau wrote: > Hi Tomasz, > > It is cool. It is very surprising to find that it has a template for a > barcode scanner. > Thanks. > Do you have the plan to make it works with Qt Creator directly? Such that it > could create a project by File -> New File or Project -> Milo Code Database > -> xxxxx Yes, that sounds great. No concrete plan yet, but I will put it on my TODO list, thanks! > > On 8 September 2017 at 15:14, Tomasz Siekierda wrote: >> >> Hi Qt! >> >> We've just released Milo Code Database: a set of small, useful helper >> classes >> for Qt, easy to include and use - plus a scaffolding for new projects we >> call >> the New Project Template. At Milo, we use them in many of our projects >> to speed up development, and we figured they may be useful to the >> community >> as well. >> >> MCDB comes in 2 forms: >> >> 1. Installer - which helps to set up a new project in just a few seconds >> (with >> scaffolding for Qt app, documentation, CI integration, versioning, config, >> logger, and more). It can even set up a git repo and do the initial commit >> for >> you :-) >> 2. Set of independent git repos, easy to clone / download / use as >> submodule. >> >> So, what are the goodies? >> >> * New project template - template for all new Qt projects (QML and >> Widgets). >> Contains a scaffolding for Qt app with documentation, tests, resources >> etc. >> * MScripts - scripts for integration with GitLab CI, Mattermost, >> versioning >> scripts for all platforms (including git SHA), and more >> * MConfig - a small class which helps storing user app settings and >> configuration >> * MLog - small but powerful log class, with full categorized logging >> support >> * MRestAPI - handy set of base classes for talking with REST web servers >> * MBarcodeScanner - small helper class that allows to scan barcodes and QR >> codes >> using QZXing library >> * MCharts - implements some simple yet meaningful and eye friendly charts >> to >> display data >> * MSentry - remote logger support (Sentry/ Raven). Easy to turn on and >> off, just >> with a config switch >> * MCrypto - AES module, containing simple and convenient API, based on >> OpenSSL >> and it has a fallback if OSSL is not available >> >> License? >> >> MIT. We are actually open to other licenses, too, if somebody thinks that >> MIT >> is not proper in this case. >> >> Where can I get it? >> >> MCDB installers (Linux and Windows): >> https://seafile.milosolutions.com/d/2c50614e1e/ >> MCDB docs are available online: >> https://docs.milosolutions.com/milo-code-db/main/ >> And all submodules are listed and documented here: >> https://docs.milosolutions.com/milo-code-db/main/subprojects.html >> Each module's source code is on GitHub: https://github.com/milosolutions >> >> Contributing >> >> We're open to any comments, critique, pull requests etc. We can also help >> if >> any piece of the documentation is unclear. It's all being constantly >> updated and >> we're happy to get some external input. >> >> Feel free to reach me at sierdzio at gmail.com or more officially at >> ppm-qt at milosolutions.com >> >> Cheerio, >> sierdzio >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > > From giuseppe.dangelo at kdab.com Mon Sep 18 22:20:17 2017 From: giuseppe.dangelo at kdab.com (Giuseppe D'Angelo) Date: Mon, 18 Sep 2017 22:20:17 +0200 Subject: [Interest] Qt at CppCon 2017 Message-ID: <17620e9c-7349-a763-b3a8-fd3c92dd4f2e@kdab.com> Howdy, at this year's CppCon [1] there is going to be a bigger Qt presence: Lars Knoll will deliver a keynote [2]; I am going to host a pre-conference 2-days training, as well as a regular session; and Simon Hausmann is also joining with another talk [3]. There is also going to be a happy hour sponsored by The Qt Company [4]. Like last year, if you're at the conference, feel free to say hello! [1] https://cppcon.org/ [2] https://cppcon.org/2017-plenary-announced-knoll/ [3] https://www.kdab.com/qt-talks-cppcon-2017/ [4] https://www.qt.io/event/cppcon/ Cheers, -- Giuseppe D'Angelo | giuseppe.dangelo at kdab.com | Senior Software Engineer KDAB (UK) Ltd., a KDAB Group company | Tel: UK +44-1625-809908 KDAB - Qt, C++ and OpenGL Experts -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4007 bytes Desc: Firma crittografica S/MIME URL: From rbehm at hushmail.com Tue Sep 19 04:27:48 2017 From: rbehm at hushmail.com (Reinhardt Behm) Date: Tue, 19 Sep 2017 10:27:48 +0800 Subject: [Interest] QWebSocket/Server Message-ID: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> Hi all, I am using QWebSocket and QWebSocketServer (Qt5.5) Is it possible to have another Server (as HTTP server) listening at the same port? QWebSocket supports binaryFrameReceived(). Is there any support to send Frames? QWebSocket has a signal bytesWritten(). Any support for receive progress? -- Best Regards Reinhardt Behm From rainer_wiesenfarth at trimble.com Tue Sep 19 08:44:13 2017 From: rainer_wiesenfarth at trimble.com (Rainer Wiesenfarth) Date: Tue, 19 Sep 2017 08:44:13 +0200 Subject: [Interest] QWebSocket/Server In-Reply-To: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> References: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> Message-ID: On Tue, Sep 19, 2017 at 4:27 AM, Reinhardt Behm wrote: > I am using QWebSocket and QWebSocketServer (Qt5.5) > Is it possible to have another Server (as HTTP server) listening at the > same > port? > ​No, each port can be listened to only by one server. This is by design of the underlying protocol (TCP, but also UDP). ​ > QWebSocket supports binaryFrameReceived(). Is there any support to send > Frames? > ​sendBinaryMessage()?​ QWebSocket has a signal bytesWritten(). Any support for receive progress? > ​No, this is also not supported by the underlying protocol. You will receive ​an error() signal if something went wrong. If you need acknowledge messages, you have to add these to your own protocol. ​Cheers, Rainer​ -- Software Engineer | Trimble Imaging Division Rotebühlstraße 81 | 70178 Stuttgart | Germany Office +49 711 22881 0 | Fax +49 711 22881 11 http://www.trimble.com/imaging/ | http://www.inpho.de/ Trimble Germany GmbH, Am Prime Parc 11, 65479 Raunheim Eingetragen beim Amtsgericht Darmstadt unter HRB 83893, Geschäftsführer: Dr. Frank Heimberg, Jürgen Kesper -------------- next part -------------- An HTML attachment was scrubbed... URL: From thiago.macieira at intel.com Tue Sep 19 08:51:15 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Mon, 18 Sep 2017 23:51:15 -0700 Subject: [Interest] QWebSocket/Server In-Reply-To: References: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> Message-ID: <1604819.zAPVLyGY8D@tjmaciei-mobl1> On Monday, 18 September 2017 23:44:13 PDT Rainer Wiesenfarth wrote: > ​No, each port can be listened to only by one server. This is by design of > the underlying protocol (TCP, but also UDP). Unless you can find a way to multiplex different protocols at an intermediary layer. The TLS header has an extension called Next Protocol Negotiation (NPN), so it's possible for the client to declare intent and the server react by serving different protocols. This is will be used for HTTP/2, but it's possible to use for other protocols too. Note how this requires client support. This doesn't mean the QWebServer class is able to react to the NPN. But it's possible to do so in theory. An alternative is to detect the protocol by fingerprinting what it actually sends on the wire. For example, if it sent HELO instead of GET, you could conclude it's an SMTP client instead of HTTP. This does not require client support, but does require that the two protocols be different enough that it's not possible to mistake one for the other. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From rbehm at hushmail.com Tue Sep 19 09:20:51 2017 From: rbehm at hushmail.com (Reinhardt Behm) Date: Tue, 19 Sep 2017 15:20:51 +0800 Subject: [Interest] QWebSocket/Server In-Reply-To: References: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> Message-ID: <90e83b2a18ee8f210ba9bceb4342b19f@smtp.hushmail.com> On Tuesday 19 September 2017 08:44:13 Rainer Wiesenfarth wrote: > On Tue, Sep 19, 2017 at 4:27 AM, Reinhardt Behm wrote: > > I am using QWebSocket and QWebSocketServer (Qt5.5) > > Is it possible to have another Server (as HTTP server) listening at the > > same > > port? > > ​No, each port can be listened to only by one server. This is by design of > the underlying protocol (TCP, but also UDP). That was also my understanding. I hoped there is some trick as the WebSocket ​ protocol sits on top of HTTP. > > QWebSocket supports binaryFrameReceived(). Is there any support to send > > Frames? > > ​sendBinaryMessage()?​ That is just a complete message, signaled by binaryMessageReceived(). The WebSocket ​protocol can split messages into frames. But it seems there is currently no support to send frames, only the reception is supported. My idea was to split large messages and also have the receiver being able to report progress and not have a large message fully in memory. > > QWebSocket has a signal bytesWritten(). Any support for receive progress? > > > ​No, this is also not supported by the underlying protocol. You will > receive ​an error() signal if something went wrong. > > If you need acknowledge messages, you have to add these to your own > protocol. Ok, I can make my own protocol on top of WebSocket, send chunks and replies from the receiver. > > ​Cheers, Rainer ​-- Best Regards Reinhardt Behm From annulen at yandex.ru Tue Sep 19 11:30:19 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Tue, 19 Sep 2017 12:30:19 +0300 Subject: [Interest] QWebSocket/Server In-Reply-To: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> References: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> Message-ID: <103261505813419@web45o.yandex.ru> 19.09.2017, 05:28, "Reinhardt Behm" : > Hi all, > > I am using QWebSocket and QWebSocketServer (Qt5.5) > Is it possible to have another Server (as HTTP server) listening at the same > port? For example, you can start Nginx on that port and set up proxying for WebSocket protocol: http://nginx.org/en/docs/http/websocket.html > > QWebSocket supports binaryFrameReceived(). Is there any support to send > Frames? > > QWebSocket has a signal bytesWritten(). Any support for receive progress? > > -- > Best Regards > > Reinhardt Behm > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Regards, Konstantin From rbehm at hushmail.com Tue Sep 19 11:49:05 2017 From: rbehm at hushmail.com (Reinhardt Behm) Date: Tue, 19 Sep 2017 17:49:05 +0800 Subject: [Interest] QWebSocket/Server In-Reply-To: <103261505813419@web45o.yandex.ru> References: <6eeab427ce0c4133e69dd8e99d4eb81a@smtp.hushmail.com> <103261505813419@web45o.yandex.ru> Message-ID: <95793cd5ea9e222327950f1e428a51f1@smtp.hushmail.com> On Tuesday 19 September 2017 12:30:19 Konstantin Tokarev wrote: > 19.09.2017, 05:28, "Reinhardt Behm" : > > Hi all, > > > > I am using QWebSocket and QWebSocketServer (Qt5.5) > > Is it possible to have another Server (as HTTP server) listening at the > > same port? > > For example, you can start Nginx on that port and set up proxying for > WebSocket protocol: > > http://nginx.org/en/docs/http/websocket.html > > > QWebSocket supports binaryFrameReceived(). Is there any support to send > > Frames? > > > > QWebSocket has a signal bytesWritten(). Any support for receive progress? > > > > -- > > Best Regards > > > > Reinhardt Behm > > > > _______________________________________________ > > Interest mailing list > > Interest at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/interest Interesting idea. Thanks Konstantin, I will have a look into that. Of course I also could use another port, but that would require the customer to open that in their firewall. -- Best Regards Reinhardt Behm From rjvbertin at gmail.com Tue Sep 19 12:04:20 2017 From: rjvbertin at gmail.com (=?ISO-8859-1?Q?Ren=E9_J=2EV=2E?= Bertin) Date: Tue, 19 Sep 2017 12:04:20 +0200 Subject: [Interest] QFileSystemWatcher performance (Mac vs. Linux) Message-ID: <1929064.CSFymMOeDV@bola> Hi, I've been running into importing slowness in an IDE that I've traced to the creation of a dirwatcher service that monitors the entire source tree for file changes. Creating that service using KF5's KDirWatch has already taken up to about 15 minutes (for the GCC source tree). KDirWatch *can* use QFileSystemWatcher, and comes with an autotest (kdirwatch_qfswatch_unittest). Comparing timing of that test on a Mac with a 2.7Ghz i7 (and using HFS+) vs. a Linux system with a 1.6Ghz N3150 Celeron (and using ZFS...) shows the Linux system to be roughly 25% faster (20s vs. 15s). I'm looking for a basic test utility that I can use to benchmark performance (preferable something "done right" to achieve optimum efficiency = not thrown together by me) in order to compare QFileSystemWatcher performance directly, without the KDirWatch overhead. Qt doesn't include an example just for this class, but maybe one exists elsewhere? Cross-posting this to the Creator ML because the underlying use case is right up their alley. Thanks, René From rjvbertin at gmail.com Tue Sep 19 16:06:04 2017 From: rjvbertin at gmail.com (=?UTF-8?B?UmVuw6kgSi4gVi4=?= Bertin) Date: Tue, 19 Sep 2017 16:06:04 +0200 Subject: [Interest] QFileSystemWatcher performance (Mac vs. Linux) References: <1929064.CSFymMOeDV@bola> Message-ID: <14233724.GS5AhK4rpL@patux.local> This NSLog message appears to be related, but how and to what exactly is still the question... (CarbonCore.framework) implementation_callback_rpc: ERROR: CFDictionaryGetValue(FSEvents_streamDict, cfPortNumber) => NULL R. From evilruff at gmail.com Tue Sep 19 20:26:06 2017 From: evilruff at gmail.com (Yuri Alexandrov) Date: Tue, 19 Sep 2017 21:26:06 +0300 Subject: [Interest] QML Menu disappeared on latest Windows 10 updates Message-ID: <59c16146.0b042e0a.d086a.3253@mx.google.com> Hi, I just noticed very weird thing started to happed on one of my laptop. After last Windows 10 update all QML applications (5.7, 5.9, 5.10) got menu control gone. That also affected QtCreator – QML Designer page. As soon as you return back to any ‘widget-based’ page, menu is back again. Screenshots attached. I also opened support ticket (as I have Qt Commercial license), but just wondering if anybody saw similar things. Only thing which still has menu working properly is ‘texteditor’ example using QtQuickControls2 in Qt 5.10 Regards, Yuri -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: pasted_image_at_2017_09_18_01_23_pm.png Type: image/png Size: 156278 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: pasted_image_at_2017_09_15_06_34_pm.png Type: image/png Size: 221846 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: pasted_image_at_2017_09_18_01_21_pm.png Type: image/png Size: 9216 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: pasted_image_at_2017_09_18_01_22_pm_1024.png Type: image/png Size: 8028 bytes Desc: not available URL: From alvinhochun at gmail.com Tue Sep 19 20:50:45 2017 From: alvinhochun at gmail.com (Alvin Wong) Date: Wed, 20 Sep 2017 02:50:45 +0800 Subject: [Interest] QML Menu disappeared on latest Windows 10 updates Message-ID: Ha! You can thank Intel for that. Intel graphics driver builds between 4636 and 4729 (exclusive) are to blame. The workaround is probably either to use ANGLE instead of desktop OpenGL or ask users to update their Intel graphics driver. See: https://communities.intel.com/thread/116003 > From: Yuri Alexandrov > To: "interest at qt-project.org" > Cc: > Bcc: > Date: Tue, 19 Sep 2017 21:26:06 +0300 > Subject: [Interest] QML Menu disappeared on latest Windows 10 updates > > > > Hi, > > > > I just noticed very weird thing started to happed on one of my laptop. After last Windows 10 update all QML applications (5.7, 5.9, 5.10) got menu control gone. That also affected QtCreator – QML Designer page. As soon as you return back to any ‘widget-based’ page, menu is back again. > > > > Screenshots attached. I also opened support ticket (as I have Qt Commercial license), but just wondering if anybody saw similar things. > > > > Only thing which still has menu working properly is ‘texteditor’ example using QtQuickControls2 in Qt 5.10 > > > > Regards, > > Yuri From evilruff at gmail.com Tue Sep 19 21:19:56 2017 From: evilruff at gmail.com (Yuri Alexandrov) Date: Tue, 19 Sep 2017 22:19:56 +0300 Subject: [Interest] QML Menu disappeared on latest Windows 10 updates In-Reply-To: References: Message-ID: <59c16ddb.0c542e0a.4882a.01b2@mx.google.com> Hi You made my day, Alvin! Thank you so much! Regards, Yuri -------------- next part -------------- An HTML attachment was scrubbed... URL: From patrickkidd at gmail.com Tue Sep 19 22:47:37 2017 From: patrickkidd at gmail.com (Patrick Stinson) Date: Tue, 19 Sep 2017 13:47:37 -0700 Subject: [Interest] Pan-scroll gesture for QGraphicsView on iOS? Message-ID: The pan gesture scrolls qgraphicsview just fine on macOS, but the same code doesn't work on iOS. Is there something special I need to do there? Thank you! -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 1403 bytes Desc: not available URL: From nus1998 at yeah.net Wed Sep 20 10:59:27 2017 From: nus1998 at yeah.net (nus1998) Date: Wed, 20 Sep 2017 16:59:27 +0800 (CST) Subject: [Interest] Qt cross compile issue on pkg-config Message-ID: <72e73ab5.253d.15e9e836db8.Coremail.nus1998@yeah.net> Hi All, I exported PKG_CONFIG_LIBDIR(/sysroot/lib/pkgconfig:/sysroot/usr/lib/pkgconfig) and PKG_CONFIG_SYSROOT_DIR(/sysroot) respectively, and configure with "-sysroot /sysroot -pkg-config", many features test like "Using C++ standard" failed. I think it's due to compiler files missing in /sysroot, so I copy all compiler relevant files to /sysroot, then configure is ok and modules such as openssl can be found, but qmake will link the libc/libm in /sysroot but not the path the compiler installed. if I add "-no-gcc-sysroot" in configure, then the module like openssl can't be found. I wonder is there a way to use 3rd libraries by pkg-config but not change the compile path? ps: the qt is 5.6.2 and the compiler is arm-none-linux-gnueabi for arm B.R. Sun -------------- next part -------------- An HTML attachment was scrubbed... URL: From jani.heikkinen at qt.io Wed Sep 20 13:42:45 2017 From: jani.heikkinen at qt.io (Jani Heikkinen) Date: Wed, 20 Sep 2017 11:42:45 +0000 Subject: [Interest] QT 5.9.2 Release schedule In-Reply-To: References: Message-ID: Hi, We just started branching from '5.9' to '5.9.2' and target is to finalize it at the beginning of next week. So during September is still possible if there isn't any surprises. It will be tight so it is possible that we need to postpone the release to the beginning of October (week 40). Let's see... ;) br, Jani ________________________________________ From: Satunin, Konstantin Sent: Wednesday, September 20, 2017 2:37 PM To: interest at qt-project.org Cc: Jani Heikkinen Subject: QT 5.9.2 Release schedule Hi All, We are interested in the bug fix for memory leak https://bugreports.qt.io/browse/QTBUG-61754 which will be available in 5.9.2 release. QT 5.9.2 snapshot version is available since August 30, and QT 5.10 Alpha was released last week. Can anybody estimate when 5.9.2 might be released? Is it still according to the planned “September 2017” window (only 2 weeks left)? Best wishes, Satunin, Konstantin (Keeper) SSG - Sensor Developer Products -------------------------------------------------------------------- Joint Stock Company Intel A/O Registered legal address: Krylatsky Hills Business Park, 17 Krylatskaya Str., Bldg 4, Moscow 121614, Russian Federation This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. From rmanni at gmail.com Wed Sep 20 14:16:51 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Wed, 20 Sep 2017 14:16:51 +0200 Subject: [Interest] represent QHostAddress as locale dependent string Message-ID: Hi, is there a way convert an ip address locale dependent as string? I mean, similarly as QLocale().toString() does for numbers or dates: for eg arabic locale it gives back arabic "hieroglyphs". I have tried QHostAddress::toString(), but it still returns european numbers for eg arabic. Thanks for any idea in advance, Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From annulen at yandex.ru Wed Sep 20 14:35:52 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Wed, 20 Sep 2017 15:35:52 +0300 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: References: Message-ID: <23001505910952@web46j.yandex.ru> 20.09.2017, 15:17, "Manner Róbert" : > Hi, > > is there a way convert an ip address locale dependent as string? I mean, similarly as QLocale().toString() does for numbers or dates: for eg arabic locale it gives back arabic "hieroglyphs". > > I have tried QHostAddress::toString(), but it still returns european numbers for eg arabic. May be a bit off-topic, but note that those "european numbers" are in fact arabic :) > > Thanks for any idea in advance, > > Robert > , > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest --  Regards, Konstantin From alikin.sergey at gmail.com Wed Sep 20 15:09:56 2017 From: alikin.sergey at gmail.com (=?UTF-8?B?0KHQtdGA0LPQtdC5INCQ0LvQuNC60LjQvQ==?=) Date: Wed, 20 Sep 2017 18:09:56 +0500 Subject: [Interest] VNC plugin at the same time with EGLFS plugin Message-ID: Hello everybody. Is it possible to start at the same time two plugins? Because we are using vnc for user's support, but in Qt5 I can start only one plugin in one moment, so user see black screen instead our application. Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From patrickkidd at gmail.com Wed Sep 20 16:56:12 2017 From: patrickkidd at gmail.com (Patrick Stinson) Date: Wed, 20 Sep 2017 07:56:12 -0700 Subject: [Interest] Building & Running on High Sierra Message-ID: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). Qt-5.9.1 fails to build in qcoregraphics.mm like so: dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier 'InvalidContext' require_action(inContext != NULL, InvalidContext, err = paramErr); ^ /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier 'InvalidBounds' require_action(inBounds != NULL, InvalidBounds, err = paramErr); ^ /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier 'InvalidImage' require_action(inImage != NULL, InvalidImage, err = paramErr); ^ 3 errors generated. Thoughts? -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 1403 bytes Desc: not available URL: From mitch.curtis at qt.io Wed Sep 20 16:59:23 2017 From: mitch.curtis at qt.io (Mitch Curtis) Date: Wed, 20 Sep 2017 14:59:23 +0000 Subject: [Interest] Building & Running on High Sierra In-Reply-To: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> References: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> Message-ID: Did you recently update Xcode? I just ran into a failure on the 5.10 branch after installing updates: https://bugreports.qt.io/browse/QTBUG-63323 From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org] On Behalf Of Patrick Stinson Sent: Wednesday, 20 September 2017 4:56 PM To: Qt Interest Subject: [Interest] Building & Running on High Sierra Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). Qt-5.9.1 fails to build in qcoregraphics.mm like so: dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier 'InvalidContext' require_action(inContext != NULL, InvalidContext, err = paramErr); ^ /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier 'InvalidBounds' require_action(inBounds != NULL, InvalidBounds, err = paramErr); ^ /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier 'InvalidImage' require_action(inImage != NULL, InvalidImage, err = paramErr); ^ 3 errors generated. Thoughts? -------------- next part -------------- An HTML attachment was scrubbed... URL: From patrickkidd at gmail.com Wed Sep 20 17:00:35 2017 From: patrickkidd at gmail.com (Patrick Stinson) Date: Wed, 20 Sep 2017 08:00:35 -0700 Subject: [Interest] Building & Running on High Sierra In-Reply-To: References: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> Message-ID: <3B6E9D50-92A9-4A45-8395-E3B461687A15@gmail.com> Actually yes, I did. Xcode 9 is required for iOS 11. > On Sep 20, 2017, at 7:59 AM, Mitch Curtis wrote: > > Did you recently update Xcode? I just ran into a failure on the 5.10 branch after installing updates: > > https://bugreports.qt.io/browse/QTBUG-63323 > > From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org ] On Behalf Of Patrick Stinson > Sent: Wednesday, 20 September 2017 4:56 PM > To: Qt Interest > > Subject: [Interest] Building & Running on High Sierra > > Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). > > Qt-5.9.1 fails to build in qcoregraphics.mm like so: > > dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm > /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier > 'InvalidContext' > require_action(inContext != NULL, InvalidContext, err = paramErr); > ^ > /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier > 'InvalidBounds' > require_action(inBounds != NULL, InvalidBounds, err = paramErr); > ^ > /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier > 'InvalidImage' > require_action(inImage != NULL, InvalidImage, err = paramErr); > ^ > 3 errors generated. > > Thoughts? -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 1403 bytes Desc: not available URL: From thiago.macieira at intel.com Wed Sep 20 17:47:58 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Wed, 20 Sep 2017 08:47:58 -0700 Subject: [Interest] Qt cross compile issue on pkg-config In-Reply-To: <72e73ab5.253d.15e9e836db8.Coremail.nus1998@yeah.net> References: <72e73ab5.253d.15e9e836db8.Coremail.nus1998@yeah.net> Message-ID: <1652297.nnqYhn9lEB@tjmaciei-mobl1> On quarta-feira, 20 de setembro de 2017 01:59:27 PDT nus1998 wrote: > I wonder is there a way to use 3rd libraries by pkg-config but not change > the compile path? Set PKG_CONFIG_PATH, not the sysroot. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From thiago.macieira at intel.com Wed Sep 20 17:49:56 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Wed, 20 Sep 2017 08:49:56 -0700 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: References: Message-ID: <6495725.yOvVXRAzdP@tjmaciei-mobl1> On quarta-feira, 20 de setembro de 2017 05:16:51 PDT Manner Róbert wrote: > Hi, > > is there a way convert an ip address locale dependent as string? I mean, > similarly as QLocale().toString() does for numbers or dates: for eg arabic > locale it gives back arabic "hieroglyphs". > > I have tried QHostAddress::toString(), but it still returns european > numbers for eg arabic. And will continue to do so, just as the socket classes and QUrl will never understand require Latin script digits. IP addresses are not localiseable. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From annulen at yandex.ru Wed Sep 20 17:50:58 2017 From: annulen at yandex.ru (Konstantin Tokarev) Date: Wed, 20 Sep 2017 18:50:58 +0300 Subject: [Interest] Qt cross compile issue on pkg-config In-Reply-To: <1652297.nnqYhn9lEB@tjmaciei-mobl1> References: <72e73ab5.253d.15e9e836db8.Coremail.nus1998@yeah.net> <1652297.nnqYhn9lEB@tjmaciei-mobl1> Message-ID: <601981505922658@web34g.yandex.ru> 20.09.2017, 18:48, "Thiago Macieira" : > On quarta-feira, 20 de setembro de 2017 01:59:27 PDT nus1998 wrote: >>  I wonder is there a way to use 3rd libraries by pkg-config but not change >>  the compile path? > > Set PKG_CONFIG_PATH, not the sysroot. If pkg-config files contain paths relative to sysroot, sysroot needs to be set; if they are relative to host file system, PKG_CONFIG_PATH should be used. > > -- > Thiago Macieira - thiago.macieira (AT) intel.com >   Software Architect - Intel Open Source Technology Center > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Regards, Konstantin From thiago.macieira at intel.com Wed Sep 20 18:35:22 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Wed, 20 Sep 2017 09:35:22 -0700 Subject: [Interest] Qt cross compile issue on pkg-config In-Reply-To: <601981505922658@web34g.yandex.ru> References: <72e73ab5.253d.15e9e836db8.Coremail.nus1998@yeah.net> <1652297.nnqYhn9lEB@tjmaciei-mobl1> <601981505922658@web34g.yandex.ru> Message-ID: <5864311.H0MUKFvDG0@tjmaciei-mobl1> On quarta-feira, 20 de setembro de 2017 08:50:58 PDT Konstantin Tokarev wrote: > 20.09.2017, 18:48, "Thiago Macieira" : > > On quarta-feira, 20 de setembro de 2017 01:59:27 PDT nus1998 wrote: > >> I wonder is there a way to use 3rd libraries by pkg-config but not > >> change > >> the compile path? > > > > Set PKG_CONFIG_PATH, not the sysroot. > > If pkg-config files contain paths relative to sysroot, sysroot needs to be > set; if they are relative to host file system, PKG_CONFIG_PATH should be > used. Right. But if you're using a sysroot, you have to make sure ALL your libraries are in that sysroot, whether found by pkg-config or not. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From qt at tastytech.ca Wed Sep 20 19:43:04 2017 From: qt at tastytech.ca (Gil H) Date: Wed, 20 Sep 2017 17:43:04 +0000 (UTC) Subject: [Interest] Qt3D Stencil example? Message-ID: <5CA2F584-891D-40F1-8436-D1246CB11101@tastytech.ca> Hi everyone, between StencilMask, StencilTest, and StencilOperation QML items and their sparse documentation, I have been unable to achieve something fairly straightforward with a custom framegraph. Specifically, I am trying to render one object (say, a torus) to the stencil buffer, and then use that to mask out the rendering of the rest of the scene in the second pass. Can anyone provide an example frame graph that does this? Thanks, Gil -------------- next part -------------- An HTML attachment was scrubbed... URL: From patrickkidd at gmail.com Wed Sep 20 20:19:21 2017 From: patrickkidd at gmail.com (Patrick Stinson) Date: Wed, 20 Sep 2017 11:19:21 -0700 Subject: [Interest] Building & Running on High Sierra In-Reply-To: <0264E785-12F4-43F8-ACF9-B9BA44EBAFDC@qt.io> References: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> <3B6E9D50-92A9-4A45-8395-E3B461687A15@gmail.com> <0264E785-12F4-43F8-ACF9-B9BA44EBAFDC@qt.io> Message-ID: Thank you. Any thoughts about the branch problem? > On Sep 20, 2017, at 10:45 AM, Jake Petroules wrote: > > This is QTBUG-62266, which has been fixed by https://codereview.qt-project.org/#/c/201546/ for Qt 5.9.2. > >> On Sep 20, 2017, at 8:00 AM, Patrick Stinson wrote: >> >> Actually yes, I did. Xcode 9 is required for iOS 11. >> >> >> >>> On Sep 20, 2017, at 7:59 AM, Mitch Curtis wrote: >>> >>> Did you recently update Xcode? I just ran into a failure on the 5.10 branch after installing updates: >>> >>> https://bugreports.qt.io/browse/QTBUG-63323 >>> >>> From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org] On Behalf Of Patrick Stinson >>> Sent: Wednesday, 20 September 2017 4:56 PM >>> To: Qt Interest >>> Subject: [Interest] Building & Running on High Sierra >>> >>> Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). >>> >>> Qt-5.9.1 fails to build in qcoregraphics.mm like so: >>> >>> dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm >>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier >>> 'InvalidContext' >>> require_action(inContext != NULL, InvalidContext, err = paramErr); >>> ^ >>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier >>> 'InvalidBounds' >>> require_action(inBounds != NULL, InvalidBounds, err = paramErr); >>> ^ >>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier >>> 'InvalidImage' >>> require_action(inImage != NULL, InvalidImage, err = paramErr); >>> ^ >>> 3 errors generated. >>> >>> Thoughts? >> >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > > -- > Jake Petroules - jake.petroules at qt.io > The Qt Company - Silicon Valley > Qbs build tool evangelist - qbs.io > From patrickkidd at gmail.com Wed Sep 20 21:40:56 2017 From: patrickkidd at gmail.com (Patrick Stinson) Date: Wed, 20 Sep 2017 12:40:56 -0700 Subject: [Interest] Building & Running on High Sierra In-Reply-To: References: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> <3B6E9D50-92A9-4A45-8395-E3B461687A15@gmail.com> <0264E785-12F4-43F8-ACF9-B9BA44EBAFDC@qt.io> Message-ID: <655E1914-1538-4C65-990B-FAEF0BBC54E0@gmail.com> that looks like it. I guess i should have searched for 10.13 instead of high sierra > On Sep 20, 2017, at 11:28 AM, Jake Petroules wrote: > > Likely https://codereview.qt-project.org/#/c/205954/ > >> On Sep 20, 2017, at 11:19 AM, Patrick Stinson wrote: >> >> Thank you. Any thoughts about the branch problem? >> >>> On Sep 20, 2017, at 10:45 AM, Jake Petroules wrote: >>> >>> This is QTBUG-62266, which has been fixed by https://codereview.qt-project.org/#/c/201546/ for Qt 5.9.2. >>> >>>> On Sep 20, 2017, at 8:00 AM, Patrick Stinson wrote: >>>> >>>> Actually yes, I did. Xcode 9 is required for iOS 11. >>>> >>>> >>>> >>>>> On Sep 20, 2017, at 7:59 AM, Mitch Curtis wrote: >>>>> >>>>> Did you recently update Xcode? I just ran into a failure on the 5.10 branch after installing updates: >>>>> >>>>> https://bugreports.qt.io/browse/QTBUG-63323 >>>>> >>>>> From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org] On Behalf Of Patrick Stinson >>>>> Sent: Wednesday, 20 September 2017 4:56 PM >>>>> To: Qt Interest >>>>> Subject: [Interest] Building & Running on High Sierra >>>>> >>>>> Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). >>>>> >>>>> Qt-5.9.1 fails to build in qcoregraphics.mm like so: >>>>> >>>>> dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm >>>>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier >>>>> 'InvalidContext' >>>>> require_action(inContext != NULL, InvalidContext, err = paramErr); >>>>> ^ >>>>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier >>>>> 'InvalidBounds' >>>>> require_action(inBounds != NULL, InvalidBounds, err = paramErr); >>>>> ^ >>>>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier >>>>> 'InvalidImage' >>>>> require_action(inImage != NULL, InvalidImage, err = paramErr); >>>>> ^ >>>>> 3 errors generated. >>>>> >>>>> Thoughts? >>>> >>>> _______________________________________________ >>>> Interest mailing list >>>> Interest at qt-project.org >>>> http://lists.qt-project.org/mailman/listinfo/interest >>> >>> -- >>> Jake Petroules - jake.petroules at qt.io >>> The Qt Company - Silicon Valley >>> Qbs build tool evangelist - qbs.io >>> >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 1403 bytes Desc: not available URL: From nus1998 at yeah.net Thu Sep 21 08:07:12 2017 From: nus1998 at yeah.net (nus1998) Date: Thu, 21 Sep 2017 14:07:12 +0800 (CST) Subject: [Interest] Qt cross compile issue on pkg-config In-Reply-To: <601981505922658@web34g.yandex.ru> References: <72e73ab5.253d.15e9e836db8.Coremail.nus1998@yeah.net> <1652297.nnqYhn9lEB@tjmaciei-mobl1> <601981505922658@web34g.yandex.ru> Message-ID: <164c10cf.20d4.15ea30c19cc.Coremail.nus1998@yeah.net> OK, I set PKG_CONFIG_LIBDIR or PKG_CONFIG_PATH, variables such as QT_LIBS_GLIB are set correctly, however, there are many error on compiling such as compiling qttools\src\qdbus\qdbus, I must add followed text to *.pro file manually. contains(QT_CONFIG, glib) { QMAKE_CXXFLAGS += $$QT_CFLAGS_GLIB LIBS_PRIVATE +=$$QT_LIBS_GLIB } At 2017-09-20 23:50:58, "Konstantin Tokarev" wrote: > > >20.09.2017, 18:48, "Thiago Macieira" : >> On quarta-feira, 20 de setembro de 2017 01:59:27 PDT nus1998 wrote: >>> I wonder is there a way to use 3rd libraries by pkg-config but not change >>> the compile path? >> >> Set PKG_CONFIG_PATH, not the sysroot. > >If pkg-config files contain paths relative to sysroot, sysroot needs to be set; >if they are relative to host file system, PKG_CONFIG_PATH should be used. > >> >> -- >> Thiago Macieira - thiago.macieira (AT) intel.com >> Software Architect - Intel Open Source Technology Center >> >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > >-- >Regards, >Konstantin >_______________________________________________ >Interest mailing list >Interest at qt-project.org >http://lists.qt-project.org/mailman/listinfo/interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From rmanni at gmail.com Thu Sep 21 09:42:12 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Thu, 21 Sep 2017 09:42:12 +0200 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: <6495725.yOvVXRAzdP@tjmaciei-mobl1> References: <6495725.yOvVXRAzdP@tjmaciei-mobl1> Message-ID: On Wed, Sep 20, 2017 at 5:49 PM, Thiago Macieira wrote: > On quarta-feira, 20 de setembro de 2017 05:16:51 PDT Manner Róbert wrote: > > Hi, > > > > is there a way convert an ip address locale dependent as string? I mean, > > similarly as QLocale().toString() does for numbers or dates: for eg > arabic > > locale it gives back arabic "hieroglyphs". > > > > I have tried QHostAddress::toString(), but it still returns european > > numbers for eg arabic. > > And will continue to do so, just as the socket classes and QUrl will never > understand require Latin script digits. > > IP addresses are not localiseable. > I understand your point, and most probably there are bigger problems in the world than this... I just wish if I knew if this is really a problem for arabic users, never using myself apps other than english. I would suspect arabic users got used to that numbers look sometimes arabic sometimes not, similarly like us hungarians got used to the partial translations half the app remaining in english, or even worse. Still, it could be a useful feature to add such a helper function somewhere in Qt: even if ip addresses are not localisable, numbers are. And ip addresses are multiple numbers. If you would have such, perhaps more applications would represent them correctly. > -- > Thiago Macieira - thiago.macieira (AT) intel.com > Software Architect - Intel Open Source Technology Center > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rmanni at gmail.com Thu Sep 21 09:43:17 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Thu, 21 Sep 2017 09:43:17 +0200 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: <23001505910952@web46j.yandex.ru> References: <23001505910952@web46j.yandex.ru> Message-ID: On Wed, Sep 20, 2017 at 2:35 PM, Konstantin Tokarev wrote: > > > 20.09.2017, 15:17, "Manner Róbert" : > > Hi, > > > > is there a way convert an ip address locale dependent as string? I mean, > similarly as QLocale().toString() does for numbers or dates: for eg arabic > locale it gives back arabic "hieroglyphs". > > > > I have tried QHostAddress::toString(), but it still returns european > numbers for eg arabic. > > May be a bit off-topic, but note that those "european numbers" are in fact > arabic :) > Good point, but if I talk about arabic numbers compared to arabic numbers, noone would understand my problem :) > > > > > Thanks for any idea in advance, > > > > Robert > > , > > > > _______________________________________________ > > Interest mailing list > > Interest at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/interest > > > -- > Regards, > Konstantin > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rbehm at hushmail.com Thu Sep 21 10:07:27 2017 From: rbehm at hushmail.com (Reinhardt Behm) Date: Thu, 21 Sep 2017 16:07:27 +0800 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: References: <23001505910952@web46j.yandex.ru> Message-ID: <74c4ee518e6ee38d15ff6055ccc250ae@smtp.hushmail.com> On Thursday 21 September 2017 09:43:17 Manner Róbert wrote: > On Wed, Sep 20, 2017 at 2:35 PM, Konstantin Tokarev > wrote: > 20.09.2017, 15:17, "Manner Róbert" : > > Hi, > > > > is there a way convert an ip address locale dependent as string? I mean, > > similarly as QLocale().toString() does for numbers or dates: for eg > > arabic locale it gives back arabic "hieroglyphs". > > > > I have tried QHostAddress::toString(), but it still returns european > > numbers for eg arabic. > May be a bit off-topic, but note that those "european numbers" are in fact > arabic :) > > Good point, but if I talk about arabic numbers compared to arabic numbers, > noone would understand my problem :) Then call them Indian numbers. Because that's what they really are. -- Best Regards Reinhardt Behm From konrad at silmor.de Thu Sep 21 11:13:07 2017 From: konrad at silmor.de (Konrad Rosenbaum) Date: Thu, 21 Sep 2017 11:13:07 +0200 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: References: <6495725.yOvVXRAzdP@tjmaciei-mobl1> Message-ID: Hi, On Thu, September 21, 2017 09:42, Manner Róbert wrote: > On Wed, Sep 20, 2017 at 5:49 PM, Thiago Macieira > > wrote: >> IP addresses are not localiseable. >> > > I understand your point, and most probably there are bigger problems in > the > world than this... I just wish if I knew if this is really a problem for > arabic users, never using myself apps other than english. > I would suspect arabic users got used to that numbers look sometimes > arabic > sometimes not, similarly like us hungarians got used to the partial > translations half the app remaining in english, or even worse. > > Still, it could be a useful feature to add such a helper function > somewhere > in Qt: even if ip addresses are not localisable, numbers are. And ip > addresses are multiple numbers. If you would have such, perhaps more > applications would represent them correctly. You really do not want to open this can of worms! IPv4 uses dot separated decimal numbers. So far so good. European/Latin derived scripts are left to right. Arabic and other Semitic scripts are right to left. Fortunately they use the dot in a very similar way to Latin based scripts. However: after you translated the digits from Latin looking arabic numerals to arabic looking arabic numerals - do you sort the 4 numbers from left to right or from right to left? Japanese, Chinese and several other languages have similar properties. Some of them use different symbols instead of dots where Latin based scripts use the dot. How do you display IPv4 addresses there? IPv6 is even more interesting: it uses hexadecimal numbers separated by colons. In addition to the IPv4 problems: How do you translate the letters and what do you do with the colons? No other program tries to localize IP addresses precisely because it leads to these kinds of problems. Don't start! Keep it simple. Also: always using digits/letters/symbols from the latin plane of Unicode is a pretty good signal to the user that this is formatted left-to-right. BTW: we Europeans still use the same arabic order of digits (highest left) which is quite normal for arabic speakers (speaking the lowest digit first), but awkward for some European languages (like German where in most cases we also speak the lowest digit first in two-digit numbers but read from left to right or we need to count digits before we know what the first word is). But we think it is perfectly normal to do so! In short: stop worrying! The locals have developed ways to deal with those quirks. Konrad From rmanni at gmail.com Thu Sep 21 14:01:31 2017 From: rmanni at gmail.com (=?UTF-8?Q?Manner_R=C3=B3bert?=) Date: Thu, 21 Sep 2017 14:01:31 +0200 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: References: <6495725.yOvVXRAzdP@tjmaciei-mobl1> Message-ID: On Thu, Sep 21, 2017 at 11:13 AM, Konrad Rosenbaum wrote: > Hi, > > On Thu, September 21, 2017 09:42, Manner Róbert wrote: > > On Wed, Sep 20, 2017 at 5:49 PM, Thiago Macieira > > > > wrote: > >> IP addresses are not localiseable. > >> > > > > I understand your point, and most probably there are bigger problems in > > the > > world than this... I just wish if I knew if this is really a problem for > > arabic users, never using myself apps other than english. > > I would suspect arabic users got used to that numbers look sometimes > > arabic > > sometimes not, similarly like us hungarians got used to the partial > > translations half the app remaining in english, or even worse. > > > > Still, it could be a useful feature to add such a helper function > > somewhere > > in Qt: even if ip addresses are not localisable, numbers are. And ip > > addresses are multiple numbers. If you would have such, perhaps more > > applications would represent them correctly. > > You really do not want to open this can of worms! > > IPv4 uses dot separated decimal numbers. So far so good. > > European/Latin derived scripts are left to right. Arabic and other Semitic > scripts are right to left. Fortunately they use the dot in a very similar > way to Latin based scripts. However: after you translated the digits from > Latin looking arabic numerals to arabic looking arabic numerals - do you > sort the 4 numbers from left to right or from right to left? > > Japanese, Chinese and several other languages have similar properties. > Some of them use different symbols instead of dots where Latin based > scripts use the dot. How do you display IPv4 addresses there? > > IPv6 is even more interesting: it uses hexadecimal numbers separated by > colons. In addition to the IPv4 problems: How do you translate the letters > and what do you do with the colons? > > No other program tries to localize IP addresses precisely because it leads > to these kinds of problems. Don't start! Keep it simple. > > Also: always using digits/letters/symbols from the latin plane of Unicode > is a pretty good signal to the user that this is formatted left-to-right. > > BTW: we Europeans still use the same arabic order of digits (highest left) > which is quite normal for arabic speakers (speaking the lowest digit > first), but awkward for some European languages (like German where in most > cases we also speak the lowest digit first in two-digit numbers but read > from left to right or we need to count digits before we know what the > first word is). But we think it is perfectly normal to do so! > > In short: stop worrying! The locals have developed ways to deal with those > quirks. > > > Ok, I am convinced. Thanks for your time! Robert > > Konrad > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From patrickkidd at gmail.com Thu Sep 21 17:35:54 2017 From: patrickkidd at gmail.com (Patrick Stinson) Date: Thu, 21 Sep 2017 08:35:54 -0700 Subject: [Interest] Project ERROR: Could not find feature static Message-ID: I pulled the 5.10 (also tried with ) branch from git into folder A, then configured and built qt from folder B like so: ../../../pyqt-sysroot-base/src/qt5-src/configure -opensource -confirm-license -nomake examples -nomake tests -prefix $PWD But when I try to run qmake on any project file I get the following: Project ERROR: Could not find feature static I imagine this has something to do with a path setting not being set somewhere? Thanks! -Patrick -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 1403 bytes Desc: not available URL: From thiago.macieira at intel.com Thu Sep 21 18:00:30 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Thu, 21 Sep 2017 09:00:30 -0700 Subject: [Interest] represent QHostAddress as locale dependent string In-Reply-To: References: <6495725.yOvVXRAzdP@tjmaciei-mobl1> Message-ID: <2179550.dpcWUY3K5K@tjmaciei-mobl1> On quinta-feira, 21 de setembro de 2017 00:42:12 PDT Manner Róbert wrote: > I understand your point, and most probably there are bigger problems in the > world than this... I just wish if I knew if this is really a problem for > arabic users, never using myself apps other than english. > I would suspect arabic users got used to that numbers look sometimes arabic > sometimes not, similarly like us hungarians got used to the partial > translations half the app remaining in english, or even worse. > > Still, it could be a useful feature to add such a helper function somewhere > in Qt: even if ip addresses are not localisable, numbers are. And ip > addresses are multiple numbers. If you would have such, perhaps more > applications would represent them correctly. You're assuming people even want this. Please point me any reasonable website showing a localised IPv4 addresses. And one with IPv6 too. And after you find that, copy & paste it to your browser location bar. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From thiago.macieira at intel.com Thu Sep 21 18:02:13 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Thu, 21 Sep 2017 09:02:13 -0700 Subject: [Interest] Project ERROR: Could not find feature static In-Reply-To: References: Message-ID: <3425985.Vg15AG0Ime@tjmaciei-mobl1> On quinta-feira, 21 de setembro de 2017 08:35:54 PDT Patrick Stinson wrote: > ../../../pyqt-sysroot-base/src/qt5-src/configure -opensource > -confirm-license -nomake examples -nomake tests -prefix $PWD -prefix $PWD is only allowed if you run directly the qtbase configure. If you're running the top-level one, install elsehwere. -prefix $PWD/qtbase is allowed in that case. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From fpiechocki at gmail.com Fri Sep 22 15:10:27 2017 From: fpiechocki at gmail.com (Filip Piechocki) Date: Fri, 22 Sep 2017 15:10:27 +0200 Subject: [Interest] QtMqtt Message-ID: Hi, It's been over a month since the QtMqtt blog post ( http://blog.qt.io/blog/2017/08/14/introducing-qtmqtt-protocol/) but I can't find the code to download. Is it already available somewhere? Best regards, Filip Piechocki -------------- next part -------------- An HTML attachment was scrubbed... URL: From carlon.luca at gmail.com Fri Sep 22 17:21:56 2017 From: carlon.luca at gmail.com (Luca Carlon) Date: Fri, 22 Sep 2017 17:21:56 +0200 Subject: [Interest] QML profiling shows scene graph not rendering Message-ID: Hello! My app is affected by some stuttering and I'm trying to investigate the reason through the profiler. I'm attaching a shot of what I see with the profiler in vaapi.png. What I see is that there is a hole in the scene graph row, but just a few milliseconds of execution of JS. As I know some MediaPlayer's are restarted in that instant, I created some tests to reproduce the behavior, and what I notice is that one video player stops rendering for some ms when another MediaPlayer is set to another media. I attach my test code in test.qml. The stuttering is not very noticeable with only two players, but I have more than two. I did the same test without the use of gstreamer vaapi (I'm using nVidia decoding here through the gstreamer vaapi plugin), using software decoding, and in that case the situation seems a bit better: attachment software.png. So is it possible that the scene graph is stuck waiting for something in QtMultimedia being done/allocated? Is there any way to move that "processing" in another thread? Thank you! Regards. Luca -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: vaapi.png Type: image/png Size: 16820 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: software.png Type: image/png Size: 17851 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: test.qml Type: text/x-qml Size: 668 bytes Desc: not available URL: From sean.harmer at kdab.com Sat Sep 23 13:22:19 2017 From: sean.harmer at kdab.com (Sean Harmer) Date: Sat, 23 Sep 2017 12:22:19 +0100 Subject: [Interest] Qt3D Stencil example? In-Reply-To: <5CA2F584-891D-40F1-8436-D1246CB11101@tastytech.ca> References: <5CA2F584-891D-40F1-8436-D1246CB11101@tastytech.ca> Message-ID: <3bc16b22-7dd4-a048-a7fd-b3897f3e9c95@kdab.com> Hi, On 20/09/2017 18:43, Gil H wrote: > Hi everyone, between StencilMask, StencilTest, and StencilOperation QML > items and their sparse documentation, I have been unable to achieve > something fairly straightforward with a custom framegraph. The render states work in exactly the same way as raw OpenGL states just in an object oriented fashion. A nice page with a stencil example is: https://open.gl/depthstencils > Specifically, I am trying to render one object (say, a torus) to the > stencil buffer, and then use that to mask out the rendering of the rest > of the scene in the second pass. Can anyone provide an example frame > graph that does this? I've created the small attached example that demonstrates this for you. It renders a torus shaped entity as a mask, followed by a simple cube as the actual scene content. Left clicking on the cube toggles whether the stencil buffers masks in or masks out the scene layer content. I've briefly documented the MaskedForwardRenderer.qml frame graph for you. To control which entities are rendered in each phase of the frame, we use a Layer and LayerFilter combination - don't forget to assign the scene layer to your lights too. To improve the stencil writing pass, you could use a trivial shader program that does the bare minimum of work rather than the phong material I've used here for convenience. Hope this helps, Sean -- Dr Sean Harmer | sean.harmer at kdab.com | Managing Director UK KDAB (UK) Ltd, a KDAB Group company Tel. +44 (0)1625 809908; Sweden (HQ) +46-563-540090 Mobile: +44 (0)7545 140604 KDAB - Qt Experts -------------- next part -------------- A non-text attachment was scrubbed... Name: stencil-mask.zip Type: application/x-zip-compressed Size: 7232 bytes Desc: not available URL: From smurphy at walbro.com Sun Sep 24 02:00:01 2017 From: smurphy at walbro.com (Murphy, Sean) Date: Sun, 24 Sep 2017 00:00:01 +0000 Subject: [Interest] Qml MessageDialog not centered over window Message-ID: <954F3AC9B636FB4F933D586E76E307EC6DF44561@USWCCEXC04> I'm not sure if this is a bug, or if I'm doing something wrong, but I'm trying to display a Qml dialog when an error occurs and it's coming up centered on my monitor, not centered over the application. Relevant code: Window { MessageDialog { id: errorDialog objectName: "errorDialog" title: "Error" text: "" } Item { id: functionItem objectName: "item" function receivedError(text) { console.log("Error received: " + text); errorDialog.text = "Error, " + text; errorDialog.open(); } } } Then on the C++ side I do QObject::connect(feed, SIGNAL(sigError(QVariant)), cv, SLOT(receivedError (QVariant))); The code works - whenever I emit sigError(errorString), the Qml function is called and the dialog opens with the correct message text. It's just that the position of the dialog is wrong. This is occurring on Windows 7, Qt 5.9.1. I'm running on a dual monitor setup, and the dialog is showing up centered on the same monitor (left) that the application is running on, not centered across my entire desktop. If this were a pure C++ application, I would say that the dialog isn't parented, but at least from this documentation, http://doc.qt.io/qt-5/qml-qtquick-dialogs-messagedialog.html, I think I'm parenting it correctly on the Qml side. The documentation states that " A MessageDialog window is automatically transient for its parent window. So whether you declare the dialog inside an Item or inside a Window, the dialog will appear centered over the window containing the item, or over the Window that you declared." Sean From Shawn.Rutledge at qt.io Mon Sep 25 08:37:05 2017 From: Shawn.Rutledge at qt.io (Shawn Rutledge) Date: Mon, 25 Sep 2017 06:37:05 +0000 Subject: [Interest] Qml MessageDialog not centered over window In-Reply-To: <954F3AC9B636FB4F933D586E76E307EC6DF44561@USWCCEXC04> References: <954F3AC9B636FB4F933D586E76E307EC6DF44561@USWCCEXC04> Message-ID: <751EBB66-B354-46B8-A19F-816EB15002FA@qt.io> > On 24 Sep 2017, at 02:00, Murphy, Sean wrote: > The documentation states that " A MessageDialog window is automatically transient for its parent window. So whether you declare the dialog inside an Item or inside a Window, the dialog will appear centered over the window containing the item, or over the Window that you declared.” Yes, it should be automatically transient for its parent. For me it works on Linux and macOS. So I wonder if we have a bug about this transient-parenting in general on Windows. Have you tried any older Qt versions? import QtQuick 2.6 import QtQuick.Window 2.1 import QtQuick.Dialogs 1.2 Window { visible: true MessageDialog { id: errorDialog objectName: "errorDialog" title: "Error" text: "" modality: Qt.NonModal // optional, to make it a separate window on macOS } Item { id: functionItem objectName: "item" function receivedError(text) { console.log("Error received: " + text); errorDialog.text = "Error, " + text; errorDialog.open(); } } Timer { interval: 5000; running: true onTriggered: functionItem.receivedError("timer fired") } } From daniel.dandrada at luxoft.com Mon Sep 25 09:28:28 2017 From: daniel.dandrada at luxoft.com (Daniel d'Andrada) Date: Mon, 25 Sep 2017 09:28:28 +0200 Subject: [Interest] propagateComposedEvents-like behavior with qt controls 2 components (no C++) Message-ID: <4f84bb4e-488f-9f5f-2ca8-a07c88c14ebb@luxoft.com> Hi All, I wanna have the following (A): Foo { Button {} } or (B): Item { Button {} Foo {} } Where "Foo" is an item that is interested only in drag gestures. So when one is performed it does its thing. It ignores clicks, letting them go to the Button element inside/behind it. So A is a QQuickItem::filtersChildMouseEvents way of doing things (like Flickable), whereas B is a MouseArea::propagateComposedEvents approach. B is not an option as QtQuick.Controls2 are not MouseArea based, so it just won't work. To implement Foo as in option A I would have to resort to C++. But I would like a pure-QML solution for this. Is it possible? Best regards, Daniel ________________________________ This e-mail and any attachment(s) are intended only for the recipient(s) named above and others who have been specifically authorized to receive them. They may contain confidential information. If you are not the intended recipient, please do not read this email or its attachment(s). Furthermore, you are hereby notified that any dissemination, distribution or copying of this e-mail and any attachment(s) is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail and then delete this e-mail and any attachment(s) or copies thereof from your system. Thank you. From andrew at dkai.dk Mon Sep 25 09:58:47 2017 From: andrew at dkai.dk (Andrew Ialacci) Date: Mon, 25 Sep 2017 07:58:47 +0000 Subject: [Interest] propagateComposedEvents-like behavior with qt controls 2 components (no C++) In-Reply-To: <4f84bb4e-488f-9f5f-2ca8-a07c88c14ebb@luxoft.com> References: <4f84bb4e-488f-9f5f-2ca8-a07c88c14ebb@luxoft.com> Message-ID: <100FA5C7-3B3D-4DE3-BFCD-62C3C003AF5B@dkai.dk> I think you can chain the signal calls using signal.connect(otherSignal) and then set the drag target to null. I had to implement something like this while writing an MDI windowing system. Admittedly though I did resort to moving 99% of the geometry manipulation and event handling to CPP. I wish there was a way to subclass the QQC2 controls completely that allowed lower level access to each controls event system. Sent from my iPhone > On Sep 25, 2017, at 9:28 AM, Daniel d'Andrada wrote: > > Hi All, > > I wanna have the following (A): > > Foo { > Button {} > } > > or (B): > > Item { > Button {} > Foo {} > } > > Where "Foo" is an item that is interested only in drag gestures. So when > one is performed it does its thing. It ignores clicks, letting them go > to the Button element inside/behind it. > > So A is a QQuickItem::filtersChildMouseEvents way of doing things (like > Flickable), whereas B is a MouseArea::propagateComposedEvents approach. > > B is not an option as QtQuick.Controls2 are not MouseArea based, so it > just won't work. > > To implement Foo as in option A I would have to resort to C++. But I > would like a pure-QML solution for this. Is it possible? > > Best regards, > Daniel > > > ________________________________ > > This e-mail and any attachment(s) are intended only for the recipient(s) named above and others who have been specifically authorized to receive them. They may contain confidential information. If you are not the intended recipient, please do not read this email or its attachment(s). Furthermore, you are hereby notified that any dissemination, distribution or copying of this e-mail and any attachment(s) is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail and then delete this e-mail and any attachment(s) or copies thereof from your system. Thank you. > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From daniel.dandrada at luxoft.com Mon Sep 25 10:35:30 2017 From: daniel.dandrada at luxoft.com (Daniel d'Andrada) Date: Mon, 25 Sep 2017 10:35:30 +0200 Subject: [Interest] propagateComposedEvents-like behavior with qt controls 2 components (no C++) In-Reply-To: <100FA5C7-3B3D-4DE3-BFCD-62C3C003AF5B@dkai.dk> References: <4f84bb4e-488f-9f5f-2ca8-a07c88c14ebb@luxoft.com> <100FA5C7-3B3D-4DE3-BFCD-62C3C003AF5B@dkai.dk> Message-ID: Actually MouseArea.drag.filterChildren is just what I was looking for (so you can implement option A in pure QML). Sorry for the noise. - Daniel On 09/25/2017 09:58 AM, Andrew Ialacci wrote: > I think you can chain the signal calls using signal.connect(otherSignal) and then set the drag target to null. > > I had to implement something like this while writing an MDI windowing system. Admittedly though I did resort to moving 99% of the geometry manipulation and event handling to CPP. > > I wish there was a way to subclass the QQC2 controls completely that allowed lower level access to each controls event system. > > > > Sent from my iPhone > >> On Sep 25, 2017, at 9:28 AM, Daniel d'Andrada wrote: >> >> Hi All, >> >> I wanna have the following (A): >> >> Foo { >> Button {} >> } >> >> or (B): >> >> Item { >> Button {} >> Foo {} >> } >> >> Where "Foo" is an item that is interested only in drag gestures. So when >> one is performed it does its thing. It ignores clicks, letting them go >> to the Button element inside/behind it. >> >> So A is a QQuickItem::filtersChildMouseEvents way of doing things (like >> Flickable), whereas B is a MouseArea::propagateComposedEvents approach. >> >> B is not an option as QtQuick.Controls2 are not MouseArea based, so it >> just won't work. >> >> To implement Foo as in option A I would have to resort to C++. But I >> would like a pure-QML solution for this. Is it possible? >> >> Best regards, >> Daniel >> >> >> ________________________________ >> >> This e-mail and any attachment(s) are intended only for the recipient(s) named above and others who have been specifically authorized to receive them. They may contain confidential information. If you are not the intended recipient, please do not read this email or its attachment(s). Furthermore, you are hereby notified that any dissemination, distribution or copying of this e-mail and any attachment(s) is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail and then delete this e-mail and any attachment(s) or copies thereof from your system. Thank you. >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest ________________________________ This e-mail and any attachment(s) are intended only for the recipient(s) named above and others who have been specifically authorized to receive them. They may contain confidential information. If you are not the intended recipient, please do not read this email or its attachment(s). Furthermore, you are hereby notified that any dissemination, distribution or copying of this e-mail and any attachment(s) is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail and then delete this e-mail and any attachment(s) or copies thereof from your system. Thank you. From Maurice.Kalinowski at qt.io Mon Sep 25 11:05:28 2017 From: Maurice.Kalinowski at qt.io (Maurice Kalinowski) Date: Mon, 25 Sep 2017 09:05:28 +0000 Subject: [Interest] QtMqtt In-Reply-To: References: Message-ID: Hi Filip, I’m afraid to say “not yet”. We’ve been struggling with getting the infrastructure in place (with positive updates over the weekend). Currently there are logistical items to be done (license header updates, etc.), but we are progressing nicely now. The aim is to have everything available (also Qt Knx) within the Qt World Summit timeframe. Probably, or hopefully, earlier. BR, Maurice From: Interest [mailto:interest-bounces+maurice.kalinowski=qt.io at qt-project.org] On Behalf Of Filip Piechocki Sent: Friday, September 22, 2017 3:10 PM To: interest at qt-project.org Interest Subject: [Interest] QtMqtt Hi, It's been over a month since the QtMqtt blog post (http://blog.qt.io/blog/2017/08/14/introducing-qtmqtt-protocol/) but I can't find the code to download. Is it already available somewhere? Best regards, Filip Piechocki -------------- next part -------------- An HTML attachment was scrubbed... URL: From Shawn.Rutledge at qt.io Mon Sep 25 12:00:16 2017 From: Shawn.Rutledge at qt.io (Shawn Rutledge) Date: Mon, 25 Sep 2017 10:00:16 +0000 Subject: [Interest] propagateComposedEvents-like behavior with qt controls 2 components (no C++) In-Reply-To: <4f84bb4e-488f-9f5f-2ca8-a07c88c14ebb@luxoft.com> References: <4f84bb4e-488f-9f5f-2ca8-a07c88c14ebb@luxoft.com> Message-ID: > On 25 Sep 2017, at 09:28, Daniel d'Andrada wrote: > > Hi All, > > I wanna have the following (A): > > Foo { > Button {} > } > > or (B): > > Item { > Button {} > Foo {} > } > > Where "Foo" is an item that is interested only in drag gestures. So when > one is performed it does its thing. It ignores clicks, letting them go > to the Button element inside/behind it. > So A is a QQuickItem::filtersChildMouseEvents way of doing things (like > Flickable), whereas B is a MouseArea::propagateComposedEvents approach. > > B is not an option as QtQuick.Controls2 are not MouseArea based, so it > just won't work. So you just want a draggable Button, right? If Foo is MouseArea, it must grab the press in order to get updates at all. If it does not grab, then Button grabs. It can work if MouseArea is a filtering parent, as documented for MouseArea’s drag.filterChildren property: import QtQuick 2.0 import QtQuick.Controls 2.2 Item { width: 200; height: 200 MouseArea { width: button.width; height: button.height drag.target: button drag.filterChildren: true Button { id: button text: "Drag me" onClicked: console.log("clicked") } } } although ATM I’m seeing a bug that it only works once. The documented way, having a wrapper item so that you can set drag.target: parent, works better: import QtQuick 2.0 import QtQuick.Controls 2.2 Item { width: 200; height: 200 Item { width: button.width; height: button.height MouseArea { anchors.fill: parent drag.target: parent drag.filterChildren: true Button { id: button text: "Drag me" onClicked: console.log("clicked") } } } } > To implement Foo as in option A I would have to resort to C++. But I > would like a pure-QML solution for this. Is it possible? A nicer way: theoretically you ought to be able to do import QtQuick 2.0 import QtQuick.Controls 2.2 import Qt.labs.handlers 1.0 Item { Button { DragHandler { } } } on 5.10 branch; but ATM if you do that, you can drag the button, but you can’t click it. IMO we need to make that work in 5.10. But we still have some trouble with grab conflict: the DragHandler gets a passive grab on press, which is fine, and is intended to mean that it will get the updates without preventing the Button from grabbing; but it also accepts the press event, and therefore Button doesn’t see it. If we make it so that Button does see the press (keep delivering even though the event was already accepted), then Button does an exclusive grab, and then DragHandler doesn’t get the updates when you start dragging. So I’ve been trying to sell the idea for several months that events need to be delivered to passive grabbers regardless of the exclusive grab. The patches languish because people aren’t convinced yet that it’s the right approach. The alternative I can see is for the DragHandler not to accept the event, so that it keeps propagating. QTBUG-63395 to keep track of that. I hope we get it fixed soon one way or another. From iman72411 at gmail.com Mon Sep 25 12:39:37 2017 From: iman72411 at gmail.com (iman ahmadvand) Date: Mon, 25 Sep 2017 14:09:37 +0330 Subject: [Interest] Dealing with unexported class in private API Message-ID: Hi. Nowadays I'm dealing with problem of the symbols absence for some private part classes. for example in qabstractscrollarea_p.h class QAbstractScrollAreaScrollBarContainer is not exported while QAbstractScrollAreaPrivate is. So if you try to access QAbstractScrollAreaPrivate::scrollBarContainers you get linker error ! (seems to happen on windows only) I need to access that class member (scrollBarContainers->layout) without compiling dependencies! Is there a solution ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From kshegunov at gmail.com Mon Sep 25 12:54:18 2017 From: kshegunov at gmail.com (Konstantin Shegunov) Date: Mon, 25 Sep 2017 13:54:18 +0300 Subject: [Interest] Dealing with unexported class in private API In-Reply-To: References: Message-ID: Hi, I'd hate to put a dent in your cool, but there's a big fat warning on top of the private headers - they are private and subject to change without warning or any promise of compatibility. -------------- next part -------------- An HTML attachment was scrubbed... URL: From smurphy at walbro.com Mon Sep 25 13:17:05 2017 From: smurphy at walbro.com (Murphy, Sean) Date: Mon, 25 Sep 2017 11:17:05 +0000 Subject: [Interest] Qml MessageDialog not centered over window In-Reply-To: <751EBB66-B354-46B8-A19F-816EB15002FA@qt.io> References: <954F3AC9B636FB4F933D586E76E307EC6DF44561@USWCCEXC04> <751EBB66-B354-46B8-A19F-816EB15002FA@qt.io> Message-ID: <954F3AC9B636FB4F933D586E76E307EC6DF44708@USWCCEXC04> > Yes, it should be automatically transient for its parent. For me it works on > Linux and macOS. So I wonder if we have a bug about this transient- > parenting in general on Windows. Have you tried any older Qt versions? Well, the example code you sent me worked correctly on the same version of Qt (5.9.1) that I'm using for me real application, so I must be doing something else wrong. First I tried your code as-is in a brand new project. On first run, that had the main window show up centered in the monitor, and the dialog was also centered, so I couldn't tell if the dialog was actually being centered over its parent, or if they were both just being centered on my screen. So I modified Window to be Window { visible: true x: 60 y: 100 .... } And that worked correctly, Window is in upper left corner of my screen, and when the error dialog pops up it is centered over Window. After that, I copied your Timer object into my real application and ran it. That fails - my main Window object is off-center on my screen, the dialog is shown centered on the screen. One thing I noticed is that the error dialogs look different from each other, the attached images show that. The image "testError.png" is what I get when running your sample code in its own application. "actualError.png" is what it looks like when run from inside my actual application, so something else is going on, but I'm not sure what yet. Other differences I can see are my imports import QtQuick 2.6 import QtQuick.Window 2.1 import QtQuick.Controls 2.2 import QtQuick.Controls.Styles 1.4 import QtQuick.Extras 1.4 import QtQuick.Dialogs 1.2 import QtQuick.Layouts 1.3 import QtCharts 2.2 And in my real main.cpp I have QApplication app(argc, argv); QQmlEngine engine; QQmlComponent component(&engine, QUrl("qrc:/qml/dashboard.qml")); component.create(); Whereas the project that was generated by Creator when I was testing your example is QGuiApplication app(argc, argv); QQmlApplicationEngine engine; engine.load(QUrl(QStringLiteral("qrc:/main.qml"))); When I changed the test app to match my actual one, the incorrect behavior returned. I know that I changed the code in main from a QGuiApplication to a QApplication when I started getting an assert when running ASSERT: "!"No style available without QApplication!"" in file kernel\qapplication.cpp, line 1060 I'm not sure what I'm doing wrong to incur that assertion? Unfortunately I can't post my entire actual code here, as it has some intellectual property in it but if you have any ideas on what I should look for I'm open to suggestions! The only "style" things I see in my Qml is that my application is using the CircularGaugeStyle (https://doc.qt.io/qt-5/qml-qtquick-controls-styles-circulargaugestyle.html) which I do need to keep in my application. Still digging... Sean -------------- next part -------------- A non-text attachment was scrubbed... Name: actualError.png Type: image/png Size: 8982 bytes Desc: actualError.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: testError.png Type: image/png Size: 14299 bytes Desc: testError.png URL: From Shawn.Rutledge at qt.io Mon Sep 25 14:09:34 2017 From: Shawn.Rutledge at qt.io (Shawn Rutledge) Date: Mon, 25 Sep 2017 12:09:34 +0000 Subject: [Interest] Qml MessageDialog not centered over window In-Reply-To: <954F3AC9B636FB4F933D586E76E307EC6DF44708@USWCCEXC04> References: <954F3AC9B636FB4F933D586E76E307EC6DF44561@USWCCEXC04> <751EBB66-B354-46B8-A19F-816EB15002FA@qt.io> <954F3AC9B636FB4F933D586E76E307EC6DF44708@USWCCEXC04> Message-ID: <5C50FC56-44BA-4993-BC9C-9BFD11727D94@qt.io> > On 25 Sep 2017, at 13:17, Murphy, Sean wrote: > > And in my real main.cpp I have > QApplication app(argc, argv); > QQmlEngine engine; > QQmlComponent component(&engine, QUrl("qrc:/qml/dashboard.qml")); > component.create(); > Whereas the project that was generated by Creator when I was testing your > example is > QGuiApplication app(argc, argv); > QQmlApplicationEngine engine; > engine.load(QUrl(QStringLiteral("qrc:/main.qml"))); Aha, it happens when using a QApplication; the MessageDialog is actually a QMessageBox in that case. I can get the same result with the test qml like this: qml -apptype widget messagedialog.qml (that makes the qml runtime use a QApplication instead of QGuiApplication) and maybe we have some trouble making a QWidget dialog transient for a QWindow. In QMessageBoxHelper::show(), it does try to set its transient parent: virtual bool show(Qt::WindowFlags f, Qt::WindowModality m, QWindow *parent) { m_dialog.winId(); QWindow *window = m_dialog.windowHandle(); Q_ASSERT(window); window->setTransientParent(parent); qDebug() << window << "transient for" << parent; … } QMessageBoxHelper::show - QWidgetWindow(0x7fcf52d80a40, name="QMessageBoxClassWindow") transient for QQuickWindowQmlImpl(0x7fcf54a192c0) I wrote it up as QTBUG-63406. From smurphy at walbro.com Mon Sep 25 14:57:25 2017 From: smurphy at walbro.com (Murphy, Sean) Date: Mon, 25 Sep 2017 12:57:25 +0000 Subject: [Interest] Qml MessageDialog not centered over window In-Reply-To: <5C50FC56-44BA-4993-BC9C-9BFD11727D94@qt.io> References: <954F3AC9B636FB4F933D586E76E307EC6DF44561@USWCCEXC04> <751EBB66-B354-46B8-A19F-816EB15002FA@qt.io> <954F3AC9B636FB4F933D586E76E307EC6DF44708@USWCCEXC04> <5C50FC56-44BA-4993-BC9C-9BFD11727D94@qt.io> Message-ID: <954F3AC9B636FB4F933D586E76E307EC6DF4474F@USWCCEXC04> > Aha, it happens when using a QApplication; the MessageDialog is actually a > QMessageBox in that case. I can get the same result with the test qml like > this: > > qml -apptype widget messagedialog.qml > > (that makes the qml runtime use a QApplication instead of QGuiApplication) > and maybe we have some trouble making a QWidget dialog transient for a > QWindow. In QMessageBoxHelper::show(), it does try to set its transient > parent: > > virtual bool show(Qt::WindowFlags f, Qt::WindowModality m, QWindow > *parent) { > m_dialog.winId(); > QWindow *window = m_dialog.windowHandle(); > Q_ASSERT(window); > window->setTransientParent(parent); > qDebug() << window << "transient for" << parent; > … } > > QMessageBoxHelper::show - QWidgetWindow(0x7fcf52d80a40, > name="QMessageBoxClassWindow") transient for > QQuickWindowQmlImpl(0x7fcf54a192c0) > > I wrote it up as QTBUG-63406. Ok. I still need to figure out what I'm doing wrong in my application that I'm not able to use a QGuiApplication without getting that assert for some reason. I really would like it to be a MessageDialog, not a QMessageBox, but as I've got things right now, I can't seem to get QGuiApplication to work. My application is pretty lightweight, and it is based off from the Qt Quick Extras - Dashboard in the examples folder. That example runs just fine using QGuiApplication, so I don't get what I've done to it that makes it assert now. I guess I need to start from the working example and begin adding in chunks of my actual application until I find out what the problems is. Sean From thiago.macieira at intel.com Mon Sep 25 18:03:00 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Mon, 25 Sep 2017 09:03:00 -0700 Subject: [Interest] Dealing with unexported class in private API In-Reply-To: References: Message-ID: <3472037.0WcTUEUyEr@tjmaciei-mobl1> On Monday, 25 September 2017 03:39:37 PDT iman ahmadvand wrote: > Hi. > Nowadays I'm dealing with problem of the symbols absence for some private > part classes. > for example in qabstractscrollarea_p.h class > QAbstractScrollAreaScrollBarContainer > is not exported while QAbstractScrollAreaPrivate is. > So if you try to access QAbstractScrollAreaPrivate::scrollBarContainers you > get linker error ! (seems to happen on windows only) > > I need to access that class member (scrollBarContainers->layout) without > compiling dependencies! > Is there a solution ? Yes, there are two possibilities: 1) put up with the problem. Figure out a way to do what you want to do without modifying the header. And please don't ask questions about the private API either: you can use them at your own risk, but you need to understand them on your own. Please read the "We mean it" comment. 2) import your fuctionality into Qt. That way, we can review your use of the priate APi and decide whether it makes sense to export more private API. Note that there's a variant to solution (1), which is to add *public* API that allows you to achieve what you want to achieve. That can be submitted to the code review for Qt 5.11 (dev branch) with an explanation of what it does and why it's good, without submitting the use of it. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From qt at tastytech.ca Mon Sep 25 22:25:33 2017 From: qt at tastytech.ca (Gil H) Date: Mon, 25 Sep 2017 20:25:33 +0000 (UTC) Subject: [Interest] Qt3D Stencil example? In-Reply-To: <3bc16b22-7dd4-a048-a7fd-b3897f3e9c95@kdab.com> References: <5CA2F584-891D-40F1-8436-D1246CB11101@tastytech.ca> <3bc16b22-7dd4-a048-a7fd-b3897f3e9c95@kdab.com> Message-ID: <3D137EEB-9DBE-46AF-A211-C5592B3BEC00@tastytech.ca> Thanks very much for the example, Sean. It’s much clearer now that I see RenderStates in use (for some reason I thought they only applied in Effect)… There is a pretty serious issue that is blocking me from using Qt3D for augmented reality, which is that QtMultimedia’s VideoOutput doesn’t work when the window surface uses CoreProfile (tested on Mac). This issue was given a low priority and is a year and a half old: https://u5810871.ct.sendgrid.net/wf/click?upn=RgE1ASyOg54bH5KXOAIcDCNgCWqccrQ-2FpQuIrXIrfIJMy2TtuL9LVEN2mNDqCmUOgEyuwohNEwt6hzEhA36JiQ-3D-3D_wBwg3POqus3ycpk-2F0rxnkTh6GGX9m8uP6qHpajPv1FEqh6WFzqVsUSHUCdA-2BOqQYKo2R18TsAPAgMZjZ6A-2FM4U6WJnaOkGJxNWR4KicbxdQdC6786gVblgUviLccyhvIpdD-2BkA71HbGmYT5QFX2aFDmXO7E20GBfRlbnRM7NWG6ojJiCN5kpbQtvBboIHkFMI5Tsp1XKcf0z1BUkFrlLbHa-2BJlA8eQGlDKDhKJcwMKs-3D Cheers, Gil On 2017-09-23, 7:22 AM, "Sean Harmer" wrote: Hi, On 20/09/2017 18:43, Gil H wrote: > Hi everyone, between StencilMask, StencilTest, and StencilOperation QML > items and their sparse documentation, I have been unable to achieve > something fairly straightforward with a custom framegraph. The render states work in exactly the same way as raw OpenGL states just in an object oriented fashion. A nice page with a stencil example is: https://u5810871.ct.sendgrid.net/wf/click?upn=RgE1ASyOg54bH5KXOAIcDFYXmPTHFqeoF2wSuQEunwWDSxtesUqiG-2FgN-2BSdVCPuz_wBwg3POqus3ycpk-2F0rxnkTh6GGX9m8uP6qHpajPv1FEqh6WFzqVsUSHUCdA-2BOqQYKo2R18TsAPAgMZjZ6A-2FM4YMOVTBDEAM-2Br0R3h6mOurpsacDIqxI15OsfB3PV-2B6lxAP1emuhCZmmCFiSBd-2FOvOSw8-2FPtkkzJbzCK3mwgnzX63ZNn2-2FZalCel4DsnTFwO7wtRM-2Fwx-2FLYZuBPYL4cRj7w9b0Ys-2BCPkvhdmcb6VTfJA-3D > Specifically, I am trying to render one object (say, a torus) to the > stencil buffer, and then use that to mask out the rendering of the rest > of the scene in the second pass. Can anyone provide an example frame > graph that does this? I've created the small attached example that demonstrates this for you. It renders a torus shaped entity as a mask, followed by a simple cube as the actual scene content. Left clicking on the cube toggles whether the stencil buffers masks in or masks out the scene layer content. I've briefly documented the MaskedForwardRenderer.qml frame graph for you. To control which entities are rendered in each phase of the frame, we use a Layer and LayerFilter combination - don't forget to assign the scene layer to your lights too. To improve the stencil writing pass, you could use a trivial shader program that does the bare minimum of work rather than the phong material I've used here for convenience. Hope this helps, Sean -- Dr Sean Harmer | sean.harmer at kdab.com | Managing Director UK KDAB (UK) Ltd, a KDAB Group company Tel. +44 (0)1625 809908; Sweden (HQ) +46-563-540090 Mobile: +44 (0)7545 140604 KDAB - Qt Experts -------------- next part -------------- An HTML attachment was scrubbed... URL: From Sudhir.Sharma at mindtree.com Tue Sep 26 07:44:58 2017 From: Sudhir.Sharma at mindtree.com (Sudhir Sharma) Date: Tue, 26 Sep 2017 05:44:58 +0000 Subject: [Interest] WebEngineView crashing in 5.9.1 Message-ID: Hi, I have compiled 5.9.1 in win 64, and using it in a machine which also has previous version of QT (4.8) installed. On starting my application I am getting exception at following line, "webEngineView->setUrl(QUrl(QStringLiteral("about:blank")));" Whereas in a fresh machine which only has 5.9.1, it does no crash. Any idea what would the conflict? Regards, Sudhir ________________________________ http://www.mindtree.com/email/disclaimer.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From thiago.macieira at intel.com Tue Sep 26 08:15:56 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Mon, 25 Sep 2017 23:15:56 -0700 Subject: [Interest] WebEngineView crashing in 5.9.1 In-Reply-To: References: Message-ID: <1684337.SoLi0gTVxo@tjmaciei-mobl1> On segunda-feira, 25 de setembro de 2017 22:44:58 PDT Sudhir Sharma wrote: > "webEngineView->setUrl(QUrl(QStringLiteral("about:blank")));" What's the value of webEngineView at this point? Are you sure it's valid? -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From Sudhir.Sharma at mindtree.com Tue Sep 26 08:41:09 2017 From: Sudhir.Sharma at mindtree.com (Sudhir Sharma) Date: Tue, 26 Sep 2017 06:41:09 +0000 Subject: [Interest] WebEngineView crashing in 5.9.1 In-Reply-To: <1684337.SoLi0gTVxo@tjmaciei-mobl1> References: <1684337.SoLi0gTVxo@tjmaciei-mobl1> Message-ID: I see following in call trace: -webEngineView0x000000d92c8381e0 {d_ptr=scoped pointer to a dynamically allocated object of type "QWebEngineViewPrivate" }QWebEngineView * -QWidget{data=0x000000d941693430 {winid=0 widget_attributes=0 window_flags={i=-2013204480 } ...} }QWidget +QObject{d_ptr=scoped pointer to a dynamically allocated object of type "QObjectData" }QObject +QPaintDevice{painters=0 reserved=0x0000000000000000 }QPaintDevice +data0x000000d941693430 {winid=0 widget_attributes=0 window_flags={i=-2013204480 } ...}QWidgetData * -d_ptrscoped pointer to a dynamically allocated object of type "QWebEngineViewPrivate"QScopedPointer > [is null]falsebool -[Raw View]{d=0x000000d930609bf0 {...} }QScopedPointer > d0x000000d930609bf0 {...}QWebEngineViewPrivate * Regards, Sudhir -----Original Message----- From: Interest [mailto:interest-bounces+sudhir.sharma=mindtree.com at qt-project.org] On Behalf Of Thiago Macieira Sent: Tuesday, September 26, 2017 11:46 AM To: interest at qt-project.org Subject: Re: [Interest] WebEngineView crashing in 5.9.1 * You are receiving this mail from an external source * On segunda-feira, 25 de setembro de 2017 22:44:58 PDT Sudhir Sharma wrote: > "webEngineView->setUrl(QUrl(QStringLiteral("about:blank")));" What's the value of webEngineView at this point? Are you sure it's valid? -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center _______________________________________________ Interest mailing list Interest at qt-project.org http://lists.qt-project.org/mailman/listinfo/interest ________________________________ http://www.mindtree.com/email/disclaimer.html From Kai.Koehne at qt.io Tue Sep 26 09:42:15 2017 From: Kai.Koehne at qt.io (Kai Koehne) Date: Tue, 26 Sep 2017 07:42:15 +0000 Subject: [Interest] WebEngineView crashing in 5.9.1 In-Reply-To: References: Message-ID: > -----Original Message----- > From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > On Behalf Of Sudhir Sharma > Sent: Tuesday, September 26, 2017 7:45 AM > To: interest at qt-project.org > Subject: [Interest] WebEngineView crashing in 5.9.1 > > Hi, > > I have compiled 5.9.1 in win 64, and using it in a machine which also has > previous version of QT (4.8) installed. > > On starting my application I am getting exception at following line, > > "webEngineView->setUrl(QUrl(QStringLiteral("about:blank")));" > > Whereas in a fresh machine which only has 5.9.1, it does no crash. > Any idea what would the conflict? No idea. But you can load your application into the Dependency Walker (depends.exe), and start it in profiling mode (F7). Make sure that you disable profiling of sub-processes ... Feel free to attach the output to a bug report then. Also, please try to reproduce with a minimal example (e.g. webenginewidgets/minimal) first. Regards Kai From coroberti at gmail.com Tue Sep 26 09:45:23 2017 From: coroberti at gmail.com (coroberti .) Date: Tue, 26 Sep 2017 10:45:23 +0300 Subject: [Interest] Qt iOS - Customization of EditMenu Message-ID: Hi, iOS developers are able to customize EditMenu (the one used to Select, Select All, Copy, Cut, Paste, Speak, etc.) using clear Apple's API. Once I've used the approach recommended by Richard Moe Gustavsen: https://bugreports.qt.io/browse/QTBUG-41275 that uses private API and is not working since Qt 5.5 Since that I cannot get rid from the standard items, for example, or provide my custom actions on selecting them. There is more mess related: https://bugreports.qt.io/browse/QTBUG-49426 Any clues or directions or the maintainer would like this to be filed as a feature request? Thank you in advance! Kind regards, Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From julius.bullinger at intel.com Tue Sep 26 11:37:25 2017 From: julius.bullinger at intel.com (Bullinger, Julius) Date: Tue, 26 Sep 2017 09:37:25 +0000 Subject: [Interest] QGeoPositionInfo in QML Message-ID: <7FC8D04D3E708B4AA8248F6D64E0CB2136BC5B72@IRSMSX104.ger.corp.intel.com> Hi there, in Qt Location, there's the QGeoPositionInfo class for storing geo positions. There's also the Position QML type with a matching interface to do the same in QML. However, I haven't had success trying to expose a QGeoPosition object as Q_PROPERTY. In QML, the type will be reported as QVariant, and I cannot access any properties. What is the supposed way to use QGeoPositionInfo in QML? There's QDeclarativePosition, but it's private. Thanks and best regards, Julius Intel Deutschland GmbH Registered Address: Am Campeon 10-12, 85579 Neubiberg, Germany Tel: +49 89 99 8853-0, www.intel.de Managing Directors: Christin Eisenschmid, Christian Lamprechter Chairperson of the Supervisory Board: Nicole Lau Registered Office: Munich Commercial Register: Amtsgericht Muenchen HRB 186928 -------------- next part -------------- An HTML attachment was scrubbed... URL: From roland at logikalsolutions.com Tue Sep 26 19:56:33 2017 From: roland at logikalsolutions.com (Roland Hughes) Date: Tue, 26 Sep 2017 12:56:33 -0500 Subject: [Interest] Cross compiling for Pi with PostgreSQL In-Reply-To: References: Message-ID: All, Has anybody on this list managed to use the cross compilation instructions from either here: https://www.ics.com/blog/configuring-qt-creator-raspberry-pi or here: https://wiki.qt.io/RaspberryPi2EGLFS and gotten PostgreSQL support to compile when the host is a 64-bit system and the target is 32-bit Pi? Both the wiki and that link seem horribly broken. You will notice the environment/path configuration is incorrect: ==== checking for PostgreSQL... /home/roland/piBuild/pitools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-g++ -c -pipe -march=armv7-a -marm -mthumb-interwork -mfpu=neon-vfpv4 -mtune=cortex-a7 -mabi=aapcs-linux -mfloat-abi=hard --sysroot=/home/roland/piBuild/sysroot -O2 -Wall -W -fPIC  -I. -I/usr/include/postgresql -I../../../mkspecs/devices/linux-rasp-pi2-g++ -o psql.o psql.cpp /home/roland/piBuild/pitools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-g++ -Wl,-rpath-link,/home/roland/piBuild/sysroot/opt/vc/lib -Wl,-rpath-link,/home/roland/piBuild/sysroot/usr/lib/arm-linux-gnueabihf -Wl,-rpath-link,/home/roland/piBuild/sysroot/lib/arm-linux-gnueabihf -mfloat-abi=hard -Wl,-O1 -o psql psql.o   --sysroot=/home/roland/piBuild/sysroot -L/usr/lib/x86_64-linux-gnu -lpq /usr/lib/x86_64-linux-gnu/libpq.so: file not recognized: File format not recognized collect2: error: ld returned 1 exit status Makefile:92: recipe for target 'psql' failed make: *** [psql] Error 1 PostgreSQL disabled. PostgreSQL support cannot be enabled due to functionality tests!  Turn on verbose messaging (-v) to /home/roland/piBuild/qt-everywhere-opensource-src-5.7.1/qtbase/configure to see the final report.  If you believe this message is in error you may use the continue  switch (-continue) to /home/roland/piBuild/qt-everywhere-opensource-src-5.7.1/qtbase/configure to continue. roland at roland-desktop:~/piBuild/qt-everywhere-opensource-src-5.7.1$ cd / roland at roland-desktop:/$ sudo find -iname libpq.so [sudo] password for roland: ./usr/lib/x86_64-linux-gnu/libpq.so ./home/roland/piBuild/sysroot/usr/lib/arm-linux-gnueabihf/libpq.so ==== Even after the library and required headers are installed on the Pi and the entire process is started from scratch the cross compile looks in the wrong place for the header (on the 64-bit machine) and the tries to link with the 64-bit version of the lib. Has anybody found the work around for this? Preferably a work around which does not require loop mounting the SD image. -- Roland Hughes, President Logikal Solutions (630)-205-1593 http://www.theminimumyouneedtoknow.com http://www.infiniteexposure.net http://www.johnsmith-book.com http://www.logikalblog.com http://www.interestingauthors.com/blog http://lesedi.us/ http://onedollarcontentstore.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jtykka at broadsoft.com Tue Sep 26 20:55:35 2017 From: jtykka at broadsoft.com (Jani Tykka) Date: Tue, 26 Sep 2017 21:55:35 +0300 Subject: [Interest] Issue with concurrent QNetworkRequests Message-ID: Hi, I'd ask if the following issues would ring anyones bells, before I start doing more deeper investigation and create test application to reproduce the issue. Basically the issue in our application is reproduced following way. 1. Have 1 big download (https) 100MB (It uses own QNam instance) - file is being written from the data got from QNetworkReply::readyRead 2. During the download execute other requests (https) using different QNam instance What is experienced that if during download other requests are executed the data received from readyRead contains extra bytes in the middle of chunked responses. It can be that the total file size received grows to 120MB. Those extra bytes are always zeros. If there are no other request executed at the same time the download will always be successful. This issue is only reproducible on Windows, not on MacOS. There is no code relation (other than inside Qt) on those modules where concurrent requests are being made. Any help or ideas on this are warmly welcome, thanks. -- This email is intended solely for the person or entity to which it is addressed and may contain confidential and/or privileged information. If you are not the intended recipient and have received this email in error, please notify BroadSoft, Inc. immediately by replying to this message, and destroy all copies of this message, along with any attachment, prior to reading, distributing or copying it. Ce message confidentiel et les éventuelles pièces jointes sont à l’usage exclusif de son ou de sa destinataire. Il est interdit, pour toute autre personne, de le distribuer, d’en dévoiler le contenu ou de le reproduire. Si vous avez reçu cette communication par erreur, veuillez en informer immédiatement l’expéditeur par courrier électronique et détruire l’original de ce message ainsi que toute copie. -------------- next part -------------- An HTML attachment was scrubbed... URL: From chinander at gmail.com Tue Sep 26 21:03:07 2017 From: chinander at gmail.com (Mike Chinander) Date: Tue, 26 Sep 2017 14:03:07 -0500 Subject: [Interest] Issue with concurrent QNetworkRequests In-Reply-To: References: Message-ID: Why can't you use only one QNetworkAccessManager instance for your application? On Tue, Sep 26, 2017 at 1:55 PM, Jani Tykka wrote: > Hi, > > I'd ask if the following issues would ring anyones bells, before I start > doing more deeper investigation and create test application to reproduce > the issue. Basically the issue in our application is reproduced following > way. > > 1. Have 1 big download (https) 100MB (It uses own QNam instance) > - file is being written from the data got from QNetworkReply::readyRead > 2. During the download execute other requests (https) using different QNam > instance > > What is experienced that if during download other requests are executed > the data received from readyRead contains extra bytes in the middle of > chunked responses. It can be that the total file size received grows to > 120MB. Those extra bytes are always zeros. > > If there are no other request executed at the same time the download will > always be successful. This issue is only reproducible on Windows, not on > MacOS. There is no code relation (other than inside Qt) on those modules > where concurrent requests are being made. > > Any help or ideas on this are warmly welcome, thanks. > > > > This email is intended solely for the person or entity to which it is > addressed and may contain confidential and/or privileged information. If > you are not the intended recipient and have received this email in error, > please notify BroadSoft, Inc. immediately by replying to this message, and > destroy all copies of this message, along with any attachment, prior to > reading, distributing or copying it. > > Ce message confidentiel et les éventuelles pièces jointes sont à l’usage > exclusif de son ou de sa destinataire. Il est interdit, pour toute autre > personne, de le distribuer, d’en dévoiler le contenu ou de le reproduire. > Si vous avez reçu cette communication par erreur, veuillez en informer > immédiatement l’expéditeur par courrier électronique et détruire l’original > de ce message ainsi que toute copie. > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From thiago.macieira at intel.com Tue Sep 26 21:03:08 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Tue, 26 Sep 2017 12:03:08 -0700 Subject: [Interest] Issue with concurrent QNetworkRequests In-Reply-To: References: Message-ID: <3434634.edeP40DyBC@tjmaciei-mobl1> On terça-feira, 26 de setembro de 2017 11:55:35 PDT Jani Tykka wrote: > Hi, > > I'd ask if the following issues would ring anyones bells, before I start > doing more deeper investigation and create test application to reproduce > the issue. Basically the issue in our application is reproduced following > way. > > 1. Have 1 big download (https) 100MB (It uses own QNam instance) > - file is being written from the data got from QNetworkReply::readyRead > 2. During the download execute other requests (https) using different QNam > instance Why are you using a separate instance? > What is experienced that if during download other requests are executed the > data received from readyRead contains extra bytes in the middle of chunked > responses. It can be that the total file size received grows to 120MB. > Those extra bytes are always zeros. If they were the same instance, I could maybe see how there could be a problem. With separate instances, I'm doubting there could be a problem in Qt. Anyway, please make sure by packet sniffing that the transfer sizes are correct. You don't need to decode the transfer, you just need to compare the size of the transfer during a correct write to the size of the transfer during an incorrect one. > Any help or ideas on this are warmly welcome, thanks. The most likely scenario is that your slot is doing stupid things. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From jtykka at broadsoft.com Tue Sep 26 22:12:43 2017 From: jtykka at broadsoft.com (Jani Tykka) Date: Tue, 26 Sep 2017 20:12:43 +0000 Subject: [Interest] Issue with concurrent QNetworkRequests In-Reply-To: <3434634.edeP40DyBC@tjmaciei-mobl1> References: <3434634.edeP40DyBC@tjmaciei-mobl1> Message-ID: No special reason why I couldnt use 1 nam instance, I will try and see if it makes any difference. Thanks ti 26.9.2017 klo 22.04 Thiago Macieira kirjoitti: > On terça-feira, 26 de setembro de 2017 11:55:35 PDT Jani Tykka wrote: > > Hi, > > > > I'd ask if the following issues would ring anyones bells, before I start > > doing more deeper investigation and create test application to reproduce > > the issue. Basically the issue in our application is reproduced following > > way. > > > > 1. Have 1 big download (https) 100MB (It uses own QNam instance) > > - file is being written from the data got from QNetworkReply::readyRead > > 2. During the download execute other requests (https) using different > QNam > > instance > > Why are you using a separate instance? > > > What is experienced that if during download other requests are executed > the > > data received from readyRead contains extra bytes in the middle of > chunked > > responses. It can be that the total file size received grows to 120MB. > > Those extra bytes are always zeros. > > If they were the same instance, I could maybe see how there could be a > problem. With separate instances, I'm doubting there could be a problem in > Qt. > > Anyway, please make sure by packet sniffing that the transfer sizes are > correct. > You don't need to decode the transfer, you just need to compare the size of > the transfer during a correct write to the size of the transfer during an > incorrect one. > > > Any help or ideas on this are warmly welcome, thanks. > > The most likely scenario is that your slot is doing stupid things. > > -- > Thiago Macieira - thiago.macieira (AT) > https://urldefense.proofpoint.com/v2/url?u=http-3A__intel.com&d=DwIFAw&c=uiYkEnhlQB0H-gDwErXr4Q&r=wvIt_JuLp9axQq3NQWhY0UHgd4iE-wQTwUWS6jf3nRM&m=JZQbjY6e3maVFM0dGfDn-LDG5oLBQn8Gk3Gg2IkOilw&s=Lkmzg50z7jqtNsRMBPBWTS9T0uGAWzRK5LZ-BlNl-bQ&e= > Software Architect - Intel Open Source Technology Center > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > > https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.qt-2Dproject.org_mailman_listinfo_interest&d=DwIFAw&c=uiYkEnhlQB0H-gDwErXr4Q&r=wvIt_JuLp9axQq3NQWhY0UHgd4iE-wQTwUWS6jf3nRM&m=JZQbjY6e3maVFM0dGfDn-LDG5oLBQn8Gk3Gg2IkOilw&s=VT-NTaxmjcbiOPGiMabQ83hcoxb8AZBFwu5At5WMfZ4&e= > -- *Jani Tykkä* Development Manager | BroadSoft, Inc. | +358 44 596 0587 | jtykka at broadsoft.com -- This email is intended solely for the person or entity to which it is addressed and may contain confidential and/or privileged information. If you are not the intended recipient and have received this email in error, please notify BroadSoft, Inc. immediately by replying to this message, and destroy all copies of this message, along with any attachment, prior to reading, distributing or copying it. Ce message confidentiel et les éventuelles pièces jointes sont à l’usage exclusif de son ou de sa destinataire. Il est interdit, pour toute autre personne, de le distribuer, d’en dévoiler le contenu ou de le reproduire. Si vous avez reçu cette communication par erreur, veuillez en informer immédiatement l’expéditeur par courrier électronique et détruire l’original de ce message ainsi que toute copie. -------------- next part -------------- An HTML attachment was scrubbed... URL: From phil_lor at bigpond.com Tue Sep 26 23:23:14 2017 From: phil_lor at bigpond.com (Phil) Date: Wed, 27 Sep 2017 07:23:14 +1000 Subject: [Interest] Qdial ticks Message-ID: Thank you for reading this. I have created a form that includes a dial with Qt Designer. I have enabled ticks but the ticks aren't shown in my final program. However, they do show in Designer's preview window if I select Fusion or Windows style. So, it looks I need to enable a style to show the ticks. How do I do that? -- Regards, Phil From thiago.macieira at intel.com Wed Sep 27 00:13:39 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Tue, 26 Sep 2017 15:13:39 -0700 Subject: [Interest] Issue with concurrent QNetworkRequests In-Reply-To: References: <3434634.edeP40DyBC@tjmaciei-mobl1> Message-ID: <2702836.HYZIEjoZ3t@tjmaciei-mobl1> On terça-feira, 26 de setembro de 2017 13:12:43 PDT Jani Tykka wrote: > No special reason why I couldnt use 1 nam instance, I will try and see if > it makes any difference. Unlikely. The problem is most likely in your own code. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From thiago.macieira at intel.com Wed Sep 27 00:14:17 2017 From: thiago.macieira at intel.com (Thiago Macieira) Date: Tue, 26 Sep 2017 15:14:17 -0700 Subject: [Interest] Qdial ticks In-Reply-To: References: Message-ID: <6524569.feq0egM8Ps@tjmaciei-mobl1> On terça-feira, 26 de setembro de 2017 14:23:14 PDT Phil wrote: > Thank you for reading this. > > I have created a form that includes a dial with Qt Designer. I have > enabled ticks but the ticks aren't shown in my final program. However, > they do show in Designer's preview window if I select Fusion or Windows > style. > > So, it looks I need to enable a style to show the ticks. How do I do that? Don't do anything. The style chooses how to display the dial and it decided on your system that the ticks shouldn't be shown. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center From Sudhir.Sharma at mindtree.com Wed Sep 27 07:03:58 2017 From: Sudhir.Sharma at mindtree.com (Sudhir Sharma) Date: Wed, 27 Sep 2017 05:03:58 +0000 Subject: [Interest] WebEngineView crashing in 5.9.1 In-Reply-To: References: Message-ID: Thanks for response. Fixed the issue. Looks like there were some old dlls (from QT4) creating conflict. After cleanup it worked. Regards, Sudhir -----Original Message----- From: Kai Koehne [mailto:Kai.Koehne at qt.io] Sent: Tuesday, September 26, 2017 1:12 PM To: Sudhir Sharma ; interest at qt-project.org Subject: RE: WebEngineView crashing in 5.9.1 * You are receiving this mail from an external source * > -----Original Message----- > From: Interest > [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > On Behalf Of Sudhir Sharma > Sent: Tuesday, September 26, 2017 7:45 AM > To: interest at qt-project.org > Subject: [Interest] WebEngineView crashing in 5.9.1 > > Hi, > > I have compiled 5.9.1 in win 64, and using it in a machine which also > has previous version of QT (4.8) installed. > > On starting my application I am getting exception at following line, > > "webEngineView->setUrl(QUrl(QStringLiteral("about:blank")));" > > Whereas in a fresh machine which only has 5.9.1, it does no crash. > Any idea what would the conflict? No idea. But you can load your application into the Dependency Walker (depends.exe), and start it in profiling mode (F7). Make sure that you disable profiling of sub-processes ... Feel free to attach the output to a bug report then. Also, please try to reproduce with a minimal example (e.g. webenginewidgets/minimal) first. Regards Kai ________________________________ http://www.mindtree.com/email/disclaimer.html From Sudhir.Sharma at mindtree.com Wed Sep 27 11:32:58 2017 From: Sudhir.Sharma at mindtree.com (Sudhir Sharma) Date: Wed, 27 Sep 2017 09:32:58 +0000 Subject: [Interest] 5.9.1 WebEngineView issues with Windows 10 Message-ID: Hi, In a HTML in windows 10, observing issues with Dropdown. They appear transparent and not visible normally. Same thing works fine in other OS like windows 7 or 8.1 Any idea if anything specific required for windows 10? Regards, Sudhir ________________________________ http://www.mindtree.com/email/disclaimer.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From ola at silentwings.no Wed Sep 27 12:59:44 2017 From: ola at silentwings.no (=?UTF-8?Q?Ola_R=C3=B8er_Thorsen?=) Date: Wed, 27 Sep 2017 12:59:44 +0200 Subject: [Interest] How to get in touch with the Qt Company sales people? Message-ID: Hi, my employer has now tried to get in touch with sales at the Qt Company to get a license quote for the last two weeks. They tried using the contact form at the web page as well as the phone numbers listed on the website (HQ and Nordic). Nobody answers the phone, and the forms filled out appear to go unnoticed. If anyone on this list from the Qt Company can give me some valid contact information please send me an email. Best regards, Ola Røer Thorsen ola at silentwings.no -------------- next part -------------- An HTML attachment was scrubbed... URL: From andy.shaw at qt.io Wed Sep 27 13:06:29 2017 From: andy.shaw at qt.io (Andy Shaw) Date: Wed, 27 Sep 2017 11:06:29 +0000 Subject: [Interest] How to get in touch with the Qt Company sales people? In-Reply-To: References: Message-ID: Hi Ola, I’ll send a separate email to you and CC those concerned. Sorry about this! Regards, Andy Fra: Interest på vegne av Ola Røer Thorsen Dato: onsdag 27. september 2017 12.59 Til: "interest at qt-project.org" Emne: [Interest] How to get in touch with the Qt Company sales people? Hi, my employer has now tried to get in touch with sales at the Qt Company to get a license quote for the last two weeks. They tried using the contact form at the web page as well as the phone numbers listed on the website (HQ and Nordic). Nobody answers the phone, and the forms filled out appear to go unnoticed. If anyone on this list from the Qt Company can give me some valid contact information please send me an email. Best regards, Ola Røer Thorsen ola at silentwings.no -------------- next part -------------- An HTML attachment was scrubbed... URL: From nunosantos at imaginando.pt Wed Sep 27 13:09:34 2017 From: nunosantos at imaginando.pt (Nuno Santos) Date: Wed, 27 Sep 2017 12:09:34 +0100 Subject: [Interest] How to get in touch with the Qt Company sales people? In-Reply-To: References: Message-ID: Andy, I have experienced from the same problem several times. For two times I have tried to obtain a trial for Qt Device Creation through the website and I have never received a reply. Just upvoting Ola’s complain! :) Best, Nuno > On 27 Sep 2017, at 12:06, Andy Shaw wrote: > > Hi Ola, > > I’ll send a separate email to you and CC those concerned. Sorry about this! > > Regards, > Andy > > Fra: Interest på vegne av Ola Røer Thorsen > Dato: onsdag 27. september 2017 12.59 > Til: "interest at qt-project.org" > Emne: [Interest] How to get in touch with the Qt Company sales people? > > Hi, > > my employer has now tried to get in touch with sales at the Qt Company to get a license quote for the last two weeks. They tried using the contact form at the web page as well as the phone numbers listed on the website (HQ and Nordic). Nobody answers the phone, and the forms filled out appear to go unnoticed. > > If anyone on this list from the Qt Company can give me some valid contact information please send me an email. > > Best regards, > Ola Røer Thorsen > ola at silentwings.no > > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Wed Sep 27 13:23:42 2017 From: realnc at gmail.com (Nikos Chantziaras) Date: Wed, 27 Sep 2017 14:23:42 +0300 Subject: [Interest] Scrolling past end of documents Message-ID: Is there some quick trick to allow QTextBrowser to scroll past the beginning/end of the document? Basically have a pagefull of emptiness before and after the contents. From evilruff at gmail.com Wed Sep 27 15:13:14 2017 From: evilruff at gmail.com (Yuri Alexandrov) Date: Wed, 27 Sep 2017 16:13:14 +0300 Subject: [Interest] How to get in touch with the Qt Company sales people? In-Reply-To: References: Message-ID: <59cba414.14e9190a.c82db.bc59@mx.google.com> In that respect I can say that Qt department in Russia – St-Petersburg works very =) at least for last 3 years I am working with them Regards, Yuri From: Nuno Santos Sent: Wednesday, September 27, 2017 2:09 PM To: Andy Shaw Cc: interest at qt-project.org Subject: Re: [Interest] How to get in touch with the Qt Company sales people? Andy, I have experienced from the same problem several times.  For two times I have tried to obtain a trial for Qt Device Creation through the website and I have never received a reply.  Just upvoting Ola’s complain! :) Best, Nuno On 27 Sep 2017, at 12:06, Andy Shaw wrote: Hi Ola,   I’ll send a separate email to you and CC those concerned. Sorry about this!   Regards, Andy   Fra: Interest på vegne av Ola Røer Thorsen Dato: onsdag 27. september 2017 12.59 Til: "interest at qt-project.org" Emne: [Interest] How to get in touch with the Qt Company sales people?   Hi,     my employer has now tried to get in touch with sales at the Qt Company to get a license quote for the last two weeks. They tried using the contact form at the web page as well as the phone numbers listed on the website (HQ and Nordic). Nobody answers the phone, and the forms filled out appear to go unnoticed.    If anyone on this list from the Qt Company can give me some valid contact information please send me an email.    Best regards, Ola Røer Thorsen ola at silentwings.no     _______________________________________________ Interest mailing list Interest at qt-project.org http://lists.qt-project.org/mailman/listinfo/interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From jhihn at gmx.com Wed Sep 27 17:23:56 2017 From: jhihn at gmx.com (Jason H) Date: Wed, 27 Sep 2017 17:23:56 +0200 Subject: [Interest] How to get in touch with the Qt Company sales people? In-Reply-To: References: Message-ID: An HTML attachment was scrubbed... URL: From dmytro.haponov at gmail.com Wed Sep 27 21:51:45 2017 From: dmytro.haponov at gmail.com (Dmytro Haponov) Date: Wed, 27 Sep 2017 22:51:45 +0300 Subject: [Interest] qt example of virtual keyboard on virtualbox ubuntu&kubuntu Message-ID: example of virtual keyboard in qt5.7,5.9 has an issue on both ubuntu & kubuntu that are guests on virtualbox - keyboard is invisible, that is it's somewhere at the bottom of editLists that form the app built from example, and you can press it so that symbols are typed, but you cannot see the keyboard itself. And the app is hanging quite seriously on kubuntu in case of full screen. Latest version of virtualbox on windows is used. ubuntu 16.04, kubuntu can look at tomorrow (writing from home). qt virtual keyboard example works fine on ubuntu and windows that are not on virtual box (don't know about kubuntu). I managed to find the source of issue on virtualbox, it's in .pro file here: disable-desktop|android-embedded|!isEmpty(CROSS_COMPILE)|qnx { DEFINES += MAIN_QML=\\\"basic-b2qt.qml\\\" } else { DEFINES += MAIN_QML=\\\"Basic.qml\\\" } I have the second define always working. If I force the first one to be used, then keyboard works fine. Btw, host windows works on both files. So the question is - is there any lib lacking on my non-windows guests (disable-xcb flag does not seem to be set right) or is this a problem of virtual box? -------------- next part -------------- An HTML attachment was scrubbed... URL: From chgans at gmail.com Thu Sep 28 09:28:30 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Thu, 28 Sep 2017 20:28:30 +1300 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io Message-ID: Hi there, I would like to customise the Qt installer, not only by adding new "resources", but i would like to disable some/most "*.qt.io" resources. The reason for removing qt.io are multiples: - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch "*.qt.io" stuff, and i'm getting tired of it (blame NZ, unfortunately), and i think it is unacceptable for my potential users (co-workers). - I don't need all what qt.io has to offer for download, only a small sub-part of it - I would like to base our own internal SDK installer on "the Qt online installer" - I want to add my own "download/update" stuff - nobody i know of uses the Qt installer, yet they develop with Qt every day, having to deal with complex dependencies and awkward toolchain install/updates. - To my opinion, we need an SDK installer/updater and I think that hacking the Qt installer is an easy win. Basically I would like the installer to pull QtCreator and Qbs from qt.io, and all the rest is pulled from a server i can manage, this download server is tied with the CI output "repository". Is it currently doable? I haven't spend that much time on the installer "internals", so this is a bit of a blind question. PS: I have a Qt license, could this be useful on that topic? Chris From nicolas.krieger at medecom.fr Thu Sep 28 10:48:06 2017 From: nicolas.krieger at medecom.fr (Nicolas Krieger) Date: Thu, 28 Sep 2017 10:48:06 +0200 Subject: [Interest] How to embed pdf.js Message-ID: <1f67ed05-dab5-3bb3-ad71-85001bfc0d74@medecom.fr> Hi, I am trying to embed pdf.js in a qrc file. I can use pdf.js in a QWebView in using the complete filepath. Everything is working fine, I can use buttons and load a file using "runJavaScript" on a QWebEnginePage. I can also load a pdf file using pdf.js user interface. But when I try to embed pdf.js in a qrc file, I just see that the pdf viewer has been loaded, but I can't neither load a file nor use buttons. Nothing happens. I have attached the qrc file I use. I d'ont understand what is wrong with qrc method. Nicolas Krieger -------------- next part -------------- pdf.js/debugger.js pdf.js/l10n.js pdf.js/viewer.css pdf.js/viewer.html pdf.js/viewer.js pdf.js/cmaps/78-EUC-H.bcmap pdf.js/cmaps/78-EUC-V.bcmap pdf.js/cmaps/78-H.bcmap pdf.js/cmaps/78ms-RKSJ-H.bcmap pdf.js/cmaps/78ms-RKSJ-V.bcmap pdf.js/cmaps/78-RKSJ-H.bcmap pdf.js/cmaps/78-RKSJ-V.bcmap pdf.js/cmaps/78-V.bcmap pdf.js/cmaps/83pv-RKSJ-H.bcmap pdf.js/cmaps/90msp-RKSJ-H.bcmap pdf.js/cmaps/90msp-RKSJ-V.bcmap pdf.js/cmaps/90ms-RKSJ-H.bcmap pdf.js/cmaps/90ms-RKSJ-V.bcmap pdf.js/cmaps/90pv-RKSJ-H.bcmap pdf.js/cmaps/90pv-RKSJ-V.bcmap pdf.js/cmaps/Add-H.bcmap pdf.js/cmaps/Add-RKSJ-H.bcmap pdf.js/cmaps/Add-RKSJ-V.bcmap pdf.js/cmaps/Add-V.bcmap pdf.js/cmaps/Adobe-CNS1-0.bcmap pdf.js/cmaps/Adobe-CNS1-1.bcmap pdf.js/cmaps/Adobe-CNS1-2.bcmap pdf.js/cmaps/Adobe-CNS1-3.bcmap pdf.js/cmaps/Adobe-CNS1-4.bcmap pdf.js/cmaps/Adobe-CNS1-5.bcmap pdf.js/cmaps/Adobe-CNS1-6.bcmap pdf.js/cmaps/Adobe-CNS1-UCS2.bcmap pdf.js/cmaps/Adobe-GB1-0.bcmap pdf.js/cmaps/Adobe-GB1-1.bcmap pdf.js/cmaps/Adobe-GB1-2.bcmap pdf.js/cmaps/Adobe-GB1-3.bcmap pdf.js/cmaps/Adobe-GB1-4.bcmap pdf.js/cmaps/Adobe-GB1-5.bcmap pdf.js/cmaps/Adobe-GB1-UCS2.bcmap pdf.js/cmaps/Adobe-Japan1-0.bcmap pdf.js/cmaps/Adobe-Japan1-1.bcmap pdf.js/cmaps/Adobe-Japan1-2.bcmap pdf.js/cmaps/Adobe-Japan1-3.bcmap pdf.js/cmaps/Adobe-Japan1-4.bcmap pdf.js/cmaps/Adobe-Japan1-5.bcmap pdf.js/cmaps/Adobe-Japan1-6.bcmap pdf.js/cmaps/Adobe-Japan1-UCS2.bcmap pdf.js/cmaps/Adobe-Korea1-0.bcmap pdf.js/cmaps/Adobe-Korea1-1.bcmap pdf.js/cmaps/Adobe-Korea1-2.bcmap pdf.js/cmaps/Adobe-Korea1-UCS2.bcmap pdf.js/cmaps/B5-H.bcmap pdf.js/cmaps/B5pc-H.bcmap pdf.js/cmaps/B5pc-V.bcmap pdf.js/cmaps/B5-V.bcmap pdf.js/cmaps/CNS1-H.bcmap pdf.js/cmaps/CNS1-V.bcmap pdf.js/cmaps/CNS2-H.bcmap pdf.js/cmaps/CNS2-V.bcmap pdf.js/cmaps/CNS-EUC-H.bcmap pdf.js/cmaps/CNS-EUC-V.bcmap pdf.js/cmaps/ETen-B5-H.bcmap pdf.js/cmaps/ETen-B5-V.bcmap pdf.js/cmaps/ETenms-B5-H.bcmap pdf.js/cmaps/ETenms-B5-V.bcmap pdf.js/cmaps/ETHK-B5-H.bcmap pdf.js/cmaps/ETHK-B5-V.bcmap pdf.js/cmaps/EUC-H.bcmap pdf.js/cmaps/EUC-V.bcmap pdf.js/cmaps/Ext-H.bcmap pdf.js/cmaps/Ext-RKSJ-H.bcmap pdf.js/cmaps/Ext-RKSJ-V.bcmap pdf.js/cmaps/Ext-V.bcmap pdf.js/cmaps/GB-EUC-H.bcmap pdf.js/cmaps/GB-EUC-V.bcmap pdf.js/cmaps/GB-H.bcmap pdf.js/cmaps/GBK2K-H.bcmap pdf.js/cmaps/GBK2K-V.bcmap pdf.js/cmaps/GBK-EUC-H.bcmap pdf.js/cmaps/GBK-EUC-V.bcmap pdf.js/cmaps/GBKp-EUC-H.bcmap pdf.js/cmaps/GBKp-EUC-V.bcmap pdf.js/cmaps/GBpc-EUC-H.bcmap pdf.js/cmaps/GBpc-EUC-V.bcmap pdf.js/cmaps/GBT-EUC-H.bcmap pdf.js/cmaps/GBT-EUC-V.bcmap pdf.js/cmaps/GBT-H.bcmap pdf.js/cmaps/GBTpc-EUC-H.bcmap pdf.js/cmaps/GBTpc-EUC-V.bcmap pdf.js/cmaps/GBT-V.bcmap pdf.js/cmaps/GB-V.bcmap pdf.js/cmaps/H.bcmap pdf.js/cmaps/Hankaku.bcmap pdf.js/cmaps/Hiragana.bcmap pdf.js/cmaps/HKdla-B5-H.bcmap pdf.js/cmaps/HKdla-B5-V.bcmap pdf.js/cmaps/HKdlb-B5-H.bcmap pdf.js/cmaps/HKdlb-B5-V.bcmap pdf.js/cmaps/HKgccs-B5-H.bcmap pdf.js/cmaps/HKgccs-B5-V.bcmap pdf.js/cmaps/HKm314-B5-H.bcmap pdf.js/cmaps/HKm314-B5-V.bcmap pdf.js/cmaps/HKm471-B5-H.bcmap pdf.js/cmaps/HKm471-B5-V.bcmap pdf.js/cmaps/HKscs-B5-H.bcmap pdf.js/cmaps/HKscs-B5-V.bcmap pdf.js/cmaps/Katakana.bcmap pdf.js/cmaps/KSC-EUC-H.bcmap pdf.js/cmaps/KSC-EUC-V.bcmap pdf.js/cmaps/KSC-H.bcmap pdf.js/cmaps/KSC-Johab-H.bcmap pdf.js/cmaps/KSC-Johab-V.bcmap pdf.js/cmaps/KSCms-UHC-H.bcmap pdf.js/cmaps/KSCms-UHC-HW-H.bcmap pdf.js/cmaps/KSCms-UHC-HW-V.bcmap pdf.js/cmaps/KSCms-UHC-V.bcmap pdf.js/cmaps/KSCpc-EUC-H.bcmap pdf.js/cmaps/KSCpc-EUC-V.bcmap pdf.js/cmaps/KSC-V.bcmap pdf.js/cmaps/LICENSE pdf.js/cmaps/NWP-H.bcmap pdf.js/cmaps/NWP-V.bcmap pdf.js/cmaps/RKSJ-H.bcmap pdf.js/cmaps/RKSJ-V.bcmap pdf.js/cmaps/Roman.bcmap pdf.js/cmaps/UniCNS-UCS2-H.bcmap pdf.js/cmaps/UniCNS-UCS2-V.bcmap pdf.js/cmaps/UniCNS-UTF8-H.bcmap pdf.js/cmaps/UniCNS-UTF8-V.bcmap pdf.js/cmaps/UniCNS-UTF16-H.bcmap pdf.js/cmaps/UniCNS-UTF16-V.bcmap pdf.js/cmaps/UniCNS-UTF32-H.bcmap pdf.js/cmaps/UniCNS-UTF32-V.bcmap pdf.js/cmaps/UniGB-UCS2-H.bcmap pdf.js/cmaps/UniGB-UCS2-V.bcmap pdf.js/cmaps/UniGB-UTF8-H.bcmap pdf.js/cmaps/UniGB-UTF8-V.bcmap pdf.js/cmaps/UniGB-UTF16-H.bcmap pdf.js/cmaps/UniGB-UTF16-V.bcmap pdf.js/cmaps/UniGB-UTF32-H.bcmap pdf.js/cmaps/UniGB-UTF32-V.bcmap pdf.js/cmaps/UniJIS2004-UTF8-H.bcmap pdf.js/cmaps/UniJIS2004-UTF8-V.bcmap pdf.js/cmaps/UniJIS2004-UTF16-H.bcmap pdf.js/cmaps/UniJIS2004-UTF16-V.bcmap pdf.js/cmaps/UniJIS2004-UTF32-H.bcmap pdf.js/cmaps/UniJIS2004-UTF32-V.bcmap pdf.js/cmaps/UniJISPro-UCS2-HW-V.bcmap pdf.js/cmaps/UniJISPro-UCS2-V.bcmap pdf.js/cmaps/UniJISPro-UTF8-V.bcmap pdf.js/cmaps/UniJIS-UCS2-H.bcmap pdf.js/cmaps/UniJIS-UCS2-HW-H.bcmap pdf.js/cmaps/UniJIS-UCS2-HW-V.bcmap pdf.js/cmaps/UniJIS-UCS2-V.bcmap pdf.js/cmaps/UniJIS-UTF8-H.bcmap pdf.js/cmaps/UniJIS-UTF8-V.bcmap pdf.js/cmaps/UniJIS-UTF16-H.bcmap pdf.js/cmaps/UniJIS-UTF16-V.bcmap pdf.js/cmaps/UniJIS-UTF32-H.bcmap pdf.js/cmaps/UniJIS-UTF32-V.bcmap pdf.js/cmaps/UniJISX0213-UTF32-H.bcmap pdf.js/cmaps/UniJISX0213-UTF32-V.bcmap pdf.js/cmaps/UniJISX02132004-UTF32-H.bcmap pdf.js/cmaps/UniJISX02132004-UTF32-V.bcmap pdf.js/cmaps/UniKS-UCS2-H.bcmap pdf.js/cmaps/UniKS-UCS2-V.bcmap pdf.js/cmaps/UniKS-UTF8-H.bcmap pdf.js/cmaps/UniKS-UTF8-V.bcmap pdf.js/cmaps/UniKS-UTF16-H.bcmap pdf.js/cmaps/UniKS-UTF16-V.bcmap pdf.js/cmaps/UniKS-UTF32-H.bcmap pdf.js/cmaps/UniKS-UTF32-V.bcmap pdf.js/cmaps/V.bcmap pdf.js/cmaps/WP-Symbol.bcmap pdf.js/images/annotation-check.svg pdf.js/images/annotation-comment.svg pdf.js/images/annotation-help.svg pdf.js/images/annotation-insert.svg pdf.js/images/annotation-key.svg pdf.js/images/annotation-newparagraph.svg pdf.js/images/annotation-noicon.svg pdf.js/images/annotation-note.svg pdf.js/images/annotation-paragraph.svg pdf.js/images/findbarButton-next.png pdf.js/images/findbarButton-next at 2x.png pdf.js/images/findbarButton-next-rtl.png pdf.js/images/findbarButton-next-rtl at 2x.png pdf.js/images/findbarButton-previous.png pdf.js/images/findbarButton-previous at 2x.png pdf.js/images/findbarButton-previous-rtl.png pdf.js/images/findbarButton-previous-rtl at 2x.png pdf.js/images/grab.cur pdf.js/images/grabbing.cur pdf.js/images/loading-icon.gif pdf.js/images/loading-small.png pdf.js/images/loading-small at 2x.png pdf.js/images/secondaryToolbarButton-documentProperties.png pdf.js/images/secondaryToolbarButton-documentProperties at 2x.png pdf.js/images/secondaryToolbarButton-firstPage.png pdf.js/images/secondaryToolbarButton-firstPage at 2x.png pdf.js/images/secondaryToolbarButton-handTool.png pdf.js/images/secondaryToolbarButton-handTool at 2x.png pdf.js/images/secondaryToolbarButton-lastPage.png pdf.js/images/secondaryToolbarButton-lastPage at 2x.png pdf.js/images/secondaryToolbarButton-rotateCcw.png pdf.js/images/secondaryToolbarButton-rotateCcw at 2x.png pdf.js/images/secondaryToolbarButton-rotateCw.png pdf.js/images/secondaryToolbarButton-rotateCw at 2x.png pdf.js/images/shadow.png pdf.js/images/texture.png pdf.js/images/toolbarButton-bookmark.png pdf.js/images/toolbarButton-bookmark at 2x.png pdf.js/images/toolbarButton-download.png pdf.js/images/toolbarButton-download at 2x.png pdf.js/images/toolbarButton-menuArrows.png pdf.js/images/toolbarButton-menuArrows at 2x.png pdf.js/images/toolbarButton-openFile.png pdf.js/images/toolbarButton-openFile at 2x.png pdf.js/images/toolbarButton-pageDown.png pdf.js/images/toolbarButton-pageDown at 2x.png pdf.js/images/toolbarButton-pageDown-rtl.png pdf.js/images/toolbarButton-pageDown-rtl at 2x.png pdf.js/images/toolbarButton-pageUp.png pdf.js/images/toolbarButton-pageUp at 2x.png pdf.js/images/toolbarButton-pageUp-rtl.png pdf.js/images/toolbarButton-pageUp-rtl at 2x.png pdf.js/images/toolbarButton-presentationMode.png pdf.js/images/toolbarButton-presentationMode at 2x.png pdf.js/images/toolbarButton-print.png pdf.js/images/toolbarButton-print at 2x.png pdf.js/images/toolbarButton-search.png pdf.js/images/toolbarButton-search at 2x.png pdf.js/images/toolbarButton-secondaryToolbarToggle.png pdf.js/images/toolbarButton-secondaryToolbarToggle at 2x.png pdf.js/images/toolbarButton-secondaryToolbarToggle-rtl.png pdf.js/images/toolbarButton-secondaryToolbarToggle-rtl at 2x.png pdf.js/images/toolbarButton-sidebarToggle.png pdf.js/images/toolbarButton-sidebarToggle at 2x.png pdf.js/images/toolbarButton-sidebarToggle-rtl.png pdf.js/images/toolbarButton-sidebarToggle-rtl at 2x.png pdf.js/images/toolbarButton-viewAttachments.png pdf.js/images/toolbarButton-viewAttachments at 2x.png pdf.js/images/toolbarButton-viewOutline.png pdf.js/images/toolbarButton-viewOutline at 2x.png pdf.js/images/toolbarButton-viewOutline-rtl.png pdf.js/images/toolbarButton-viewOutline-rtl at 2x.png pdf.js/images/toolbarButton-viewThumbnail.png pdf.js/images/toolbarButton-viewThumbnail at 2x.png pdf.js/images/toolbarButton-zoomIn.png pdf.js/images/toolbarButton-zoomIn at 2x.png pdf.js/images/toolbarButton-zoomOut.png pdf.js/images/toolbarButton-zoomOut at 2x.png pdf.js/images/treeitem-collapsed.png pdf.js/images/treeitem-collapsed at 2x.png pdf.js/images/treeitem-collapsed-rtl.png pdf.js/images/treeitem-collapsed-rtl at 2x.png pdf.js/images/treeitem-expanded.png pdf.js/images/treeitem-expanded at 2x.png pdf.js/locale/locale.properties pdf.js/locale/fr/viewer.properties pdf.js/locale/en-GB/viewer.properties pdf.js/locale/en-US/viewer.properties From nicolas.krieger at medecom.fr Thu Sep 28 11:13:54 2017 From: nicolas.krieger at medecom.fr (Nicolas Krieger) Date: Thu, 28 Sep 2017 11:13:54 +0200 Subject: [Interest] How to embed pdf.js In-Reply-To: <1f67ed05-dab5-3bb3-ad71-85001bfc0d74@medecom.fr> References: <1f67ed05-dab5-3bb3-ad71-85001bfc0d74@medecom.fr> Message-ID: <3870df1b-3c8f-55d9-eba2-1af168f24415@medecom.fr> Oups, I deleted the "build" directory of pdf.js. This was the mistake. Everything is working fine now. Le 28/09/2017 à 10:48, Nicolas Krieger a écrit : > Hi, > > I am trying to embed pdf.js in a qrc file. > > I can use pdf.js in a QWebView in using the complete filepath. > Everything is working fine, I can use buttons and load a file using > "runJavaScript" on a QWebEnginePage. I can also load a pdf file using > pdf.js user interface. > > But when I try to embed pdf.js in a qrc file, I just see that the pdf > viewer has been loaded, but I can't neither load a file nor use > buttons. Nothing happens. > > I have attached the qrc file I use. > > I d'ont understand what is wrong with qrc method. > > Nicolas Krieger > > > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From artem.sidyakin at qt.io Thu Sep 28 11:34:56 2017 From: artem.sidyakin at qt.io (Artem Sidyakin) Date: Thu, 28 Sep 2017 09:34:56 +0000 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: Message-ID: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> > I would like to disable some/most “*.qt.io" resources You can uncheck “unwanted” repositories in settings, and then installer will fetch only the stuff you need. For example, that’s how you get Qt Creator only: 1) https://i.imgur.com/EC9c6S5.png 2) https://i.imgur.com/lteXbdm.png > and all the rest is pulled from a server i can manage On the first screenshot (https://i.imgur.com/EC9c6S5.png) you can see that it is possible to add your own repositories. There is also a documentation for Qt Installer Framework that might help: http://doc.qt.io/qtinstallerframework/ifw-online-installers.html --- Artem Sidyakin > On 28 Sep 2017, at 09:28, Christian Gagneraud wrote: > > Hi there, > > I would like to customise the Qt installer, not only by adding new > "resources", but i would like to disable some/most "*.qt.io" > resources. > > The reason for removing qt.io are multiples: > - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch > "*.qt.io" stuff, and i'm getting tired of it (blame NZ, > unfortunately), and i think it is unacceptable for my potential users > (co-workers). > - I don't need all what qt.io has to offer for download, only a small > sub-part of it > - I would like to base our own internal SDK installer on "the Qt > online installer" > - I want to add my own "download/update" stuff > - nobody i know of uses the Qt installer, yet they develop with Qt > every day, having to deal with complex dependencies and awkward > toolchain install/updates. > - To my opinion, we need an SDK installer/updater and I think that > hacking the Qt installer is an easy win. > > Basically I would like the installer to pull QtCreator and Qbs from > qt.io, and all the rest is pulled from a server i can manage, this > download server is tied with the CI output "repository". > > Is it currently doable? I haven't spend that much time on the > installer "internals", so this is a bit of a blind question. > > PS: I have a Qt license, could this be useful on that topic? > > Chris > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest From Kai.Koehne at qt.io Thu Sep 28 11:40:35 2017 From: Kai.Koehne at qt.io (Kai Koehne) Date: Thu, 28 Sep 2017 09:40:35 +0000 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: Message-ID: > -----Original Message----- > From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > On Behalf Of Christian Gagneraud > Sent: Thursday, September 28, 2017 9:29 AM > To: interest at qt-project.org > Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io > > Hi there, > > I would like to customise the Qt installer, not only by adding new > "resources", but i would like to disable some/most "*.qt.io" > resources. > > The reason for removing qt.io are multiples: > - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch "*.qt.io" > stuff, and i'm getting tired of it (blame NZ, unfortunately), and i think it is > unacceptable for my potential users (co-workers). > [...] > PS: I have a Qt license, could this be useful on that topic? Please also raise this with Qt Support then. Maybe we just have to add another mirror in the region then. Regards Kai From mitch.curtis at qt.io Thu Sep 28 12:02:18 2017 From: mitch.curtis at qt.io (Mitch Curtis) Date: Thu, 28 Sep 2017 10:02:18 +0000 Subject: [Interest] qt example of virtual keyboard on virtualbox ubuntu&kubuntu In-Reply-To: References: Message-ID: It could be https://bugreports.qt.io/browse/QTBUG-54088? The issue isn’t in the .pro; the integrated InputPanel approach (or ”Applictaion integration”: http://doc.qt.io/qt-5/qtvirtualkeyboard-deployment-guide.html#integration-method) will pretty much always work, as it simply puts the InputPanel in the same scene as the rest of your application. The Desktop integration is more tricky, as you can see in QTBUG-54088. From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org] On Behalf Of Dmytro Haponov Sent: Wednesday, 27 September 2017 9:52 PM To: interest at qt-project.org Subject: [Interest] qt example of virtual keyboard on virtualbox ubuntu&kubuntu example of virtual keyboard in qt5.7,5.9 has an issue on both ubuntu & kubuntu that are guests on virtualbox - keyboard is invisible, that is it's somewhere at the bottom of editLists that form the app built from example, and you can press it so that symbols are typed, but you cannot see the keyboard itself. And the app is hanging quite seriously on kubuntu in case of full screen. Latest version of virtualbox on windows is used. ubuntu 16.04, kubuntu can look at tomorrow (writing from home). qt virtual keyboard example works fine on ubuntu and windows that are not on virtual box (don't know about kubuntu). I managed to find the source of issue on virtualbox, it's in .pro file here: disable-desktop|android-embedded|!isEmpty(CROSS_COMPILE)|qnx { DEFINES += MAIN_QML=\\\"basic-b2qt.qml\\\" } else { DEFINES += MAIN_QML=\\\"Basic.qml\\\" } I have the second define always working. If I force the first one to be used, then keyboard works fine. Btw, host windows works on both files. So the question is - is there any lib lacking on my non-windows guests (disable-xcb flag does not seem to be set right) or is this a problem of virtual box? -------------- next part -------------- An HTML attachment was scrubbed... URL: From mnelson at sutron.com Thu Sep 28 16:36:41 2017 From: mnelson at sutron.com (Michael R Nelson) Date: Thu, 28 Sep 2017 10:36:41 -0400 Subject: [Interest] Crash dump related to QtCharts::QDateTimeAxis destructor Message-ID: <32C8071935AE4F4DB83A391C5D7ABA730ABE4565@mailsvr> I'm hoping someone can give me a clue as to what this crash dump is telling me. Problem started after moving to Qt 5.9.1 snapshot, on macOS Sierra. In my source, I have a QML ChartView with a DateTimeAxis and ValueAxis, and do series manipulation in C++. Thread 0 Crashed:: Dispatch queue: com.apple.main-thread 0 org.qt-project.QtQml 0x000000010bfe1a4c QQmlContextData::ContextGuard::objectDestroyed(QObject*) + 44 1 org.qt-project.QtQml 0x000000010bfc0a09 QQmlData::destroyed(QObject*) + 361 2 org.qt-project.QtCore 0x000000010caccaaf QObject::~QObject() + 303 3 libqtchartsqml2.dylib 0x000000011650fc00 QQmlPrivate::QQmlElement::~QQmlElement() + 32 4 org.qt-project.QtCore 0x000000010cacddd7 QObject::event(QEvent*) + 823 5 org.qt-project.QtWidgets 0x000000010cf82bad QApplicationPrivate::notify_helper(QObject*, QEvent*) + 269 6 org.qt-project.QtWidgets 0x000000010cf8547c QApplication::notify(QObject*, QEvent*) + 5852 7 org.qt-project.QtCore 0x000000010caa3a24 QCoreApplication::notifyInternal2(QObject*, QEvent*) + 164 8 org.qt-project.QtCore 0x000000010caa4770 QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) + 560 9 libqcocoa.dylib 0x00000001100ea19e QCocoaEventDispatcherPrivate::processPostedEvents() + 190 10 libqcocoa.dylib 0x00000001100eaa61 QCocoaEventDispatcherPrivate::postedEventsSourceCallback(void*) + 33 11 com.apple.CoreFoundation 0x00007fffb1ace3e1 __CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION__ + 17 12 com.apple.CoreFoundation 0x00007fffb1aaf5d7 __CFRunLoopDoSources0 + 423 13 com.apple.CoreFoundation 0x00007fffb1aaeb46 __CFRunLoopRun + 934 ... Thanks for any insight you can offer. Mike Please be advised that this email may contain confidential information. If you are not the intended recipient, please notify us by email by replying to the sender and delete this message. The sender disclaims that the content of this email constitutes an offer to enter into, or the acceptance of, any agreement; provided that the foregoing does not invalidate the binding effect of any digital or other electronic reproduction of a manual signature that is included in any attachment. From michael.corcoran at outlook.com Thu Sep 28 23:30:19 2017 From: michael.corcoran at outlook.com (Michael Corcoran) Date: Thu, 28 Sep 2017 21:30:19 +0000 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: , Message-ID: Hi Christian, I wonder if this is something you could achieve with a script that modifies behaviour of the official Qt installer? I have used it to automatically select desired components and install location before (https://github.com/d-ronin/dRonin/blob/next/make/scripts/qt-install.qs and running Qt installer with an extra flag: --script $(ROOT_DIR)/make/scripts/qt-install.qs, apologies in advance for lack of comments). Docs are reasonably good http://doc.qt.io/qtinstallerframework/. You'd have to dig through docs to see if you can change the source URLs etc. Out of curiosity, which mirror does the Qt installer select for you? For me it chooses one in Japan, which is quite a long way away by internet thus very slow (e.g. ~100 kiB/s). I've often preferred to download Qt offline installers from a US West-coast mirror instead (Southern Cross Cable is the only big cable out of NZ, and goes direct to Australia and California). Kai, Rather than a new mirror it might be sufficient to just alter the automatic selection to use California for New Zealand clients (I'm unsure how it works/how feasible that is)? Regards, Mike ________________________________________ From: Interest on behalf of Kai Koehne Sent: Thursday, 28 September 2017 10:40 p.m. To: Christian Gagneraud; interest at qt-project.org Subject: Re: [Interest] strip: Customise Qt intaller, remove *.qt.io > -----Original Message----- > From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > On Behalf Of Christian Gagneraud > Sent: Thursday, September 28, 2017 9:29 AM > To: interest at qt-project.org > Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io > > Hi there, > > I would like to customise the Qt installer, not only by adding new > "resources", but i would like to disable some/most "*.qt.io" > resources. > > The reason for removing qt.io are multiples: > - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch "*.qt.io" > stuff, and i'm getting tired of it (blame NZ, unfortunately), and i think it is > unacceptable for my potential users (co-workers). > [...] > PS: I have a Qt license, could this be useful on that topic? Please also raise this with Qt Support then. Maybe we just have to add another mirror in the region then. Regards Kai _______________________________________________ Interest mailing list Interest at qt-project.org http://lists.qt-project.org/mailman/listinfo/interest From Jake.Petroules at qt.io Thu Sep 28 23:57:45 2017 From: Jake.Petroules at qt.io (Jake Petroules) Date: Thu, 28 Sep 2017 21:57:45 +0000 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: Message-ID: <4F2169D4-9E0B-4AC4-99A6-78F538E62C76@qt.io> > On Sep 28, 2017, at 2:30 PM, Michael Corcoran wrote: > > Hi Christian, > > I wonder if this is something you could achieve with a script that modifies behaviour of the official Qt installer? I have used it to automatically select desired components and install location before (https://github.com/d-ronin/dRonin/blob/next/make/scripts/qt-install.qs and running Qt installer with an extra flag: --script $(ROOT_DIR)/make/scripts/qt-install.qs, apologies in advance for lack of comments). Docs are reasonably good http://doc.qt.io/qtinstallerframework/. You'd have to dig through docs to see if you can change the source URLs etc. > > Out of curiosity, which mirror does the Qt installer select for you? For me it chooses one in Japan, which is quite a long way away by internet thus very slow (e.g. ~100 kiB/s). I've often preferred to download Qt offline installers from a US West-coast mirror instead (Southern Cross Cable is the only big cable out of NZ, and goes direct to Australia and California). We recently added a Berkeley, CA (US West Coast) mirror thanks to the University of California, Berkeley: http://mirrors.ocf.berkeley.edu/qt/ Connection speed should be extremely fast (20 Mbps for me from San Francisco, which is even faster than local AWS CDNs), does that work any better for you? The Japanese mirrors are between 2.5 and 5 MB for me which is not terrible despite the distance, and CA is about as far from Japan as NZ is, so maybe Berkeley can give you an acceptable speed. If not, I'd suggest reaching out to a large AU or NZ university which has an existing Open Source software mirroring site (as many large universities do) and ask them to mirror. In my experience they are often more than happy to. > > Kai, > Rather than a new mirror it might be sufficient to just alter the automatic selection to use California for New Zealand clients (I'm unsure how it works/how feasible that is)? We use MaxMind GeoIP commercial database for IP address geolocation, and it selects the best mirror based on that. Not sure of the particulars beyond that or how easy/possible it is to "override" things like that. It's not always a very reliable system in general, since I for example get a mirror on the US East Coast despite Berkeley being a 20-minute drive from me, because my IP has the wrong location in that DB. However, if we know that NZ is only physically connected to Australia and California then it might indeed make sense to hardcode the Berkeley mirror for NZ users. > > Regards, > Mike > ________________________________________ > From: Interest on behalf of Kai Koehne > Sent: Thursday, 28 September 2017 10:40 p.m. > To: Christian Gagneraud; interest at qt-project.org > Subject: Re: [Interest] strip: Customise Qt intaller, remove *.qt.io > >> -----Original Message----- >> From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] >> On Behalf Of Christian Gagneraud >> Sent: Thursday, September 28, 2017 9:29 AM >> To: interest at qt-project.org >> Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io >> >> Hi there, >> >> I would like to customise the Qt installer, not only by adding new >> "resources", but i would like to disable some/most "*.qt.io" >> resources. >> >> The reason for removing qt.io are multiples: >> - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch "*.qt.io" >> stuff, and i'm getting tired of it (blame NZ, unfortunately), and i think it is >> unacceptable for my potential users (co-workers). >> [...] >> PS: I have a Qt license, could this be useful on that topic? > > Please also raise this with Qt Support then. Maybe we just have to add another mirror in the region then. > > Regards > > Kai > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Jake Petroules - jake.petroules at qt.io The Qt Company - Silicon Valley Qbs build tool evangelist - qbs.io From chgans at gmail.com Fri Sep 29 00:01:09 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 29 Sep 2017 11:01:09 +1300 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> References: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> Message-ID: On 28 September 2017 at 22:34, Artem Sidyakin wrote: >> I would like to disable some/most “*.qt.io" resources > > You can uncheck “unwanted” repositories in settings, and then installer will fetch only the stuff you need. > > For example, that’s how you get Qt Creator only: > > 1) https://i.imgur.com/EC9c6S5.png > 2) https://i.imgur.com/lteXbdm.png Thanks for sharing, appreciated. Still to get there i have to go first through "Preparing meta information download..." which takes a couple of minutes, I don't know why, but network speed sounds like a plausible cause. Chris From chgans at gmail.com Fri Sep 29 00:40:40 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 29 Sep 2017 11:40:40 +1300 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: <4F2169D4-9E0B-4AC4-99A6-78F538E62C76@qt.io> References: <4F2169D4-9E0B-4AC4-99A6-78F538E62C76@qt.io> Message-ID: On 29 September 2017 at 10:57, Jake Petroules wrote: >> Kai, >> Rather than a new mirror it might be sufficient to just alter the automatic selection to use California for New Zealand clients (I'm unsure how it works/how feasible that is)? > > We use MaxMind GeoIP commercial database for IP address geolocation, and it selects the best mirror based on that. Not sure of the particulars beyond that or how easy/possible it is to "override" things like that. > > It's not always a very reliable system in general, since I for example get a mirror on the US East Coast despite Berkeley being a 20-minute drive from me, because my IP has the wrong location in that DB. > > However, if we know that NZ is only physically connected to Australia and California then it might indeed make sense to hardcode the Berkeley mirror for NZ users. Here are the 2 sets of cables involved b/w NZ, US and Japan: https://en.wikipedia.org/wiki/Australia%E2%80%93Japan_Cable https://en.wikipedia.org/wiki/Southern_Cross_Cable Maybe we should ask the NSA to remove their MitM boxes... ;) I will try to see if overriding the mirror helps. Right now, the MaintenaceTool is downloading at less than 10KiB/s!?! and sometimes get stalled... Then i did a google speed test in parallel of the MaintenaceTool and I got 5 MiB/s (mega-bytes) both way to/from Google's Auckland servers. So my end point (Auckland office) is not the bottle-neck. A speed test with - DigitalOcean San Franciso gives me 3.59 Mbps (mega-bits) - DigitalOcean New York: 4.56 Mbps - DigitalOcean Singapour: 5.88 Mbps - Australia: 18.66 Mbps - Taiwan: .. stuck... - Japan: .. stuck... oh wait! na, stuck again - NZ DSL: 61.08 Mbps - 4 Mbps to London, Netherland These bits-per-second figure are not accurate, but helps painting the picture I think. It might be a good idea to contact the guys who mirrors Linux distro and co in new Zeland (ftp://ucmirror.canterbury.ac.nz comes to my mind) Chris > >> >> Regards, >> Mike >> ________________________________________ >> From: Interest on behalf of Kai Koehne >> Sent: Thursday, 28 September 2017 10:40 p.m. >> To: Christian Gagneraud; interest at qt-project.org >> Subject: Re: [Interest] strip: Customise Qt intaller, remove *.qt.io >> >>> -----Original Message----- >>> From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] >>> On Behalf Of Christian Gagneraud >>> Sent: Thursday, September 28, 2017 9:29 AM >>> To: interest at qt-project.org >>> Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io >>> >>> Hi there, >>> >>> I would like to customise the Qt installer, not only by adding new >>> "resources", but i would like to disable some/most "*.qt.io" >>> resources. >>> >>> The reason for removing qt.io are multiples: >>> - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch "*.qt.io" >>> stuff, and i'm getting tired of it (blame NZ, unfortunately), and i think it is >>> unacceptable for my potential users (co-workers). >>> [...] >>> PS: I have a Qt license, could this be useful on that topic? >> >> Please also raise this with Qt Support then. Maybe we just have to add another mirror in the region then. >> >> Regards >> >> Kai >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest >> _______________________________________________ >> Interest mailing list >> Interest at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > > -- > Jake Petroules - jake.petroules at qt.io > The Qt Company - Silicon Valley > Qbs build tool evangelist - qbs.io > From chgans at gmail.com Fri Sep 29 00:48:12 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 29 Sep 2017 11:48:12 +1300 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: Message-ID: On 29 September 2017 at 10:30, Michael Corcoran wrote: > Hi Christian, > > I wonder if this is something you could achieve with a script that modifies behaviour of the official Qt installer? I have used it to automatically select desired components and install location before (https://github.com/d-ronin/dRonin/blob/next/make/scripts/qt-install.qs and running Qt installer with an extra flag: --script $(ROOT_DIR)/make/scripts/qt-install.qs, apologies in advance for lack of comments). Docs are reasonably good http://doc.qt.io/qtinstallerframework/. You'd have to dig through docs to see if you can change the source URLs etc. Based on the doc and what you and Artem are saying, this is not exactly what i would like to achieve, thought it is interesting information. I don't wan't to automate the install process (entering login, clicking next), i wan't the installer to come with my own set of (selection of) repositories. That's why i'm thinking about repackaging the Qt Online installer. So that the starting point for our development machines is this custom online installer. Chris From szehowe.koh at gmail.com Fri Sep 29 01:29:29 2017 From: szehowe.koh at gmail.com (Sze Howe Koh) Date: Fri, 29 Sep 2017 07:29:29 +0800 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: Message-ID: On 28 September 2017 at 17:40, Kai Koehne wrote: > > > -----Original Message----- > > From: Interest [mailto:interest-bounces+kai.koehne=qt.io at qt-project.org] > > On Behalf Of Christian Gagneraud > > Sent: Thursday, September 28, 2017 9:29 AM > > To: interest at qt-project.org > > Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io > > > > Hi there, > > > > I would like to customise the Qt installer, not only by adding new > > "resources", but i would like to disable some/most "*.qt.io" > > resources. > > > > The reason for removing qt.io are multiples: > > - From New Zealand, the Qt installer spends 5 to 10 minutes to fetch "*.qt.io" > > stuff, and i'm getting tired of it (blame NZ, unfortunately), and i think it is > > unacceptable for my potential users (co-workers). > > [...] > > PS: I have a Qt license, could this be useful on that topic? > > Please also raise this with Qt Support then. Maybe we just have to add another mirror in the region then. It's not that we're too far away from the available mirrors. Rather, our link to our default mirror is slow. If I choose a different mirror (e.g. by using the Qt SDK Repo Chooser [1]), speeds are fine. --------------- QTIFW-441 [2] is reported as Done in v3.0.1. I was under the impression that this feature would allow end-users to manually select a mirror in the installer (see the comments under QTIFW-441). However, after upgrading my Qt Online Installer, I still don’t see a way to choose a mirror. Could you please confirm if the ability to select a mirror has been integrated? > Regards > > Kai Regards, Sze-Howe [1] https://forum.qt.io/topic/43349/slow-downloads-with-the-online-installer-try-this-tool [2] https://bugreports.qt.io/browse/QTIFW-441 From szehowe.koh at gmail.com Fri Sep 29 01:44:02 2017 From: szehowe.koh at gmail.com (Sze Howe Koh) Date: Fri, 29 Sep 2017 07:44:02 +0800 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> Message-ID: On 29 September 2017 at 06:01, Christian Gagneraud wrote: > On 28 September 2017 at 22:34, Artem Sidyakin wrote: >>> I would like to disable some/most “*.qt.io" resources >> >> You can uncheck “unwanted” repositories in settings, and then installer will fetch only the stuff you need. >> >> For example, that’s how you get Qt Creator only: >> >> 1) https://i.imgur.com/EC9c6S5.png >> 2) https://i.imgur.com/lteXbdm.png > > Thanks for sharing, appreciated. > > Still to get there i have to go first through "Preparing meta > information download..." which takes a couple of minutes, I don't know > why, but network speed sounds like a plausible cause. > > Chris I just did a quick test on a PC without Qt installed. I'm in Australia, and my current default mirror is ftp.jaist.ac.jp During the "Preparing meta information download..." phase, Resource Monitor told me that the data is crawling in at 10 Bytes per second (yes, bytes), and it took a few minutes to get the meta info. However, once I get past that stage, the actual Qt Creator files come in at ~300 KiB per second. Resource Monitor reports that both sets of downloads came from the same mirror, so it's not just a connection issue. I wonder what's the reason for this huge discrepancy? Is it because the meta information comes in as lots of tiny files? Regards, Sze-Howe From amosbird at gmail.com Fri Sep 29 02:55:58 2017 From: amosbird at gmail.com (Amos Bird) Date: Fri, 29 Sep 2017 08:55:58 +0800 Subject: [Interest] Add readline key bindings to all qt input boxes such as QFileDialog Message-ID: <87poaa8grl.fsf@gmail.com> #+OPTIONS: latex:t toc:nil H:3 Hi, How can I change the key bindings in all qt input boxes such as QFileDialog? I'd like to add readline keys to them. regards, -- Amos Bird amosbird at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From chgans at gmail.com Fri Sep 29 08:17:22 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Fri, 29 Sep 2017 19:17:22 +1300 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> Message-ID: On 29 September 2017 at 12:44, Sze Howe Koh wrote: > I just did a quick test on a PC without Qt installed. I'm in > Australia, and my current default mirror is ftp.jaist.ac.jp > > During the "Preparing meta information download..." phase, Resource > Monitor told me that the data is crawling in at 10 Bytes per second > (yes, bytes), and it took a few minutes to get the meta info. However, > once I get past that stage, the actual Qt Creator files come in at > ~300 KiB per second. Resource Monitor reports that both sets of > downloads came from the same mirror, so it's not just a connection > issue. I can confirm, this is exactly what happened during an install this afternoon, all the checksum files were very slow taking few of seconds each, the archive files were initially very slow, but quickly ramped up to 500 KiB/s. This was what i saw in the install phase (was away during the meta info download). Chris From jeanmichael.celerier at gmail.com Fri Sep 29 12:53:42 2017 From: jeanmichael.celerier at gmail.com (=?UTF-8?Q?Jean=2DMicha=C3=ABl_Celerier?=) Date: Fri, 29 Sep 2017 12:53:42 +0200 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> Message-ID: (If I may complain too, in France also the "download meta information" takes a good ten to fifteen seconds every time. I have a 1gbp/s optical fiber connection... downloading the whole qt sdk for a fast local mirror takes less time than it takes for updating these damn metadata, which also are updated sometimes twice when the maintenant tool also needs updating, for twice the frustration). Best, ------- Jean-Michaël Celerier http://www.jcelerier.name On Fri, Sep 29, 2017 at 8:17 AM, Christian Gagneraud wrote: > On 29 September 2017 at 12:44, Sze Howe Koh wrote: > > I just did a quick test on a PC without Qt installed. I'm in > > Australia, and my current default mirror is ftp.jaist.ac.jp > > > > During the "Preparing meta information download..." phase, Resource > > Monitor told me that the data is crawling in at 10 Bytes per second > > (yes, bytes), and it took a few minutes to get the meta info. However, > > once I get past that stage, the actual Qt Creator files come in at > > ~300 KiB per second. Resource Monitor reports that both sets of > > downloads came from the same mirror, so it's not just a connection > > issue. > > I can confirm, this is exactly what happened during an install this > afternoon, all the checksum files were very slow taking few of seconds > each, the archive files were initially very slow, but quickly ramped > up to 500 KiB/s. > This was what i saw in the install phase (was away during the meta > info download). > > > Chris > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mitch.curtis at qt.io Fri Sep 29 17:00:30 2017 From: mitch.curtis at qt.io (Mitch Curtis) Date: Fri, 29 Sep 2017 15:00:30 +0000 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: <57B695AF-6212-4C73-8C26-71F24F42267F@qt.io> , Message-ID: https://bugreports.qt.io/browse/QTIFW-975 ________________________________________ From: Interest on behalf of Jean-Michaël Celerier Sent: Friday, 29 September 2017 12:53:42 PM To: Christian Gagneraud Cc: interest at qt-project.org Subject: Re: [Interest] strip: Customise Qt intaller, remove *.qt.io (If I may complain too, in France also the "download meta information" takes a good ten to fifteen seconds every time. I have a 1gbp/s optical fiber connection... downloading the whole qt sdk for a fast local mirror takes less time than it takes for updating these damn metadata, which also are updated sometimes twice when the maintenant tool also needs updating, for twice the frustration). Best, ------- Jean-Michaël Celerier http://www.jcelerier.name On Fri, Sep 29, 2017 at 8:17 AM, Christian Gagneraud > wrote: On 29 September 2017 at 12:44, Sze Howe Koh > wrote: > I just did a quick test on a PC without Qt installed. I'm in > Australia, and my current default mirror is ftp.jaist.ac.jp > > During the "Preparing meta information download..." phase, Resource > Monitor told me that the data is crawling in at 10 Bytes per second > (yes, bytes), and it took a few minutes to get the meta info. However, > once I get past that stage, the actual Qt Creator files come in at > ~300 KiB per second. Resource Monitor reports that both sets of > downloads came from the same mirror, so it's not just a connection > issue. I can confirm, this is exactly what happened during an install this afternoon, all the checksum files were very slow taking few of seconds each, the archive files were initially very slow, but quickly ramped up to 500 KiB/s. This was what i saw in the install phase (was away during the meta info download). Chris _______________________________________________ Interest mailing list Interest at qt-project.org http://lists.qt-project.org/mailman/listinfo/interest From chgans at gmail.com Sat Sep 30 10:05:41 2017 From: chgans at gmail.com (Christian Gagneraud) Date: Sat, 30 Sep 2017 21:05:41 +1300 Subject: [Interest] strip: Customise Qt intaller, remove *.qt.io In-Reply-To: References: Message-ID: Hi, PART1: Thanks everyone for you answers, after digging into the qtsdk.git repo, i think i might be better staying away from the custom online installer... It's complex, and a custom online installer will give me maintenance nightmares. I think what i need is the QBSP facility (that might be an Enterprise feature), but basically it seems to be tailored to my needs: Add a BSP to Qt SDK, automatically registered as a Kit in QtCreator. Once i get this working to my taste i can move on online repositories. Here is the BootToQt one [1], based on Yocto (by the Linux Fundation). QBSP seems to have stem from the need of the typical foreign/proprietary BSP [2], hence their name. But actually they are just offline repositories. PART 2: The network issues are a separate problem, i have investigated further, and i'm now sure that the root problem for NZ and AU customers is that: 1 - They are directed to japanese mirrors 2 - Download from there is initially very very very slow (600ms TCP round-trip), but can ramp-up to 1.5MiB/s after ca. 30 s 3 - It is not possible to change or *choose* the mirror. 3.1 - the mirror list depends on your license, so you need to login first 4 - The installer starts with 2 online repo, after logging in you end up with hundreds of repos. (I think i have 200+ repos) 5 - If you have a dual license (Desktop + Embedded), you have twice the number of online repositories 6 - Your dual license is actually triple: you implicitly have access to "OpenSource", tripling your number of online repositories. 7 - The Qt installer has the bad habit of re-dowloading the whole "meta-data" thing again and again. 8 - "Qt Open Source" mirrors don't mirror "Qt for Device Creation". As a side note, in this situation (200+ repos), the "Repositories" setting tab becomes unusable: - a Filter line edit would be cool, with regexp support! ;) - maybe filters per "license grant" (see below), I can see a pattern on my install, all repo names starts with either: "Enterprise Linux-x86 Desktop ..." (Qt Enterprise Desktop) "Enterprise Linux-x86 Embedded ..." (Qt for Device Creation) "Qt Linux-x64 ..." (Qt Open Source) - dependencies b/w repo doesn't seem to be managed correctly, you can easily wreak havoc your install, ending with weird fatal dialog saying "foo.bar.baz depends on ...., bailing out" Anyway, as suggested by Kai Koehne, i will open a customer support request and kindly ask for an "Oceania" mirror. [That would be really cool if i could setup my own local mirror] PART 3: As mentioned by someone else (here or on jira), maybe it's all about hash management: hash your stash, per "license grant", donwload their hash per "license grant", if hash matches, your stash is good. if not, repeat with all stash entry, hash based again. The Qt Online installer is based on recursive 7z archives, so maybe the root download should be the hash of the gold root archive object, Then only if needed, the hash-identified 7z archives that contain the hash of their's children. This is exactly how git storage works [3] Chris [1] https://github.com/tmpsantos/meta-boot2qt/tree/tmpsantos-boot2qt_demo/files/qbsp [2] https://en.wikipedia.org/wiki/Board_support_package [3] https://gitlab.com/chgans/aucklug-git-september-2017/raw/master/AuckLUG-git-september-2017.pdf From realnc at gmail.com Sat Sep 30 22:55:55 2017 From: realnc at gmail.com (Nikos Chantziaras) Date: Sat, 30 Sep 2017 23:55:55 +0300 Subject: [Interest] What's the configure switch to have xcb working when deploying Qt apps? Message-ID: Deploying a 64-bit application that bundles the official 64-bit Qt builds (5.6.3) works fine. However, there are no 32-bit builds, so I've built my own. But, the deployed application doesn't work: This application failed to start because it could not find or load the Qt platform plugin "xcb" in "". Available platform plugins are eglfs, minimal, minimalegl, offscreen, wayland-egl, wayland, xcb. This is on Ubuntu 16.04 32-bit. On Ubuntu 16.04 64-bit with the offical Qt builds, it works fine. Is there a list of configure options used by the official builds so I can replicate them? I've been building Qt all day long now, experimenting with configure options, but I can't seem to hit the right combination. From Abin.Joseph at nestgroup.net Wed Sep 6 08:31:07 2017 From: Abin.Joseph at nestgroup.net (Abin Joseph) Date: Wed, 06 Sep 2017 06:31:07 -0000 Subject: [Interest] Unable to do Valgrind Memory Analyzer on i.MX6Q target Message-ID: <986519C39291F34C886B1EBB543F29635650C678@MBCHN01.chn.nestgroup.net> Hi, I am trying to do valgrind memory analysis with Qt Creator on iMX6Q board. I am able to run the application from Qt creator and also able to run application from terminal using command /opt/HelloWorld -platform wayland I have valgrind in my target board, and if I run valgrind -v /opt/HelloWorld -platform wayland, it gives me whole logs which are so difficult to trace out the exact errors. But still application launches. But if I click Analyze->Valgrind Memory Analyzer on Qt Creator, it gives me error and the application does not launch. If I do not give any run time arguments I get the error: Analyzing memory of /opt/HelloWorld QML debugging is enabled. Only use this in a safe environment. This application failed to start because it could not find or load the Qt platform plugin "xcb" in "". Available platform plugins are: eglfs, linuxfb, minimal, minimalegl, offscreen, wayland-egl, wayland. Reinstalling the application may fix this problem. Analyzing finished. ** Process killed by signal ** If I do not give any run time arguments -platform wayland, I get the error: Analyzing memory of /opt/HelloWorld QML debugging is enabled. Only use this in a safe environment. Failed to create display (No such file or directory) Analyzing finished. ** Unknown error ** Is there something I am missing to run valgrind memory analysis on iMx6Q board? With Regards, Abin Joseph ________________________________ Confidentiality Statement / Disclaimer : This message and any attachments is intended for the sole use of the intended recipient. It may contain confidential information. Any unauthorized use, dissemination or modification is strictly prohibited. If you are not the intended recipient, please notify the sender immediately then delete it from all your systems, and do not copy, use or print. Internet communications are not secure and it is the responsibility of the recipient to make sure that it is virus/malicious code exempt. The company/sender cannot be responsible for any unauthorized alterations or modifications made to the contents. If you require any form of confirmation of the contents, please contact the company/sender. The company/sender is not liable for any errors or omissions in the content of this message. ________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From sami.nurmenniemi at qt.io Mon Sep 18 10:44:31 2017 From: sami.nurmenniemi at qt.io (Sami Nurmenniemi) Date: Mon, 18 Sep 2017 08:44:31 -0000 Subject: [Interest] [Development] QRandomGenerator and boot times In-Reply-To: References: <1700113.SrpWEvuCgM@tjmaciei-mobl1> <106099e9-84be-6a11-5bb8-86758b411b80@qt.io> <19569896.oT66G5BakO@tjmaciei-mobl1> Message-ID: On 18/09/2017 7:59 pm, "Christian Gagneraud" > wrote: > > > > On 18/09/2017 7:34 pm, "Sami Nurmenniemi" > wrote: > > On 15.09.2017 18:21, Thiago Macieira wrote: > > On Friday, 15 September 2017 00:31:36 PDT Sami Nurmenniemi > wrote: > > I think we'll just have to accept blocking for the > devices without > hwrng. I don't know if we really support any such > devices. If we do and > boot time is essential for those, we'll have to figure > out some way > (probably saving entropy over reboot). > > And that would be a non-Qt job. I wasn't worried about Qt > on real devices > because I don't expect it to be run early enough to > matter.  On VMs, that's > another story, and if you don't trust your > undercloud-provided /dev/hwrng, why > are you using that cloud? (Though I'll say there's an > Intel team working on > figuring out how a VM can get trust from the actual > hardware, skipping the > hypervisor trust) > > We have made some safety critical customer demos where fast > boot time of Qt framework is essential. It was demoing boot > time of 1.2s for a Qt application to start after powering on > the device. I suppose that demo (and real safety critical use > cases) is going to have problems with the QRandomGenerator > even with hwrng in use. > > > > So basically, you have artificially "optimised" boot time by > ignoring entropy management and now you're blaming > QRandomGenerator. You can do one or the other, not both. > > Me too I can make a device boot in less than a second, but then > don't be surprised if some features are missing/buggy.... > > This sound very much like a beginner mistake, where boot time > takes over functionality. Every one can boot an unfunctional > device in less than a second. This is nothing new. > > The point of mentioning the 1.2s demo was an answer to the statement "I wasn't worried about Qt on real devices because I don't expect it to be run early enough to matter." Maybe the demo application (being the only user space process) needs to make the connection between hwrng + random pool or preserve entropy over reboots. -------------- next part -------------- An HTML attachment was scrubbed... URL: From konstantin.satunin at intel.com Wed Sep 20 13:37:16 2017 From: konstantin.satunin at intel.com (Satunin, Konstantin) Date: Wed, 20 Sep 2017 11:37:16 +0000 Subject: [Interest] QT 5.9.2 Release schedule Message-ID: Hi All, We are interested in the bug fix for memory leak https://bugreports.qt.io/browse/QTBUG-61754 which will be available in 5.9.2 release. QT 5.9.2 snapshot version is available since August 30, and QT 5.10 Alpha was released last week. Can anybody estimate when 5.9.2 might be released? Is it still according to the planned "September 2017" window (only 2 weeks left)? Best wishes, Satunin, Konstantin (Keeper) SSG - Sensor Developer Products -------------------------------------------------------------------- Joint Stock Company Intel A/O Registered legal address: Krylatsky Hills Business Park, 17 Krylatskaya Str., Bldg 4, Moscow 121614, Russian Federation This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jake.petroules at gmail.com Wed Sep 20 19:45:53 2017 From: jake.petroules at gmail.com (Jake Petroules) Date: Wed, 20 Sep 2017 10:45:53 -0700 Subject: [Interest] Building & Running on High Sierra In-Reply-To: <3B6E9D50-92A9-4A45-8395-E3B461687A15@gmail.com> References: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> <3B6E9D50-92A9-4A45-8395-E3B461687A15@gmail.com> Message-ID: <0264E785-12F4-43F8-ACF9-B9BA44EBAFDC@qt.io> This is QTBUG-62266, which has been fixed by https://codereview.qt-project.org/#/c/201546/ for Qt 5.9.2. > On Sep 20, 2017, at 8:00 AM, Patrick Stinson wrote: > > Actually yes, I did. Xcode 9 is required for iOS 11. > > > >> On Sep 20, 2017, at 7:59 AM, Mitch Curtis wrote: >> >> Did you recently update Xcode? I just ran into a failure on the 5.10 branch after installing updates: >> >> https://bugreports.qt.io/browse/QTBUG-63323 >> >> From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org] On Behalf Of Patrick Stinson >> Sent: Wednesday, 20 September 2017 4:56 PM >> To: Qt Interest >> Subject: [Interest] Building & Running on High Sierra >> >> Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). >> >> Qt-5.9.1 fails to build in qcoregraphics.mm like so: >> >> dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm >> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier >> 'InvalidContext' >> require_action(inContext != NULL, InvalidContext, err = paramErr); >> ^ >> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier >> 'InvalidBounds' >> require_action(inBounds != NULL, InvalidBounds, err = paramErr); >> ^ >> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier >> 'InvalidImage' >> require_action(inImage != NULL, InvalidImage, err = paramErr); >> ^ >> 3 errors generated. >> >> Thoughts? > > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest -- Jake Petroules - jake.petroules at qt.io The Qt Company - Silicon Valley Qbs build tool evangelist - qbs.io From jake.petroules at gmail.com Wed Sep 20 20:28:51 2017 From: jake.petroules at gmail.com (Jake Petroules) Date: Wed, 20 Sep 2017 11:28:51 -0700 Subject: [Interest] Building & Running on High Sierra In-Reply-To: References: <3C0A5DF1-2284-42B2-A591-A51680915C05@gmail.com> <3B6E9D50-92A9-4A45-8395-E3B461687A15@gmail.com> <0264E785-12F4-43F8-ACF9-B9BA44EBAFDC@qt.io> Message-ID: Likely https://codereview.qt-project.org/#/c/205954/ > On Sep 20, 2017, at 11:19 AM, Patrick Stinson wrote: > > Thank you. Any thoughts about the branch problem? > >> On Sep 20, 2017, at 10:45 AM, Jake Petroules wrote: >> >> This is QTBUG-62266, which has been fixed by https://codereview.qt-project.org/#/c/201546/ for Qt 5.9.2. >> >>> On Sep 20, 2017, at 8:00 AM, Patrick Stinson wrote: >>> >>> Actually yes, I did. Xcode 9 is required for iOS 11. >>> >>> >>> >>>> On Sep 20, 2017, at 7:59 AM, Mitch Curtis wrote: >>>> >>>> Did you recently update Xcode? I just ran into a failure on the 5.10 branch after installing updates: >>>> >>>> https://bugreports.qt.io/browse/QTBUG-63323 >>>> >>>> From: Interest [mailto:interest-bounces+mitch.curtis=qt.io at qt-project.org] On Behalf Of Patrick Stinson >>>> Sent: Wednesday, 20 September 2017 4:56 PM >>>> To: Qt Interest >>>> Subject: [Interest] Building & Running on High Sierra >>>> >>>> Is there a version or commit that runs on high sierra? I am getting segfaults in a 300-or-so deep stack of calls in libqcocoa.dylib using the branch on git (need to try again to get a useful stack trace). >>>> >>>> Qt-5.9.1 fails to build in qcoregraphics.mm like so: >>>> >>>> dor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm >>>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:81:39: error: use of undeclared identifier >>>> 'InvalidContext' >>>> require_action(inContext != NULL, InvalidContext, err = paramErr); >>>> ^ >>>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:82:38: error: use of undeclared identifier >>>> 'InvalidBounds' >>>> require_action(inBounds != NULL, InvalidBounds, err = paramErr); >>>> ^ >>>> /Users/patrick/dev/vendor/pyqt-sysroot-base/src/qt-everywhere-opensource-src-5.9.1/qtbase/src/gui/painting/qcoregraphics.mm:83:37: error: use of undeclared identifier >>>> 'InvalidImage' >>>> require_action(inImage != NULL, InvalidImage, err = paramErr); >>>> ^ >>>> 3 errors generated. >>>> >>>> Thoughts? >>> >>> _______________________________________________ >>> Interest mailing list >>> Interest at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/interest >> >> -- >> Jake Petroules - jake.petroules at qt.io >> The Qt Company - Silicon Valley >> Qbs build tool evangelist - qbs.io >> > _______________________________________________ > Interest mailing list > Interest at qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest