From Eike.Ziller at theqtcompany.com Tue Mar 1 13:37:59 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Tue, 1 Mar 2016 12:37:59 +0000 Subject: [Qt-creator] [Dev] Qt Creator 4.0 soft-branched In-Reply-To: <43102C71-D5FC-467F-BE5E-066E3F28D76E@theqtcompany.com> References: <2172B575-947C-459D-B717-DA063D0C4275@theqtcompany.com> <43102C71-D5FC-467F-BE5E-066E3F28D76E@theqtcompany.com> Message-ID: <8ED15CBD-5C30-4321-8A5C-61512BD6D31D@digia.com> > On Feb 24, 2016, at 14:36, Ziller Eike wrote: > >> >> On Feb 24, 2016, at 2:34 PM, Ziller Eike wrote: >> >> Hi, >> >> in preparation of the feature freeze for Qt Creator 4.0 > > Which is scheduled for next week: > > https://wiki.qt.io/Qt_Creator_Releases > > Qt Creator 4.0: > • Feature freeze w10 (~Mar 1 2016) > • Beta release w12 (~Mar 15 2016) > • String freeze w14 (~Mar 29 2016) > • Release candidate w16 (~Apr 12 2016) > • Final release w18 (~Apr 26 2016) Reminder: I will do a last merge from master -> 4.0 on Wednesday morning (CEST). After that, everything that goes into master is for Qt Creator 4.1. Br, Eike >> , I “soft-branched“ 4.0. >> >> That means: >> You should prefer the 4.0 branch for all your work that is supposed to end up in the 4.0 release. >> >> The master branch is regularly merged back into 4.0 until the feature freeze next week, >> so you do not have to retarget existing patches for master on codereview. >> >> Br, Eike >> >> -- >> Eike Ziller, Principle Software Engineer - The Qt Company GmbH >> >> The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin >> Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja >> Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B >> >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator > > -- > Eike Ziller, Principle Software Engineer - The Qt Company GmbH > > The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin > Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja > Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From nikolai.kosjar at theqtcompany.com Tue Mar 1 17:09:40 2016 From: nikolai.kosjar at theqtcompany.com (Nikolai Kosjar) Date: Tue, 1 Mar 2016 17:09:40 +0100 Subject: [Qt-creator] Clang Code Model completion not working In-Reply-To: <56D2D5E6.60702@gmail.com> References: <56D2D5E6.60702@gmail.com> Message-ID: <56D5BEC4.6000302@theqtcompany.com> Hi! On 02/28/2016 12:11 PM, Andrzej Telszewski wrote: > Hi, > > Qt Creator: 3.5.1 > OS: Linux > ... > Unfortunately, the completion does not work for me, that is, there is > neither pop-up for functions, classes nor the class members. > > Have I misconfigured something? If you already see diagnostic warnings from libclang, then that's unlikely. Please create a bug report with answers to: 1) Does it work with current Qt Creator snapshots? Snapshots are available at http://download.qt.io/snapshots/qtcreator/master/latest . If it fails, the remaining questions assume that you use the snapshot version. 2) Does it already fail for a main.cpp of a "Qt Console Application" (from Creator's wizard)? 3) Start Qt Creator from a terminal as follows, trigger completion and wait for a while. What's the generated output? $ QT_LOGGING_RULES="qtc.clangcodemodel.ipc=true" ./qtcreator Nikolai From Eike.Ziller at theqtcompany.com Wed Mar 2 09:26:41 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Wed, 2 Mar 2016 08:26:41 +0000 Subject: [Qt-creator] [Dev] Qt Creator 4.0 Feature Freeze (was: soft-branched) In-Reply-To: <8ED15CBD-5C30-4321-8A5C-61512BD6D31D@digia.com> References: <2172B575-947C-459D-B717-DA063D0C4275@theqtcompany.com> <43102C71-D5FC-467F-BE5E-066E3F28D76E@theqtcompany.com> <8ED15CBD-5C30-4321-8A5C-61512BD6D31D@digia.com> Message-ID: <74EF2B54-70EF-4607-A493-A06114C37885@theqtcompany.com> > On Mar 1, 2016, at 1:37 PM, Ziller Eike wrote: > >> >> On Feb 24, 2016, at 14:36, Ziller Eike wrote: >> >>> >>> On Feb 24, 2016, at 2:34 PM, Ziller Eike wrote: >>> >>> Hi, >>> >>> in preparation of the feature freeze for Qt Creator 4.0 >> >> Which is scheduled for next week: >> >> https://wiki.qt.io/Qt_Creator_Releases >> >> Qt Creator 4.0: >> • Feature freeze w10 (~Mar 1 2016) >> • Beta release w12 (~Mar 15 2016) >> • String freeze w14 (~Mar 29 2016) >> • Release candidate w16 (~Apr 12 2016) >> • Final release w18 (~Apr 26 2016) > > Reminder: > I will do a last merge from master -> 4.0 on Wednesday morning (CEST). > After that, everything that goes into master is for Qt Creator 4.1. Merge is done now. > Br, Eike > >>> , I “soft-branched“ 4.0. >>> >>> That means: >>> You should prefer the 4.0 branch for all your work that is supposed to end up in the 4.0 release. >>> >>> The master branch is regularly merged back into 4.0 until the feature freeze next week, >>> so you do not have to retarget existing patches for master on codereview. >>> >>> Br, Eike >>> >>> -- >>> Eike Ziller, Principle Software Engineer - The Qt Company GmbH >>> >>> The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin >>> Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja >>> Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B >>> >>> _______________________________________________ >>> Qt-creator mailing list >>> Qt-creator at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/qt-creator >> >> -- >> Eike Ziller, Principle Software Engineer - The Qt Company GmbH >> >> The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin >> Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja >> Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B >> >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator > > -- > Eike Ziller, Principle Software Engineer - The Qt Company GmbH > > The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin > Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja > Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Riitta-Leena.Miettinen at theqtcompany.com Wed Mar 2 14:17:30 2016 From: Riitta-Leena.Miettinen at theqtcompany.com (Miettinen Riitta-Leena) Date: Wed, 2 Mar 2016 13:17:30 +0000 Subject: [Qt-creator] Problems with wizard.json regarding resolution of variables In-Reply-To: <1456401812.8279.7.camel@theqtcompany.com> References: <56CEC3DD.9020106@stefan-hans.de> <1456401812.8279.7.camel@theqtcompany.com> Message-ID: Hello, The wizards are described in the Qt Creator Manual: http://doc.qt.io/qtcreator/creator-project-wizards.html http://doc.qt.io/qtcreator/creator-project-wizards-json.html http://doc.qt.io/qtcreator/creator-project-wizards-xml.html To help us improve the manual, please let us know if you cannot find the information you need on those pages. Cheers, Leena Leena Miettinen | Documentation Engineer | The Qt Company Digia Germany GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B Email: riitta-leena.miettinen at theqtcompany.com | Phone: +49 30 63 92 3255 | www.qt.io | Qt Blog: http://blog.qt.io/ | Twitter: @QtbyDigia, @Qtproject | Facebook: www.facebook.com/qt -----Original Message----- From: Qt-creator [mailto:qt-creator-bounces+riitta-leena.miettinen=theqtcompany.com at qt-project.org] On Behalf Of Hunger Tobias Sent: Donnerstag, 25. Februar 2016 13:04 To: qt-creator at qt-project.org Subject: Re: [Qt-creator] Problems with wizard.json regarding resolution of variables On Do, 2016-02-25 at 10:05 +0100, Stefan wrote: > What is the difference in using wizard.json resp. wizard.xml? A lot:-) There is e.g. no %ProjectName% in wizard.xml, while there is %{ProjectName} in wizard.json. But you are using that, so that is probably not what is biting you. > What am I doing wrong? What does not work for you? I could not figure that out based on your mail, sorry. To debug wizards I recommend to set a keyboard shortcut (in Tools>Options>Environment>Keyboard) to the Wizard.Inspect action. You can then trigger the wizard you are curious about and trigger the Wizard.Inspect action at any time the (JSON) wizard is running. That will get you a page with all the variables the wizard knows at that point in time. I found the Inspector a great help to debug problems with my wizards. Best Regards, Tobias -- Tobias Hunger, Senior Software Engineer | The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B _______________________________________________ Qt-creator mailing list Qt-creator at qt-project.org http://lists.qt-project.org/mailman/listinfo/qt-creator From atelszewski at gmail.com Wed Mar 2 20:27:35 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Wed, 2 Mar 2016 20:27:35 +0100 Subject: [Qt-creator] Clang Code Model completion not working In-Reply-To: <56D5BEC4.6000302@theqtcompany.com> References: <56D2D5E6.60702@gmail.com> <56D5BEC4.6000302@theqtcompany.com> Message-ID: <56D73EA7.5080507@gmail.com> On 01/03/16 17:09, Nikolai Kosjar wrote: > Hi! > > On 02/28/2016 12:11 PM, Andrzej Telszewski wrote: >> Hi, >> >> Qt Creator: 3.5.1 >> OS: Linux >> ... >> Unfortunately, the completion does not work for me, that is, there is >> neither pop-up for functions, classes nor the class members. >> >> Have I misconfigured something? > > If you already see diagnostic warnings from libclang, then that's unlikely. > > Please create a bug report with answers to: > > 1) Does it work with current Qt Creator snapshots? Snapshots are > available at http://download.qt.io/snapshots/qtcreator/master/latest . > If it fails, the remaining questions assume that you use the snapshot > version. > > 2) Does it already fail for a main.cpp of a "Qt Console Application" > (from Creator's wizard)? > > 3) Start Qt Creator from a terminal as follows, trigger completion and > wait for a while. What's the generated output? > > $ QT_LOGGING_RULES="qtc.clangcodemodel.ipc=true" ./qtcreator > > Nikolai > Upgrading to 3.6.0 solves the problem. Should I still fill the bug or 3.5.1 is no longer supported? -- Best regards, Andrzej Telszewski From Stefan.Walter at lisec.com Thu Mar 3 06:49:13 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 05:49:13 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 Message-ID: Hi, I am trying to build Qt-Creator 3.5.1 with Qt 5.5.1 on CentOS 6.7 x86_64. I am right at the begin the following issue: g++ -c -pipe -O2 -std=c++0x -fvisibility=hidden -fvisibility-inlines-hidden -Wall -W -D_REENTRANT -fPIC -DQBS_LIBRARY -DQT_NO_CAST_FROM_ASCII -DQT_NO_PROCESS_COMBINED_ARGUMENT_START -DQBS_VERSION=\"1.4.3\" -DQT_CREATOR -DQML_BUILD_STATIC_LIB -DSRCDIR=\"/home/qt-user/qt-creator-opensource-src-3.5.1/src/shared/qbs/src/lib/corelib\" -DQBS_ENABLE_PROJECT_FILE_UPDATES -DQT_NO_DEBUG -DQT_SCRIPT_LIB -DQT_GUI_LIB -DQT_XML_LIB -DQT_CORE_LIB -I. -I../../../src -I. -I../../lib -I/home/qt-user/qt-5.5/qt-5.5.1/qt-install/include -I/home/qt-user/qt-5.5/qt-5.5.1/qt-install/include/QtScript -I/home/qt-user/qt-5.5/qt-5.5.1/qt-install/include/QtGui -I/home/qt-user/qt-5.5/qt-5.5.1/qt-install/include/QtXml -I/home/qt-user/qt-5.5/qt-5.5.1/qt-install/include/QtCore -I. -I/home/qt-user/qt-5.5/qt-5.5.1/qt-install/mkspecs/linux-g++ -o runenvironment.o api/runenvironment.cpp In file included from api/runenvironment.cpp:37: ./language/scriptengine.h:128: error: \u2018nullptr\u2019 was not declared in this scope make[2]: *** [runenvironment.o] Error 1 make[2]: Leaving directory `/home/qt-user/qt-creator-opensource-src-3.5.1/src/shared/qbs/src/lib/corelib' make[1]: *** [sub-shared-qbs-src-lib-corelib-make_first-ordered] Error 2 make[1]: Leaving directory `/home/qt-user/qt-creator-opensource-src-3.5.1/src' make: *** [sub-src-make_first-ordered] Error 2 Please advise what can I do to solve this? Regards, Stefan -------------- next part -------------- An HTML attachment was scrubbed... URL: From tobias.hunger at gmail.com Thu Mar 3 07:37:37 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Thu, 3 Mar 2016 07:37:37 +0100 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: Message-ID: Am 03.03.2016 06:49 schrieb "Walter Stefan" : > I am trying to build Qt-Creator 3.5.1 with Qt 5.5.1 on CentOS 6.7 x86_64. I am right at the begin the following issue: > In file included from api/runenvironment.cpp:37: > > ./language/scriptengine.h:128: error: \u2018nullptr\u2019 was not declared in this scope You will need to upgrade your compiler to an supported version. We have started to rely on C++11 features unconditionally (it is 2016, that stuff has been around for 5 years now!). It really improves the language, you should update to something that supports it, not just for Qt Creators sake! The README file has details. IIRC that is GCC 4.7.X or later. Best regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From Stefan.Walter at lisec.com Thu Mar 3 07:45:50 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 06:45:50 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: Message-ID: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> Dear Tobias, Well, I disagree. CentOS 6.7 is by far not that old and they still ship it with g++ 4.4.7. I would not even know how to upgrade the G++ on a CentOS 6.x. Can you point me how to upgrade the g++ compiler? Therefore, I stuck at that point now. Qt 5.5.1 compiled without a single issue. Regards, Stefan From: Tobias Hunger [mailto:tobias.hunger at gmail.com] Sent: Donnerstag, 3. März 2016 10:38 To: Walter Stefan Cc: qt-creator Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 Am 03.03.2016 06:49 schrieb "Walter Stefan" >: > I am trying to build Qt-Creator 3.5.1 with Qt 5.5.1 on CentOS 6.7 x86_64. I am right at the begin the following issue: > In file included from api/runenvironment.cpp:37: > > ./language/scriptengine.h:128: error: \u2018nullptr\u2019 was not declared in this scope You will need to upgrade your compiler to an supported version. We have started to rely on C++11 features unconditionally (it is 2016, that stuff has been around for 5 years now!). It really improves the language, you should update to something that supports it, not just for Qt Creators sake! The README file has details. IIRC that is GCC 4.7.X or later. Best regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From Eike.Ziller at theqtcompany.com Thu Mar 3 08:49:26 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Thu, 3 Mar 2016 07:49:26 +0000 Subject: [Qt-creator] Clang Code Model completion not working In-Reply-To: <56D73EA7.5080507@gmail.com> References: <56D2D5E6.60702@gmail.com> <56D5BEC4.6000302@theqtcompany.com> <56D73EA7.5080507@gmail.com> Message-ID: <83EC74B7-DE3B-461A-A141-5AF93CEC0F91@theqtcompany.com> > On Mar 2, 2016, at 8:27 PM, Andrzej Telszewski wrote: > > On 01/03/16 17:09, Nikolai Kosjar wrote: >> Hi! >> >> On 02/28/2016 12:11 PM, Andrzej Telszewski wrote: >>> Hi, >>> >>> Qt Creator: 3.5.1 >>> OS: Linux >>> ... >>> Unfortunately, the completion does not work for me, that is, there is >>> neither pop-up for functions, classes nor the class members. >>> >>> Have I misconfigured something? >> >> If you already see diagnostic warnings from libclang, then that's unlikely. >> >> Please create a bug report with answers to: >> >> 1) Does it work with current Qt Creator snapshots? Snapshots are >> available at http://download.qt.io/snapshots/qtcreator/master/latest . >> If it fails, the remaining questions assume that you use the snapshot >> version. >> >> 2) Does it already fail for a main.cpp of a "Qt Console Application" >> (from Creator's wizard)? >> >> 3) Start Qt Creator from a terminal as follows, trigger completion and >> wait for a while. What's the generated output? >> >> $ QT_LOGGING_RULES="qtc.clangcodemodel.ipc=true" ./qtcreator >> >> Nikolai >> > > Upgrading to 3.6.0 solves the problem. > Should I still fill the bug or 3.5.1 is no longer supported? There will be no more releases of 3.5.x. Br, Eike > -- > Best regards, > Andrzej Telszewski > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Eike.Ziller at theqtcompany.com Thu Mar 3 09:05:44 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Thu, 3 Mar 2016 08:05:44 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> Message-ID: > On Mar 3, 2016, at 7:45 AM, Walter Stefan wrote: > > Dear Tobias, > > Well, I disagree. CentOS 6.7 is by far not that old and they still ship it with g++ 4.4.7. That is the decision of the CentOS distribution. CentOS 6.7 was released August 2015, where GCC 5.2 was already released (and GCC 4.9.3 was the latest 4.x line release). (CentOS 7 which was first released in July 2014 contains GCC 4.8 btw) > I would not even know how to upgrade the G++ on a CentOS 6.x. Can you point me how to upgrade the g++ compiler? You can install and use devtoolset http://linux.web.cern.ch/linux/devtoolset/ (probably also available through whatever package manager is used by CentOS). > Therefore, I stuck at that point now. Qt 5.5.1 compiled without a single issue. Qt has less requirements on the compiler and runs on more platforms than Qt Creator. Br, Eike > Regards, > Stefan > > > From: Tobias Hunger [mailto:tobias.hunger at gmail.com] > Sent: Donnerstag, 3. März 2016 10:38 > To: Walter Stefan > Cc: qt-creator > Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 > > > Am 03.03.2016 06:49 schrieb "Walter Stefan" : > > I am trying to build Qt-Creator 3.5.1 with Qt 5.5.1 on CentOS 6.7 x86_64. I am right at the begin the following issue: > > > In file included from api/runenvironment.cpp:37: > > > > ./language/scriptengine.h:128: error: \u2018nullptr\u2019 was not declared in this scope > > You will need to upgrade your compiler to an supported version. We have started to rely on C++11 features unconditionally (it is 2016, that stuff has been around for 5 years now!). It really improves the language, you should update to something that supports it, not just for Qt Creators sake! > > The README file has details. IIRC that is GCC 4.7.X or later. > > Best regards, > Tobias > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Stefan.Walter at lisec.com Thu Mar 3 09:19:24 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 08:19:24 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> Message-ID: If I understand you correctly, then I should compile Qt-Creator with the devtoolset but can still compile Qt 5.5.1 with the older g++? Or do you mean that I should compile both with the devtoolset? Regards, Stefan -----Original Message----- From: Ziller Eike [mailto:Eike.Ziller at theqtcompany.com] Sent: Donnerstag, 3. März 2016 12:06 To: Walter Stefan Cc: Tobias Hunger ; qt-creator Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 > On Mar 3, 2016, at 7:45 AM, Walter Stefan wrote: > > Dear Tobias, > > Well, I disagree. CentOS 6.7 is by far not that old and they still ship it with g++ 4.4.7. That is the decision of the CentOS distribution. CentOS 6.7 was released August 2015, where GCC 5.2 was already released (and GCC 4.9.3 was the latest 4.x line release). (CentOS 7 which was first released in July 2014 contains GCC 4.8 btw) > I would not even know how to upgrade the G++ on a CentOS 6.x. Can you point me how to upgrade the g++ compiler? You can install and use devtoolset http://linux.web.cern.ch/linux/devtoolset/ (probably also available through whatever package manager is used by CentOS). > Therefore, I stuck at that point now. Qt 5.5.1 compiled without a single issue. Qt has less requirements on the compiler and runs on more platforms than Qt Creator. Br, Eike > Regards, > Stefan > > > From: Tobias Hunger [mailto:tobias.hunger at gmail.com] > Sent: Donnerstag, 3. März 2016 10:38 > To: Walter Stefan > Cc: qt-creator > Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues > on CentOS 6.7 x86_64 > > > Am 03.03.2016 06:49 schrieb "Walter Stefan" : > > I am trying to build Qt-Creator 3.5.1 with Qt 5.5.1 on CentOS 6.7 x86_64. I am right at the begin the following issue: > > > In file included from api/runenvironment.cpp:37: > > > > ./language/scriptengine.h:128: error: \u2018nullptr\u2019 was not > > declared in this scope > > You will need to upgrade your compiler to an supported version. We have started to rely on C++11 features unconditionally (it is 2016, that stuff has been around for 5 years now!). It really improves the language, you should update to something that supports it, not just for Qt Creators sake! > > The README file has details. IIRC that is GCC 4.7.X or later. > > Best regards, > Tobias > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From tobias.hunger at gmail.com Thu Mar 3 11:27:35 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Thu, 3 Mar 2016 11:27:35 +0100 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> Message-ID: Hi Walter, Creator does not need to support all the platforms Qt runs on (just the main desktop ones), so we tend to try out new things there before they hit Qt. Please take a look here: http://doc-snapshots.qt.io/qt5-5.6/supported-platforms-and-configurations.html Qt 5.5 already dropped support for your compiler on RHEL 6. Qt 5.4 still supports it though. Best Regards, Tobias From Stefan.Walter at lisec.com Thu Mar 3 12:01:07 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 11:01:07 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> Message-ID: <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> Dear Tobias, Thanks a lot. I got the compilation done by using the devtools. When I try to execute the qt-creator, the screen does not open and I get a few errors. Can you help me on that as well? My Output: [walteste at aedu2-build-qt bin]$ ./qtcreator failed to get the current screen resources libGL error: No matching fbConfigs or visuals found libGL error: failed to load driver: swrast QXcbConnection: XCB error: 172 (Unknown), sequence: 162, resource id: 150, major code: 149 (Unknown), minor code: 20 WARNING: Application /home/walteste/qt-creator-opensource-src-3.5.1/bin/qtcreator. calling GLX 1.3 function "glXCreatePbuffer" when GLX 1.3 is not supported! This is an application bug! QQuickWidget: Failed to make context current QQuickWidget::resizeEvent() no OpenGL context Best Regards, Stefan -----Original Message----- From: Tobias Hunger [mailto:tobias.hunger at gmail.com] Sent: Donnerstag, 3. März 2016 14:28 To: Walter Stefan Cc: qt-creator Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 Hi Walter, Creator does not need to support all the platforms Qt runs on (just the main desktop ones), so we tend to try out new things there before they hit Qt. Please take a look here: http://doc-snapshots.qt.io/qt5-5.6/supported-platforms-and-configurations.html Qt 5.5 already dropped support for your compiler on RHEL 6. Qt 5.4 still supports it though. Best Regards, Tobias From orgads at gmail.com Thu Mar 3 12:33:08 2016 From: orgads at gmail.com (Orgad Shaneh) Date: Thu, 3 Mar 2016 13:33:08 +0200 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> Message-ID: On Thu, Mar 3, 2016 at 1:01 PM, Walter Stefan wrote: > Dear Tobias, > > Thanks a lot. I got the compilation done by using the devtools. > > When I try to execute the qt-creator, the screen does not open and I get a > few errors. Can you help me on that as well? > > My Output: > [walteste at aedu2-build-qt bin]$ ./qtcreator > failed to get the current screen resources > libGL error: No matching fbConfigs or visuals found > libGL error: failed to load driver: swrast > QXcbConnection: XCB error: 172 (Unknown), sequence: 162, resource id: 150, > major code: 149 (Unknown), minor code: 20 > WARNING: Application > /home/walteste/qt-creator-opensource-src-3.5.1/bin/qtcreator. calling GLX > 1.3 function "glXCreatePbuffer" when GLX 1.3 is not supported! This is an > application bug! > QQuickWidget: Failed to make context current > QQuickWidget::resizeEvent() no OpenGL context > > Best Regards, > Stefan > > -----Original Message----- > From: Tobias Hunger [mailto:tobias.hunger at gmail.com] > Sent: Donnerstag, 3. März 2016 14:28 > To: Walter Stefan > Cc: qt-creator > Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on > CentOS 6.7 x86_64 > > Hi Walter, > > Creator does not need to support all the platforms Qt runs on (just the > main desktop ones), so we tend to try out new things there before they hit > Qt. > > Please take a look here: > > http://doc-snapshots.qt.io/qt5-5.6/supported-platforms-and-configurations.html > > Qt 5.5 already dropped support for your compiler on RHEL 6. Qt 5.4 still > supports it though. > > Best Regards, > Tobias > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > Try qtcreator -noload Welcome - Orgad -------------- next part -------------- An HTML attachment was scrubbed... URL: From Stefan.Walter at lisec.com Thu Mar 3 12:49:38 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 11:49:38 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> Message-ID: <57418cda78b84431a831a47e6295810a@ATSE-MAIL4.lisec.internal> Thanks, that worked! From: Orgad Shaneh [mailto:orgads at gmail.com] Sent: Donnerstag, 3. März 2016 15:33 To: Walter Stefan Cc: Tobias Hunger ; qt-creator Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 On Thu, Mar 3, 2016 at 1:01 PM, Walter Stefan > wrote: Dear Tobias, Thanks a lot. I got the compilation done by using the devtools. When I try to execute the qt-creator, the screen does not open and I get a few errors. Can you help me on that as well? My Output: [walteste at aedu2-build-qt bin]$ ./qtcreator failed to get the current screen resources libGL error: No matching fbConfigs or visuals found libGL error: failed to load driver: swrast QXcbConnection: XCB error: 172 (Unknown), sequence: 162, resource id: 150, major code: 149 (Unknown), minor code: 20 WARNING: Application /home/walteste/qt-creator-opensource-src-3.5.1/bin/qtcreator. calling GLX 1.3 function "glXCreatePbuffer" when GLX 1.3 is not supported! This is an application bug! QQuickWidget: Failed to make context current QQuickWidget::resizeEvent() no OpenGL context Best Regards, Stefan -----Original Message----- From: Tobias Hunger [mailto:tobias.hunger at gmail.com] Sent: Donnerstag, 3. März 2016 14:28 To: Walter Stefan > Cc: qt-creator > Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 Hi Walter, Creator does not need to support all the platforms Qt runs on (just the main desktop ones), so we tend to try out new things there before they hit Qt. Please take a look here: http://doc-snapshots.qt.io/qt5-5.6/supported-platforms-and-configurations.html Qt 5.5 already dropped support for your compiler on RHEL 6. Qt 5.4 still supports it though. Best Regards, Tobias _______________________________________________ Qt-creator mailing list Qt-creator at qt-project.org http://lists.qt-project.org/mailman/listinfo/qt-creator Try qtcreator -noload Welcome - Orgad -------------- next part -------------- An HTML attachment was scrubbed... URL: From orgads at gmail.com Thu Mar 3 12:50:53 2016 From: orgads at gmail.com (Orgad Shaneh) Date: Thu, 3 Mar 2016 13:50:53 +0200 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: <57418cda78b84431a831a47e6295810a@ATSE-MAIL4.lisec.internal> References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> <57418cda78b84431a831a47e6295810a@ATSE-MAIL4.lisec.internal> Message-ID: On Thu, Mar 3, 2016 at 1:49 PM, Walter Stefan wrote: > Thanks, that worked! > > > > *From:* Orgad Shaneh [mailto:orgads at gmail.com] > *Sent:* Donnerstag, 3. März 2016 15:33 > *To:* Walter Stefan > *Cc:* Tobias Hunger ; qt-creator < > qt-creator at qt-project.org> > > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > > > > Try qtcreator -noload Welcome > Now you can disable Welcome in Help -> About Plugins, so you don't have to pass it every time. - Orgad -------------- next part -------------- An HTML attachment was scrubbed... URL: From Stefan.Walter at lisec.com Thu Mar 3 13:50:16 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 12:50:16 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> <57418cda78b84431a831a47e6295810a@ATSE-MAIL4.lisec.internal> Message-ID: <51bea331d39e40d0a73679d52b87110c@ATSE-MAIL4.lisec.internal> Awesome, that was a great tip! From: Orgad Shaneh [mailto:orgads at gmail.com] Sent: Donnerstag, 3. März 2016 15:51 To: Walter Stefan Cc: Tobias Hunger ; qt-creator Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 On Thu, Mar 3, 2016 at 1:49 PM, Walter Stefan > wrote: Thanks, that worked! From: Orgad Shaneh [mailto:orgads at gmail.com] Sent: Donnerstag, 3. März 2016 15:33 To: Walter Stefan > Cc: Tobias Hunger >; qt-creator > Qt-creator at qt-project.org http://lists.qt-project.org/mailman/listinfo/qt-creator Try qtcreator -noload Welcome Now you can disable Welcome in Help -> About Plugins, so you don't have to pass it every time. - Orgad -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.loehning at theqtcompany.com Thu Mar 3 16:34:08 2016 From: robert.loehning at theqtcompany.com (=?UTF-8?Q?Robert_L=c3=b6hning?=) Date: Thu, 3 Mar 2016 16:34:08 +0100 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: <51bea331d39e40d0a73679d52b87110c@ATSE-MAIL4.lisec.internal> References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> <57418cda78b84431a831a47e6295810a@ATSE-MAIL4.lisec.internal> <51bea331d39e40d0a73679d52b87110c@ATSE-MAIL4.lisec.internal> Message-ID: <56D85970.8010507@theqtcompany.com> Hi, you might want to disable the plugins QmlDesigner and QmlProfiler as well. When the Welcome screen crashes with OpenGL issues, they will probably not work either. Best Regards, Robert Am 03.03.2016 um 13:50 schrieb Walter Stefan: > Awesome, that was a great tip! > > From: Orgad Shaneh [mailto:orgads at gmail.com] > Sent: Donnerstag, 3. März 2016 15:51 > To: Walter Stefan > Cc: Tobias Hunger ; qt-creator > Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 > > On Thu, Mar 3, 2016 at 1:49 PM, Walter Stefan > wrote: > Thanks, that worked! > > From: Orgad Shaneh [mailto:orgads at gmail.com] > Sent: Donnerstag, 3. März 2016 15:33 > To: Walter Stefan > > Cc: Tobias Hunger >; qt-creator > > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > > Try qtcreator -noload Welcome > > Now you can disable Welcome in Help -> About Plugins, so you don't have to pass it every time. > > - Orgad -- Robert Löhning, Software Engineer - The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Stefan.Walter at lisec.com Thu Mar 3 17:28:33 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Thu, 3 Mar 2016 16:28:33 +0000 Subject: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 In-Reply-To: <56D85970.8010507@theqtcompany.com> References: <8d31aa7326c64a658c15f060caf0cc7d@ATSE-MAIL4.lisec.internal> <2ce594c91e744e7e9e4681da26de0948@ATSE-MAIL4.lisec.internal> <57418cda78b84431a831a47e6295810a@ATSE-MAIL4.lisec.internal> <51bea331d39e40d0a73679d52b87110c@ATSE-MAIL4.lisec.internal> <56D85970.8010507@theqtcompany.com> Message-ID: <6d96b0a70cb448c48e4ea52474558799@ATSE-MAIL4.lisec.internal> Thanks for the valuable tip. -----Original Message----- From: Qt-creator [mailto:qt-creator-bounces+stefan.walter=lisec.com at qt-project.org] On Behalf Of Robert Löhning Sent: Donnerstag, 3. März 2016 19:34 To: qt-creator at qt-project.org Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues on CentOS 6.7 x86_64 Hi, you might want to disable the plugins QmlDesigner and QmlProfiler as well. When the Welcome screen crashes with OpenGL issues, they will probably not work either. Best Regards, Robert Am 03.03.2016 um 13:50 schrieb Walter Stefan: > Awesome, that was a great tip! > > From: Orgad Shaneh [mailto:orgads at gmail.com] > Sent: Donnerstag, 3. März 2016 15:51 > To: Walter Stefan > Cc: Tobias Hunger ; qt-creator > > Subject: Re: [Qt-creator] Qt-Creator 3.5.1 with Qt 5.5.1 build issues > on CentOS 6.7 x86_64 > > On Thu, Mar 3, 2016 at 1:49 PM, Walter Stefan > wrote: > Thanks, that worked! > > From: Orgad Shaneh [mailto:orgads at gmail.com] > Sent: Donnerstag, 3. März 2016 15:33 > To: Walter Stefan > > > Cc: Tobias Hunger > >; qt-creator > > > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > > Try qtcreator -noload Welcome > > Now you can disable Welcome in Help -> About Plugins, so you don't have to pass it every time. > > - Orgad -- Robert Löhning, Software Engineer - The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B _______________________________________________ Qt-creator mailing list Qt-creator at qt-project.org http://lists.qt-project.org/mailman/listinfo/qt-creator From duane.hebert at group-upc.com Thu Mar 3 17:42:47 2016 From: duane.hebert at group-upc.com (Duane) Date: Thu, 3 Mar 2016 11:42:47 -0500 Subject: [Qt-creator] Problems switching kit between 4.85 and 5.5 Message-ID: QtCreator on Linux with Qt 4.8.5 and Qt 5.5.1 installed. We're moving our projects from Qt4 to Qt5. U I have QtCreator setup with a kit using each so I can support both for the moment. Recently when trying to build Qt4 projects I'm getting errors in files like qheaderview.h where it doesn't like setSectionsMovable() which replaced Qt4's setMovable(), so I can see it's getting the header. But if I look at the Makefile generated it seems to have the correct information. In the kit/version section I can see that the includes look correct. Any idea why QtCreator is not getting the correct path? From freddy311082 at gmail.com Thu Mar 3 18:01:31 2016 From: freddy311082 at gmail.com (Freddy Martinez Garcia) Date: Thu, 3 Mar 2016 14:01:31 -0300 Subject: [Qt-creator] Update QWebElement from different threads Message-ID: Hi guys... I need to update a QWebElement from different threads... Imagine that I have the following statement: QString htmlElemCode = "

Hello World

"; QWebElement elem = webFrame->findFirtsElement("#body_div"); elem.appendInside(htmlElemCode); well, I need to run elem.appendInside(htmlElemCode); in a thread different than the main thread. I had defined this piece of code inside one slot, which I need to call it in a different threads, not only from the main thread. The problem come when I try to call the slot from a thread different than the main thread, because the GUI doesn't update it's html content... however, if I call the slot from the main thread, I can see a success result on my html content, updating the html elements as I wish. My main goal is update my html content dynamically using a javascript functions defined on my html code, which let me add some rows and columns to my html tables, but, I need to do this using various threads... is that possible ??? what is the best way to do that ?? regards *============================================="El tamaño de tus logros depende del tamaño de tus metas." * *C++ and Qt Senior Developer* *Lic. Computer Science* *Buenos Aires, Argentina* -------------- next part -------------- An HTML attachment was scrubbed... URL: From atelszewski at gmail.com Thu Mar 3 19:30:22 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Thu, 3 Mar 2016 19:30:22 +0100 Subject: [Qt-creator] Clang Code Model: add -Wno-padded Message-ID: <56D882BE.4060006@gmail.com> Hi, Qt Creator: 3.6.0 Could _-Wno-padded_ flag be added to the Clang Code Model flags? I don't believe that warning about padding should be the default behavior. -- Best regards, Andrzej Telszewski From atelszewski at gmail.com Thu Mar 3 20:44:27 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Thu, 3 Mar 2016 20:44:27 +0100 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: References: <56D882BE.4060006@gmail.com> Message-ID: <56D8941B.7020801@gmail.com> On 03/03/16 20:23, Bubke Marco wrote: > Hi, we will add more options in 4.0 to the code model, so you can choose different sets. > > Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. > Nope, it does not tell you it's suboptimal. It tells you that compiler will add the additional padding/alignment bytes. Having structure with say 3 bool's and nothing more is perfectly fine. In such a case, to silence the warning, I would have to add some dummy fields, which most of the time is not necessary. > -- > Sent from cellphone, sorry for the typos > > > > On March 3, 2016 19:30:31 Andrzej Telszewski wrote: > >> Hi, >> >> Qt Creator: 3.6.0 >> >> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >> I don't believe that warning about padding should be the default behavior. >> >> -- >> Best regards, >> Andrzej Telszewski >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Best regards, Andrzej Telszewski From Marco.Bubke at theqtcompany.com Thu Mar 3 21:55:24 2016 From: Marco.Bubke at theqtcompany.com (Bubke Marco) Date: Thu, 3 Mar 2016 20:55:24 +0000 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> Message-ID: You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. I think it quite informative that bool x, y, z; int w; char c; Is not optimal. If you don't like that disable it in the settings. -- Sent from cellphone, sorry for the typos On March 3, 2016 20:44:28 Andrzej Telszewski wrote: > On 03/03/16 20:23, Bubke Marco wrote: >> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >> >> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >> > > Nope, it does not tell you it's suboptimal. It tells you that compiler > will add the additional padding/alignment bytes. > > Having structure with say 3 bool's and nothing more is perfectly fine. > In such a case, to silence the warning, I would have to add some dummy > fields, which most of the time is not necessary. > >> -- >> Sent from cellphone, sorry for the typos >> >> >> >> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >> >>> Hi, >>> >>> Qt Creator: 3.6.0 >>> >>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>> I don't believe that warning about padding should be the default behavior. >>> >>> -- >>> Best regards, >>> Andrzej Telszewski >>> _______________________________________________ >>> Qt-creator mailing list >>> Qt-creator at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/qt-creator >> > > > -- > Best regards, > Andrzej Telszewski -- Sent from cellphone, sorry for the typos From atelszewski at gmail.com Thu Mar 3 22:22:22 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Thu, 3 Mar 2016 22:22:22 +0100 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> Message-ID: <56D8AB0E.7020505@gmail.com> On 03/03/16 21:55, Bubke Marco wrote: > You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. > > I think it quite informative that > > bool x, y, z; > int w; > char c; > > Is not optimal. If you don't like that disable it in the settings. I do think that having this kind of information is useful, but in my opinion -Wno-padded should be default, and turning on the warning should be left to those who need it. I personally added -Wno-padded to the clang model settings. Say you have a struct like that: struct nopad { bool flag1; bool flag2; }; It is perfectly fine structure and you don't have to pad it, yet you get the warning. I would understand, if the warning told you to re-arrange the members, but it simply warns you about what is the default compiler behavior, that it will add 2 or 6 bytes of padding. This has been the default compiler behavior for ages, so I don't see the point of giving the warning by default;) In addition, the padding depends on the architecture you're compiling for, so you would have to take it into account. I'm not against the warning, I'm just saying it shouldn't be enabled by default. My real structure is like that: typedef struct { string name; bool can_span; } info_key_t; There is nothing sensible I can change to make the warning disappear. > > -- > Sent from cellphone, sorry for the typos > > > On March 3, 2016 20:44:28 Andrzej Telszewski wrote: > >> On 03/03/16 20:23, Bubke Marco wrote: >>> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >>> >>> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >>> >> >> Nope, it does not tell you it's suboptimal. It tells you that compiler >> will add the additional padding/alignment bytes. >> >> Having structure with say 3 bool's and nothing more is perfectly fine. >> In such a case, to silence the warning, I would have to add some dummy >> fields, which most of the time is not necessary. >> >>> -- >>> Sent from cellphone, sorry for the typos >>> >>> >>> >>> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >>> >>>> Hi, >>>> >>>> Qt Creator: 3.6.0 >>>> >>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>> I don't believe that warning about padding should be the default behavior. >>>> >>>> -- >>>> Best regards, >>>> Andrzej Telszewski >>>> _______________________________________________ >>>> Qt-creator mailing list >>>> Qt-creator at qt-project.org >>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>> >> >> >> -- >> Best regards, >> Andrzej Telszewski > > -- > Sent from cellphone, sorry for the typos > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Pozdrawiam, Best regards, Andrzej Telszewski From Marco.Bubke at theqtcompany.com Fri Mar 4 01:03:01 2016 From: Marco.Bubke at theqtcompany.com (Bubke Marco) Date: Fri, 4 Mar 2016 00:03:01 +0000 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: <56D8AB0E.7020505@gmail.com> References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> Message-ID: I have similar structures and have never seen the warning for that cases. Which clang version do you use? -- Sent from cellphone, sorry for the typos On March 3, 2016 22:22:26 Andrzej Telszewski wrote: > On 03/03/16 21:55, Bubke Marco wrote: >> You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. >> >> I think it quite informative that >> >> bool x, y, z; >> int w; >> char c; >> >> Is not optimal. If you don't like that disable it in the settings. > > I do think that having this kind of information is useful, but in my > opinion -Wno-padded should be default, and turning on the warning should > be left to those who need it. I personally added -Wno-padded to the > clang model settings. > > Say you have a struct like that: > struct nopad > { > bool flag1; > bool flag2; > }; > > It is perfectly fine structure and you don't have to pad it, yet you get > the warning. I would understand, if the warning told you to re-arrange > the members, but it simply warns you about what is the default compiler > behavior, that it will add 2 or 6 bytes of padding. This has been the > default compiler behavior for ages, so I don't see the point of giving > the warning by default;) In addition, the padding depends on the > architecture you're compiling for, so you would have to take it into > account. > > I'm not against the warning, I'm just saying it shouldn't be enabled by > default. > > My real structure is like that: > typedef struct > { > string name; > bool can_span; > } info_key_t; > > There is nothing sensible I can change to make the warning disappear. > >> >> -- >> Sent from cellphone, sorry for the typos >> >> >> On March 3, 2016 20:44:28 Andrzej Telszewski wrote: >> >>> On 03/03/16 20:23, Bubke Marco wrote: >>>> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >>>> >>>> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >>>> >>> >>> Nope, it does not tell you it's suboptimal. It tells you that compiler >>> will add the additional padding/alignment bytes. >>> >>> Having structure with say 3 bool's and nothing more is perfectly fine. >>> In such a case, to silence the warning, I would have to add some dummy >>> fields, which most of the time is not necessary. >>> >>>> -- >>>> Sent from cellphone, sorry for the typos >>>> >>>> >>>> >>>> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >>>> >>>>> Hi, >>>>> >>>>> Qt Creator: 3.6.0 >>>>> >>>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>>> I don't believe that warning about padding should be the default behavior. >>>>> >>>>> -- >>>>> Best regards, >>>>> Andrzej Telszewski >>>>> _______________________________________________ >>>>> Qt-creator mailing list >>>>> Qt-creator at qt-project.org >>>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>> >>> >>> >>> -- >>> Best regards, >>> Andrzej Telszewski >> >> -- >> Sent from cellphone, sorry for the typos >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator >> > > > -- > Pozdrawiam, > Best regards, > Andrzej Telszewski From Marco.Bubke at theqtcompany.com Fri Mar 4 01:05:22 2016 From: Marco.Bubke at theqtcompany.com (Bubke Marco) Date: Fri, 4 Mar 2016 00:05:22 +0000 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: <56D8AB0E.7020505@gmail.com> References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> Message-ID: Can you file a bug report? -- Sent from cellphone, sorry for the typos On March 3, 2016 22:22:26 Andrzej Telszewski wrote: > On 03/03/16 21:55, Bubke Marco wrote: >> You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. >> >> I think it quite informative that >> >> bool x, y, z; >> int w; >> char c; >> >> Is not optimal. If you don't like that disable it in the settings. > > I do think that having this kind of information is useful, but in my > opinion -Wno-padded should be default, and turning on the warning should > be left to those who need it. I personally added -Wno-padded to the > clang model settings. > > Say you have a struct like that: > struct nopad > { > bool flag1; > bool flag2; > }; > > It is perfectly fine structure and you don't have to pad it, yet you get > the warning. I would understand, if the warning told you to re-arrange > the members, but it simply warns you about what is the default compiler > behavior, that it will add 2 or 6 bytes of padding. This has been the > default compiler behavior for ages, so I don't see the point of giving > the warning by default;) In addition, the padding depends on the > architecture you're compiling for, so you would have to take it into > account. > > I'm not against the warning, I'm just saying it shouldn't be enabled by > default. > > My real structure is like that: > typedef struct > { > string name; > bool can_span; > } info_key_t; > > There is nothing sensible I can change to make the warning disappear. > >> >> -- >> Sent from cellphone, sorry for the typos >> >> >> On March 3, 2016 20:44:28 Andrzej Telszewski wrote: >> >>> On 03/03/16 20:23, Bubke Marco wrote: >>>> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >>>> >>>> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >>>> >>> >>> Nope, it does not tell you it's suboptimal. It tells you that compiler >>> will add the additional padding/alignment bytes. >>> >>> Having structure with say 3 bool's and nothing more is perfectly fine. >>> In such a case, to silence the warning, I would have to add some dummy >>> fields, which most of the time is not necessary. >>> >>>> -- >>>> Sent from cellphone, sorry for the typos >>>> >>>> >>>> >>>> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >>>> >>>>> Hi, >>>>> >>>>> Qt Creator: 3.6.0 >>>>> >>>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>>> I don't believe that warning about padding should be the default behavior. >>>>> >>>>> -- >>>>> Best regards, >>>>> Andrzej Telszewski >>>>> _______________________________________________ >>>>> Qt-creator mailing list >>>>> Qt-creator at qt-project.org >>>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>> >>> >>> >>> -- >>> Best regards, >>> Andrzej Telszewski >> >> -- >> Sent from cellphone, sorry for the typos >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator >> > > > -- > Pozdrawiam, > Best regards, > Andrzej Telszewski From Eike.Ziller at theqtcompany.com Fri Mar 4 09:17:57 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Fri, 4 Mar 2016 08:17:57 +0000 Subject: [Qt-creator] Problems switching kit between 4.85 and 5.5 In-Reply-To: References: Message-ID: <02355088-B9C2-4B42-A6A3-F8303346F2A3@theqtcompany.com> > On Mar 3, 2016, at 5:42 PM, Duane wrote: > > QtCreator on Linux with Qt 4.8.5 and Qt 5.5.1 installed. We're moving our projects from Qt4 to Qt5. U > > I have QtCreator setup with a kit using each so I can support both for the moment. Recently when trying to build Qt4 projects I'm getting errors in files like qheaderview.h where it doesn't like setSectionsMovable() which replaced Qt4's setMovable(), so I can see it's getting the header. > > But if I look at the Makefile generated it seems to have the correct information. In the kit/version section I can see that the includes look correct. Any idea why QtCreator is not getting the correct path? Sorry, I do not understand what your actual problem is. So you get errors when building. Is that expected, or the problem? Where does _Qt Creator_ not get the correct path? (To what?) And what is the result of that? Br, Eike -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From christian.kandeler at theqtcompany.com Fri Mar 4 09:46:34 2016 From: christian.kandeler at theqtcompany.com (Christian Kandeler) Date: Fri, 4 Mar 2016 09:46:34 +0100 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: <56D8AB0E.7020505@gmail.com> References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> Message-ID: <56D94B6A.7080409@theqtcompany.com> On 03/03/2016 10:22 PM, Andrzej Telszewski wrote: > I do think that having this kind of information is useful, but in my > opinion -Wno-padded should be default, and turning on the warning should > be left to those who need it. I personally added -Wno-padded to the > clang model settings. > > Say you have a struct like that: > struct nopad > { > bool flag1; > bool flag2; > }; Not to mention that the warning becomes completely pointless with C++ classes containing virtual functions etc. In my experience, the false-positive rate is 100%. Christian From duane.hebert at group-upc.com Fri Mar 4 14:10:30 2016 From: duane.hebert at group-upc.com (Duane) Date: Fri, 4 Mar 2016 08:10:30 -0500 Subject: [Qt-creator] Problems switching kit between 4.85 and 5.5 In-Reply-To: <02355088-B9C2-4B42-A6A3-F8303346F2A3@theqtcompany.com> References: <02355088-B9C2-4B42-A6A3-F8303346F2A3@theqtcompany.com> Message-ID: On 04/03/2016 3:17 AM, Ziller Eike wrote: > >> On Mar 3, 2016, at 5:42 PM, Duane wrote: >> >> QtCreator on Linux with Qt 4.8.5 and Qt 5.5.1 installed. We're moving our projects from Qt4 to Qt5. U >> >> I have QtCreator setup with a kit using each so I can support both for the moment. Recently when trying to build Qt4 projects I'm getting errors in files like qheaderview.h where it doesn't like setSectionsMovable() which replaced Qt4's setMovable(), so I can see it's getting the header. >> >> But if I look at the Makefile generated it seems to have the correct information. In the kit/version section I can see that the includes look correct. Any idea why QtCreator is not getting the correct path? > > Sorry, I do not understand what your actual problem is. So you get errors when building. Is that expected, or the problem? Where does _Qt Creator_ not get the correct path? (To what?) And what is the result of that? > > Br, Eike > Sorry if I wasn't clear. I build with Qt5 kit and it's fine. I switch to the kit using Qt4 and I get errors like: /usr/local/include/QtWidgets/qheaderview.h:119: error: missing binary operator before token "(" And it's pointing to the line: ---> bool sectionsMovable() const; #if QT_DEPRECATED_SINCE(5, 0) inline QT_DEPRECATED void setMovable(bool movable) { setSectionsMovable(movable); } inline QT_DEPRECATED bool isMovable() const { return sectionsMovable(); } #endif But Qt4 doesn't have sectionsMovable() in qheaderview. My Qt installations are in /usr/local/Trolltech/Qt-5.5.1 and /usr/local/Trolltech/Qt-4.8.5. Why would it be looking in /usr/local/include in the first place. And why would I have these in the second place? They're only links to the Qt5 headers in the installations. This was working for quite a while. From duane.hebert at group-upc.com Fri Mar 4 14:25:40 2016 From: duane.hebert at group-upc.com (Duane) Date: Fri, 4 Mar 2016 08:25:40 -0500 Subject: [Qt-creator] Problems switching kit between 4.85 and 5.5 In-Reply-To: References: <02355088-B9C2-4B42-A6A3-F8303346F2A3@theqtcompany.com> Message-ID: On 04/03/2016 8:10 AM, Duane wrote: > On 04/03/2016 3:17 AM, Ziller Eike wrote: >> >>> On Mar 3, 2016, at 5:42 PM, Duane wrote: >>> >>> QtCreator on Linux with Qt 4.8.5 and Qt 5.5.1 installed. We're >>> moving our projects from Qt4 to Qt5. U >>> >>> I have QtCreator setup with a kit using each so I can support both >>> for the moment. Recently when trying to build Qt4 projects I'm >>> getting errors in files like qheaderview.h where it doesn't like >>> setSectionsMovable() which replaced Qt4's setMovable(), so I can see >>> it's getting the header. >>> >>> But if I look at the Makefile generated it seems to have the correct >>> information. In the kit/version section I can see that the includes >>> look correct. Any idea why QtCreator is not getting the correct path? >> >> Sorry, I do not understand what your actual problem is. So you get >> errors when building. Is that expected, or the problem? Where does _Qt >> Creator_ not get the correct path? (To what?) And what is the result >> of that? >> >> Br, Eike >> > > > Sorry if I wasn't clear. I build with Qt5 kit and it's fine. I switch > to the kit using Qt4 and I get errors like: > > /usr/local/include/QtWidgets/qheaderview.h:119: error: missing binary > operator before token "(" > > And it's pointing to the line: > > ---> bool sectionsMovable() const; > #if QT_DEPRECATED_SINCE(5, 0) > inline QT_DEPRECATED void setMovable(bool movable) { > setSectionsMovable(movable); } > inline QT_DEPRECATED bool isMovable() const { return > sectionsMovable(); } > #endif > > > But Qt4 doesn't have sectionsMovable() in qheaderview. > > My Qt installations are in /usr/local/Trolltech/Qt-5.5.1 and > /usr/local/Trolltech/Qt-4.8.5. > > Why would it be looking in /usr/local/include in the first place. And > why would I have these in the second place? They're only links to the > Qt5 headers in the installations. This was working for quite a while. The Qt header files in /usr/local/include are links to Qt5.3 that was used to build QtCreator. We're using Tiny Core Linux and had to build this ourselves. This was working. I'm still not seeing why switching the kit between builds isn't picking the correct include path. From Eike.Ziller at theqtcompany.com Fri Mar 4 15:01:45 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Fri, 4 Mar 2016 14:01:45 +0000 Subject: [Qt-creator] Problems switching kit between 4.85 and 5.5 In-Reply-To: References: <02355088-B9C2-4B42-A6A3-F8303346F2A3@theqtcompany.com> Message-ID: <35866F43-3313-4CD2-8E7E-B11B2A50A883@theqtcompany.com> > On Mar 4, 2016, at 2:25 PM, Duane wrote: > > On 04/03/2016 8:10 AM, Duane wrote: >> On 04/03/2016 3:17 AM, Ziller Eike wrote: >>> >>>> On Mar 3, 2016, at 5:42 PM, Duane wrote: >>>> >>>> QtCreator on Linux with Qt 4.8.5 and Qt 5.5.1 installed. We're >>>> moving our projects from Qt4 to Qt5. U >>>> >>>> I have QtCreator setup with a kit using each so I can support both >>>> for the moment. Recently when trying to build Qt4 projects I'm >>>> getting errors in files like qheaderview.h where it doesn't like >>>> setSectionsMovable() which replaced Qt4's setMovable(), so I can see >>>> it's getting the header. >>>> >>>> But if I look at the Makefile generated it seems to have the correct >>>> information. In the kit/version section I can see that the includes >>>> look correct. Any idea why QtCreator is not getting the correct path? >>> >>> Sorry, I do not understand what your actual problem is. So you get >>> errors when building. Is that expected, or the problem? Where does _Qt >>> Creator_ not get the correct path? (To what?) And what is the result >>> of that? >>> >>> Br, Eike >>> >> >> >> Sorry if I wasn't clear. I build with Qt5 kit and it's fine. I switch >> to the kit using Qt4 and I get errors like: >> >> /usr/local/include/QtWidgets/qheaderview.h:119: error: missing binary >> operator before token "(" >> >> And it's pointing to the line: >> >> ---> bool sectionsMovable() const; >> #if QT_DEPRECATED_SINCE(5, 0) >> inline QT_DEPRECATED void setMovable(bool movable) { >> setSectionsMovable(movable); } >> inline QT_DEPRECATED bool isMovable() const { return >> sectionsMovable(); } >> #endif >> >> >> But Qt4 doesn't have sectionsMovable() in qheaderview. >> >> My Qt installations are in /usr/local/Trolltech/Qt-5.5.1 and >> /usr/local/Trolltech/Qt-4.8.5. >> >> Why would it be looking in /usr/local/include in the first place. And >> why would I have these in the second place? They're only links to the >> Qt5 headers in the installations. This was working for quite a while. > > The Qt header files in /usr/local/include are links to Qt5.3 that was used to build QtCreator. We're using Tiny Core Linux and had to build this ourselves. This was working. I'm still not seeing why switching the kit between builds isn't picking the correct include path. What’s the compiler line that produces that error? And in which file is the qheaderview.h included (which then resolves to the wrong one)? Br, Eike -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From atelszewski at gmail.com Fri Mar 4 17:28:12 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Fri, 4 Mar 2016 17:28:12 +0100 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> Message-ID: <56D9B79C.9080109@gmail.com> On 04/03/16 01:05, Bubke Marco wrote: > Can you file a bug report? But is it really a bug? Have a look at the attached image. If I add -Wno-padded, then the warning disappears. In my opinion, adding -Wno-padded to the default flags would be sufficient. > > -- > Sent from cellphone, sorry for the typos > > > > On March 3, 2016 22:22:26 Andrzej Telszewski wrote: > >> On 03/03/16 21:55, Bubke Marco wrote: >>> You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. >>> >>> I think it quite informative that >>> >>> bool x, y, z; >>> int w; >>> char c; >>> >>> Is not optimal. If you don't like that disable it in the settings. >> >> I do think that having this kind of information is useful, but in my >> opinion -Wno-padded should be default, and turning on the warning should >> be left to those who need it. I personally added -Wno-padded to the >> clang model settings. >> >> Say you have a struct like that: >> struct nopad >> { >> bool flag1; >> bool flag2; >> }; >> >> It is perfectly fine structure and you don't have to pad it, yet you get >> the warning. I would understand, if the warning told you to re-arrange >> the members, but it simply warns you about what is the default compiler >> behavior, that it will add 2 or 6 bytes of padding. This has been the >> default compiler behavior for ages, so I don't see the point of giving >> the warning by default;) In addition, the padding depends on the >> architecture you're compiling for, so you would have to take it into >> account. >> >> I'm not against the warning, I'm just saying it shouldn't be enabled by >> default. >> >> My real structure is like that: >> typedef struct >> { >> string name; >> bool can_span; >> } info_key_t; >> >> There is nothing sensible I can change to make the warning disappear. >> >>> >>> -- >>> Sent from cellphone, sorry for the typos >>> >>> >>> On March 3, 2016 20:44:28 Andrzej Telszewski wrote: >>> >>>> On 03/03/16 20:23, Bubke Marco wrote: >>>>> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >>>>> >>>>> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >>>>> >>>> >>>> Nope, it does not tell you it's suboptimal. It tells you that compiler >>>> will add the additional padding/alignment bytes. >>>> >>>> Having structure with say 3 bool's and nothing more is perfectly fine. >>>> In such a case, to silence the warning, I would have to add some dummy >>>> fields, which most of the time is not necessary. >>>> >>>>> -- >>>>> Sent from cellphone, sorry for the typos >>>>> >>>>> >>>>> >>>>> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> Qt Creator: 3.6.0 >>>>>> >>>>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>>>> I don't believe that warning about padding should be the default behavior. >>>>>> >>>>>> -- >>>>>> Best regards, >>>>>> Andrzej Telszewski >>>>>> _______________________________________________ >>>>>> Qt-creator mailing list >>>>>> Qt-creator at qt-project.org >>>>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>>> >>>> >>>> >>>> -- >>>> Best regards, >>>> Andrzej Telszewski >>> >>> -- >>> Sent from cellphone, sorry for the typos >>> _______________________________________________ >>> Qt-creator mailing list >>> Qt-creator at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>> >> >> >> -- >> Pozdrawiam, >> Best regards, >> Andrzej Telszewski > . > -- Pozdrawiam, Best regards, Andrzej Telszewski -------------- next part -------------- A non-text attachment was scrubbed... Name: struct_padding.png Type: image/png Size: 11196 bytes Desc: not available URL: From atelszewski at gmail.com Fri Mar 4 17:30:23 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Fri, 4 Mar 2016 17:30:23 +0100 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: <56D94B6A.7080409@theqtcompany.com> References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> <56D94B6A.7080409@theqtcompany.com> Message-ID: <56D9B81F.80600@gmail.com> On 04/03/16 09:46, Christian Kandeler wrote: > On 03/03/2016 10:22 PM, Andrzej Telszewski wrote: >> I do think that having this kind of information is useful, but in my >> opinion -Wno-padded should be default, and turning on the warning should >> be left to those who need it. I personally added -Wno-padded to the >> clang model settings. >> >> Say you have a struct like that: >> struct nopad >> { >> bool flag1; >> bool flag2; >> }; > > Not to mention that the warning becomes completely pointless with C++ > classes containing virtual functions etc. In my experience, the > false-positive rate is 100%. > Do you see the warning for classes? In my case, it only appears for structs. > > Christian > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Best regards, Andrzej Telszewski From realnc at gmail.com Fri Mar 4 18:05:54 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Fri, 4 Mar 2016 19:05:54 +0200 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> Message-ID: This happens for me too, on Clang 3.7.1. I never complained about it though, since it's dead-easy to just add -wno-padded to the Clang model options. On 04/03/16 02:03, Bubke Marco wrote: > I have similar structures and have never seen the warning for that cases. Which clang version do you use? > > -- > Sent from cellphone, sorry for the typos > > > > On March 3, 2016 22:22:26 Andrzej Telszewski wrote: > >> On 03/03/16 21:55, Bubke Marco wrote: >>> You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. >>> >>> I think it quite informative that >>> >>> bool x, y, z; >>> int w; >>> char c; >>> >>> Is not optimal. If you don't like that disable it in the settings. >> >> I do think that having this kind of information is useful, but in my >> opinion -Wno-padded should be default, and turning on the warning should >> be left to those who need it. I personally added -Wno-padded to the >> clang model settings. >> >> Say you have a struct like that: >> struct nopad >> { >> bool flag1; >> bool flag2; >> }; >> >> It is perfectly fine structure and you don't have to pad it, yet you get >> the warning. I would understand, if the warning told you to re-arrange >> the members, but it simply warns you about what is the default compiler >> behavior, that it will add 2 or 6 bytes of padding. This has been the >> default compiler behavior for ages, so I don't see the point of giving >> the warning by default;) In addition, the padding depends on the >> architecture you're compiling for, so you would have to take it into >> account. >> >> I'm not against the warning, I'm just saying it shouldn't be enabled by >> default. >> >> My real structure is like that: >> typedef struct >> { >> string name; >> bool can_span; >> } info_key_t; >> >> There is nothing sensible I can change to make the warning disappear. >> >>> >>> -- >>> Sent from cellphone, sorry for the typos >>> >>> >>> On March 3, 2016 20:44:28 Andrzej Telszewski wrote: >>> >>>> On 03/03/16 20:23, Bubke Marco wrote: >>>>> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >>>>> >>>>> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >>>>> >>>> >>>> Nope, it does not tell you it's suboptimal. It tells you that compiler >>>> will add the additional padding/alignment bytes. >>>> >>>> Having structure with say 3 bool's and nothing more is perfectly fine. >>>> In such a case, to silence the warning, I would have to add some dummy >>>> fields, which most of the time is not necessary. >>>> >>>>> -- >>>>> Sent from cellphone, sorry for the typos >>>>> >>>>> >>>>> >>>>> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> Qt Creator: 3.6.0 >>>>>> >>>>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>>>> I don't believe that warning about padding should be the default behavior. >>>>>> >>>>>> -- >>>>>> Best regards, >>>>>> Andrzej Telszewski >>>>>> _______________________________________________ >>>>>> Qt-creator mailing list >>>>>> Qt-creator at qt-project.org >>>>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>>> >>>> >>>> >>>> -- >>>> Best regards, >>>> Andrzej Telszewski >>> >>> -- >>> Sent from cellphone, sorry for the typos >>> _______________________________________________ >>> Qt-creator mailing list >>> Qt-creator at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>> >> >> >> -- >> Pozdrawiam, >> Best regards, >> Andrzej Telszewski From imikejackson at gmail.com Fri Mar 4 19:42:59 2016 From: imikejackson at gmail.com (Mike Jackson) Date: Fri, 04 Mar 2016 13:42:59 -0500 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go Message-ID: <56D9D733.5080608@gmail.com> I just compiled QtCreator 4.0 and what a surprise. It seems as if the QtCreator team seems to have followed everyone else in UI design lately and tossed out 30 years of Human Computer Interaction design. Namely, you tossed colorful icons to go with a flat mono-chrome color palette. I do not understand where this comes from? We programmers have been fighting and waiting for years to get able to have high res icons that can convey information quickly and when we get that capability all the UI designers seem to have just given up on the concept. Color conveys information quickly. It lets the human brain decipher information with less effort than a monochrome palette. Numerous experts in the field have written at length about this issue and where the UI design is going and just how wrong it is. I urge the QtCreator team to at least include a theme that retains the colorful icons from QtCreator 3.x in the 4.0 release. It is not too late. Let me put forth a very direct example. If you drive a white car and you park that car in a parking lot with 1000 other white cars of the same type (say 4 door passenger style cars), when you come back out to get in your car, unless you parked your car in a very specific spot (first or last) your brain needs to work much hard to remember exactly which car is yours. Now imagine your car is red, in that same lot of 1000 white cars. Finding your car is instantaneous. Thanks for listening. -- Mike Jackson -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Sat Mar 5 00:44:53 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sat, 5 Mar 2016 01:44:53 +0200 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: <56D9D733.5080608@gmail.com> References: <56D9D733.5080608@gmail.com> Message-ID: What, you're using icons instead of doing everything with keyboard shortcuts? :-P On 04/03/16 20:42, Mike Jackson wrote: > > I just compiled QtCreator 4.0 and what a surprise. It seems as if the > QtCreator team seems to have followed everyone else in UI design lately > and tossed out 30 years of Human Computer Interaction design. Namely, > you tossed colorful icons to go with a flat mono-chrome color palette. I > do not understand where this comes from? We programmers have been > fighting and waiting for years to get able to have high res icons that > can convey information quickly and when we get that capability all the > UI designers seem to have just given up on the concept. Color conveys > information quickly. It lets the human brain decipher information with > less effort than a monochrome palette. Numerous experts in the field > have written at length about this issue and where the UI design is going > and just how wrong it is. I urge the QtCreator team to at least include > a theme that retains the colorful icons from QtCreator 3.x in the 4.0 > release. It is not too late. > > > Let me put forth a very direct example. If you drive a white car and you > park that car in a parking lot with 1000 other white cars of the same > type (say 4 door passenger style cars), when you come back out to get in > your car, unless you parked your car in a very specific spot (first or > last) your brain needs to work much hard to remember exactly which car > is yours. Now imagine your car is red, in that same lot of 1000 white > cars. Finding your car is instantaneous. > > Thanks for listening. From mail at milianw.de Sat Mar 5 16:03:18 2016 From: mail at milianw.de (Milian Wolff) Date: Sat, 05 Mar 2016 16:03:18 +0100 Subject: [Qt-creator] Very slow startup / autodetecting hardening versions of gcc and clang In-Reply-To: References: <56029E19.9010005@fh-kl.de> Message-ID: <1681395.JQLAlha4rp@agathebauer> On Mittwoch, 23. September 2015 15:51:22 CET Tobias Hunger wrote: > Hi Wilhelm, > > On Wed, Sep 23, 2015 at 2:42 PM, Wilhelm wrote: > > I'm using qtcreator 3.5.0 with qt 5.5.0 on arch-linux. > > > > I experience very long startup times running qtcreator. Starting from > > > bash results in: > > > > QByteArray ProjectExplorer::runGcc(const Utils::FileName&, const > > QStringList&, const QStringList&): Timeout running > > '/usr/lib/hardening-wrapper/bin/clang++'. > > > > Questions: > > > > 1) What may be the cause to get the timeouts? > > The hardening-wrapper package. > > /usr/lib/hardening-wrapper/bin/g++ does not like the arguments we pass > to it: E.g. "-xc++ -E -dM /dev/null" does not produce any output, so > the code model would be broken for this compiler -- if it would have > been detected properly). > > Please report a bug against the tool: That wrapper should not > interfere with the output of the wrapped application in such a way. When I try to launch Qt Creator I'm also suffering from this issue on ArchLinux. But this works as expected, and indeed KDevelop has never had any issues with it: echo | /usr/lib/hardening-wrapper/bin/g++ -xc++ -E -dM - ... Compiling also works as expected. So I really think that something Qt Creator does is not correct... Looking at strace, I see that Qt Creator starts with the above and then keeps recursing into things like /usr/lib/hardening-wrapper/bin/g++ -B/usr/lib/hardening-wrapper/bin -fstack- check -fstack-protector-strong -B/usr/lib/hardening-wrapper/bin -fPIE -fstack- check -fstack-protector-strong -xc++ -E -dM - Are you tinkering with the PATH inside Qt Creator? Note again that this works as expected on the CLI and for KDevelop as well. Cheers -- Milian Wolff mail at milianw.de http://milianw.de -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part. URL: From Stefan.Walter at lisec.com Sun Mar 6 07:47:00 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Sun, 6 Mar 2016 06:47:00 +0000 Subject: [Qt-creator] Determine if started through Qt Creator Message-ID: <0fd43fa0dd9e448782d75d75559c96ec@ATSE-MAIL4.lisec.internal> Hi, I would like to determine in my application, if it was started through Qt Creator. My code should behave slightly different between running from Qt-Creator (error handling) or started directly from a terminal. What is the best approach for this? Regards, Stefan -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Sun Mar 6 11:05:47 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 6 Mar 2016 12:05:47 +0200 Subject: [Qt-creator] Determine if started through Qt Creator In-Reply-To: <0fd43fa0dd9e448782d75d75559c96ec@ATSE-MAIL4.lisec.internal> References: <0fd43fa0dd9e448782d75d75559c96ec@ATSE-MAIL4.lisec.internal> Message-ID: On 06/03/16 08:47, Walter Stefan wrote: > Hi, > > I would like to determine in my application, if it was started through > Qt Creator. > > My code should behave slightly different between running from Qt-Creator > (error handling) or started directly from a terminal. > > What is the best approach for this? In the project settings, you can modify the environment. Choose the "run" configuration, and just add an environment variable. Name it whatever you want, like IN_QTCREATOR. In your program, just use: qEnvironmentVariableIsSet("IN_QTCREATOR"); (http://doc.qt.io/qt-5/qtglobal.html#qEnvironmentVariableIsSet) to detect if the variable is set. Obviously, it only going to be set if you run the program in Creator. From Stefan.Walter at lisec.com Sun Mar 6 11:19:14 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Sun, 6 Mar 2016 10:19:14 +0000 Subject: [Qt-creator] Determine if started through Qt Creator In-Reply-To: References: <0fd43fa0dd9e448782d75d75559c96ec@ATSE-MAIL4.lisec.internal> Message-ID: <632cf818c41241ceb762206d466f128a@ATSE-MAIL4.lisec.internal> Thanks, but is there not a more generic way. I don't want to set this in every single project. Regards, Stefan -----Original Message----- From: Qt-creator [mailto:qt-creator-bounces+stefan.walter=lisec.com at qt-project.org] On Behalf Of Nikos Chantziaras Sent: Sonntag, 6. März 2016 14:06 To: qt-creator at qt-project.org Subject: Re: [Qt-creator] Determine if started through Qt Creator On 06/03/16 08:47, Walter Stefan wrote: > Hi, > > I would like to determine in my application, if it was started through > Qt Creator. > > My code should behave slightly different between running from > Qt-Creator (error handling) or started directly from a terminal. > > What is the best approach for this? In the project settings, you can modify the environment. Choose the "run" configuration, and just add an environment variable. Name it whatever you want, like IN_QTCREATOR. In your program, just use: qEnvironmentVariableIsSet("IN_QTCREATOR"); (http://doc.qt.io/qt-5/qtglobal.html#qEnvironmentVariableIsSet) to detect if the variable is set. Obviously, it only going to be set if you run the program in Creator. _______________________________________________ Qt-creator mailing list Qt-creator at qt-project.org http://lists.qt-project.org/mailman/listinfo/qt-creator From tobias.hunger at gmail.com Sun Mar 6 12:25:40 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Sun, 6 Mar 2016 12:25:40 +0100 Subject: [Qt-creator] Determine if started through Qt Creator In-Reply-To: <632cf818c41241ceb762206d466f128a@ATSE-MAIL4.lisec.internal> References: <0fd43fa0dd9e448782d75d75559c96ec@ATSE-MAIL4.lisec.internal> <632cf818c41241ceb762206d466f128a@ATSE-MAIL4.lisec.internal> Message-ID: Hello Stefan, On Sun, Mar 6, 2016 at 11:19 AM, Walter Stefan wrote: > Thanks, but is there not a more generic way. I don't want to set this in every single project. We try hard to make sure there is *no* difference between running a application in Qt Creator and outside of Qt Creator:-) If you need one, then you need to define your own conventions. Setting a environment variable is one convenient way to do that, just as Nikos suggested. You can set those variables in build-/run settings of the project itself. This offers a very fine-grained control over what is run in "QtCreator mode" and what not. You can also defined environment variables on our kits. That way they will be set by default in all projects that use those kits. You can set the variable before starting Qt Creator (e.g. in a script/bat-file) and the variable will be handed down to any build/run using the system environment. By default all projects are built using a slightly modified systems environment and will then use the same environment to run the applications later. There are probably more ways to accomplish what you want. You are a programmer, you will figure out the best approach for your individual setup:-) Best Regards, Tobias From Stefan.Walter at lisec.com Sun Mar 6 12:41:34 2016 From: Stefan.Walter at lisec.com (Walter Stefan) Date: Sun, 6 Mar 2016 11:41:34 +0000 Subject: [Qt-creator] Determine if started through Qt Creator In-Reply-To: References: <0fd43fa0dd9e448782d75d75559c96ec@ATSE-MAIL4.lisec.internal> <632cf818c41241ceb762206d466f128a@ATSE-MAIL4.lisec.internal> Message-ID: <28cb12fb338544c2b68158779fe39081@ATSE-MAIL4.lisec.internal> Dear Tobias, I was also thinking as a workaround to use a startup script that defines a environment variable. I was just asking for a probably existing out of the box solution, instead of inventing my own solution. Thanks and Regard, Stefan -----Original Message----- From: Tobias Hunger [mailto:tobias.hunger at gmail.com] Sent: Sonntag, 6. März 2016 15:26 To: Walter Stefan Cc: Nikos Chantziaras ; qt-creator at qt-project.org Subject: Re: [Qt-creator] Determine if started through Qt Creator Hello Stefan, On Sun, Mar 6, 2016 at 11:19 AM, Walter Stefan wrote: > Thanks, but is there not a more generic way. I don't want to set this in every single project. We try hard to make sure there is *no* difference between running a application in Qt Creator and outside of Qt Creator:-) If you need one, then you need to define your own conventions. Setting a environment variable is one convenient way to do that, just as Nikos suggested. You can set those variables in build-/run settings of the project itself. This offers a very fine-grained control over what is run in "QtCreator mode" and what not. You can also defined environment variables on our kits. That way they will be set by default in all projects that use those kits. You can set the variable before starting Qt Creator (e.g. in a script/bat-file) and the variable will be handed down to any build/run using the system environment. By default all projects are built using a slightly modified systems environment and will then use the same environment to run the applications later. There are probably more ways to accomplish what you want. You are a programmer, you will figure out the best approach for your individual setup:-) Best Regards, Tobias From hkarel at yandex.ru Sun Mar 6 18:15:22 2016 From: hkarel at yandex.ru (=?UTF-8?B?0JrQsNGA0LXQu9C40L0g0J/QsNCy0LXQuw==?=) Date: Sun, 6 Mar 2016 20:15:22 +0300 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode Message-ID: <56DC65AA.9060903@yandex.ru> Hello, My project consists of a client and a server parts. The client and the server interact under the TCP protocol. How I can run the client and the server in the debug mode in QtCreator? -- Best regards, Pavel Karelin -------------- next part -------------- An HTML attachment was scrubbed... URL: From bogdan.vatra at kdab.com Sun Mar 6 19:20:08 2016 From: bogdan.vatra at kdab.com (Bogdan Vatra) Date: Sun, 06 Mar 2016 20:20:08 +0200 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <56DC65AA.9060903@yandex.ru> References: <56DC65AA.9060903@yandex.ru> Message-ID: <3929879.VM2KY5d8d8@zmeu> Hi, I think you'll need to run two QtCreator instances. Cheers, BogDan. On Sunday 06 March 2016 20:15:22 Карелин Павел wrote: > Hello, > > My project consists of a client and a server parts. The client and the > server interact under the TCP protocol. How I can run the client and the > server in the debug mode in QtCreator? From hkarel at yandex.ru Sun Mar 6 19:23:38 2016 From: hkarel at yandex.ru (=?UTF-8?B?0JrQsNGA0LXQu9C40L0g0J/QsNCy0LXQuw==?=) Date: Sun, 6 Mar 2016 21:23:38 +0300 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <3929879.VM2KY5d8d8@zmeu> References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> Message-ID: <56DC75AA.2010905@yandex.ru> 06.03.2016 21:20, Bogdan Vatra пишет: > Hi, > > I think you'll need to run two QtCreator instances. I tried it, got this error: "Cannot lock build graph file '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': Already locked by 'QtCreator' (PID 4475)." > > Cheers, > BogDan. > > On Sunday 06 March 2016 20:15:22 Карелин Павел wrote: >> Hello, >> >> My project consists of a client and a server parts. The client and the >> server interact under the TCP protocol. How I can run the client and the >> server in the debug mode in QtCreator? -------------- next part -------------- An HTML attachment was scrubbed... URL: From atelszewski at gmail.com Sun Mar 6 20:10:06 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Sun, 6 Mar 2016 20:10:06 +0100 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <56DC75AA.2010905@yandex.ru> References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> <56DC75AA.2010905@yandex.ru> Message-ID: <56DC808E.2020603@gmail.com> On 06/03/16 19:23, Карелин Павел wrote: > > > 06.03.2016 21:20, Bogdan Vatra пишет: >> Hi, >> >> I think you'll need to run two QtCreator instances. > I tried it, got this error: "Cannot lock build graph file > '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': > Already locked by 'QtCreator' (PID 4475)." Is the server and client the same binary? If so, I think you'd have to separate them and then you could run two instances. >> >> Cheers, >> BogDan. >> >> On Sunday 06 March 2016 20:15:22 Карелин Павел wrote: >>> Hello, >>> >>> My project consists of a client and a server parts. The client and the >>> server interact under the TCP protocol. How I can run the client and the >>> server in the debug mode in QtCreator? > > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Best regards, Andrzej Telszewski From atelszewski at gmail.com Sun Mar 6 20:19:43 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Sun, 6 Mar 2016 20:19:43 +0100 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <56DC808E.2020603@gmail.com> References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> <56DC75AA.2010905@yandex.ru> <56DC808E.2020603@gmail.com> Message-ID: <56DC82CF.1090701@gmail.com> On 06/03/16 20:10, Andrzej Telszewski wrote: > On 06/03/16 19:23, Карелин Павел wrote: >> >> >> 06.03.2016 21:20, Bogdan Vatra пишет: >>> Hi, >>> >>> I think you'll need to run two QtCreator instances. >> I tried it, got this error: "Cannot lock build graph file >> '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': >> >> Already locked by 'QtCreator' (PID 4475)." > > Is the server and client the same binary? > If so, I think you'd have to separate them and then you could run two > instances. > Or maybe another solution. Under "Debug" -> "Start Debugging" you have various possibilities. Maybe you could "Attach to Running Application" provided that you have started your application outside of Qt Creator? I don't know how the locking is going to behave, though. >>> >>> Cheers, >>> BogDan. >>> >>> On Sunday 06 March 2016 20:15:22 Карелин Павел wrote: >>>> Hello, >>>> >>>> My project consists of a client and a server parts. The client and the >>>> server interact under the TCP protocol. How I can run the client and >>>> the >>>> server in the debug mode in QtCreator? >> >> >> >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator >> > -- Best regards, Andrzej Telszewski From orgads at gmail.com Sun Mar 6 22:29:46 2016 From: orgads at gmail.com (Orgad Shaneh) Date: Sun, 6 Mar 2016 23:29:46 +0200 Subject: [Qt-creator] Incorrect version reported for Auto-detected system debugger after debugger update In-Reply-To: <56D1D1A3.7020002@gmail.com> References: <56D1D1A3.7020002@gmail.com> Message-ID: On Sat, Feb 27, 2016 at 6:41 PM, Andrzej Telszewski wrote: > Hi, > > Qt Creator: 3.5.1 > OS: Linux > > I updated GDB and went to Build&Run -> Debuggers to see if Qt Creator > picked the change, and it hasn't. For Auto-detected, it reported the > version of the old version of GDB. > > Removing debuggers.xml triggers the auto-detection and this time the > reported version is correct. (After that you need to reassign the new > debugger to the kits, so it's not the best approach). > > Just for your information. Thanks for the report. Fix pending: https://codereview.qt-project.org/151520 - Orgad -------------- next part -------------- An HTML attachment was scrubbed... URL: From hkarel at yandex.ru Sun Mar 6 22:47:36 2016 From: hkarel at yandex.ru (=?UTF-8?B?0JrQsNGA0LXQu9C40L0g0J/QsNCy0LXQuw==?=) Date: Mon, 7 Mar 2016 00:47:36 +0300 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <56DC82CF.1090701@gmail.com> References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> <56DC75AA.2010905@yandex.ru> <56DC808E.2020603@gmail.com> <56DC82CF.1090701@gmail.com> Message-ID: <56DCA578.9000101@yandex.ru> 06.03.2016 22:19, Andrzej Telszewski пишет: > On 06/03/16 20:10, Andrzej Telszewski wrote: >> On 06/03/16 19:23, Карелин Павел wrote: >>> >>> >>> 06.03.2016 21:20, Bogdan Vatra пишет: >>>> Hi, >>>> >>>> I think you'll need to run two QtCreator instances. >>> I tried it, got this error: "Cannot lock build graph file >>> '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': >>> >>> >>> Already locked by 'QtCreator' (PID 4475)." >> >> Is the server and client the same binary? >> If so, I think you'd have to separate them and then you could run two >> instances. Client and server are separated >> > > Or maybe another solution. Under "Debug" -> "Start Debugging" you have > various possibilities. Maybe you could "Attach to Running Application" > provided that you have started your application outside of Qt Creator? I tried to attach to running application, but debbuger immediately quits > > I don't know how the locking is going to behave, though. In principle, it would suit me a running the client and server parts of my application in two different instances of QtCreator. But I can not open the same project in two different instances of QtCreator. > >>>> >>>> Cheers, >>>> BogDan. >>>> >>>> On Sunday 06 March 2016 20:15:22 Карелин Павел wrote: >>>>> Hello, >>>>> >>>>> My project consists of a client and a server parts. The client >>>>> and the >>>>> server interact under the TCP protocol. How I can run the client and >>>>> the >>>>> server in the debug mode in QtCreator? >>> >>> >>> >>> _______________________________________________ >>> Qt-creator mailing list >>> Qt-creator at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>> >> > From orgads at gmail.com Sun Mar 6 22:52:32 2016 From: orgads at gmail.com (Orgad Shaneh) Date: Sun, 6 Mar 2016 23:52:32 +0200 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <56DCA578.9000101@yandex.ru> References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> <56DC75AA.2010905@yandex.ru> <56DC808E.2020603@gmail.com> <56DC82CF.1090701@gmail.com> <56DCA578.9000101@yandex.ru> Message-ID: בתאריך 6 במרץ 2016 23:47,‏ "Карелин Павел" כתב: > > > > 06.03.2016 22:19, Andrzej Telszewski пишет: > >> On 06/03/16 20:10, Andrzej Telszewski wrote: >>> >>> On 06/03/16 19:23, Карелин Павел wrote: >>>> >>>> >>>> >>>> 06.03.2016 21:20, Bogdan Vatra пишет: >>>>> >>>>> Hi, >>>>> >>>>> I think you'll need to run two QtCreator instances. >>>> >>>> I tried it, got this error: "Cannot lock build graph file >>>> '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': >>>> >>>> Already locked by 'QtCreator' (PID 4475)." >>> >>> >>> Is the server and client the same binary? >>> If so, I think you'd have to separate them and then you could run two >>> instances. > > Client and server are separated > >>> >> >> Or maybe another solution. Under "Debug" -> "Start Debugging" you have various possibilities. Maybe you could "Attach to Running Application" provided that you have started your application outside of Qt Creator? > > I tried to attach to running application, but debbuger immediately quits > >> >> I don't know how the locking is going to behave, though. > > In principle, it would suit me a running the client and server parts of my application in two different instances of QtCreator. But I can not open the same project in two different instances of QtCreator. Why do they have to be in the same project? Use 2 projects and 2 instances. - Orgad -------------- next part -------------- An HTML attachment was scrubbed... URL: From hkarel at yandex.ru Sun Mar 6 23:07:29 2016 From: hkarel at yandex.ru (=?UTF-8?B?0JrQsNGA0LXQu9C40L0g0J/QsNCy0LXQuw==?=) Date: Mon, 7 Mar 2016 01:07:29 +0300 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> <56DC75AA.2010905@yandex.ru> <56DC808E.2020603@gmail.com> <56DC82CF.1090701@gmail.com> <56DCA578.9000101@yandex.ru> Message-ID: <56DCAA21.40806@yandex.ru> 07.03.2016 00:52, Orgad Shaneh пишет: > > > בתאריך 6 במרץ 2016 23:47,‏ "Карелин Павел" > כתב: > > > > > > > > 06.03.2016 22:19, Andrzej Telszewski пишет: > > > >> On 06/03/16 20:10, Andrzej Telszewski wrote: > >>> > >>> On 06/03/16 19:23, Карелин Павел wrote: > >>>> > >>>> > >>>> > >>>> 06.03.2016 21:20, Bogdan Vatra пишет: > >>>>> > >>>>> Hi, > >>>>> > >>>>> I think you'll need to run two QtCreator instances. > >>>> > >>>> I tried it, got this error: "Cannot lock build graph file > >>>> > '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': > > >>>> > >>>> Already locked by 'QtCreator' (PID 4475)." > >>> > >>> > >>> Is the server and client the same binary? > >>> If so, I think you'd have to separate them and then you could run two > >>> instances. > > > > Client and server are separated > > > >>> > >> > >> Or maybe another solution. Under "Debug" -> "Start Debugging" you > have various possibilities. Maybe you could "Attach to Running > Application" provided that you have started your application outside > of Qt Creator? > > > > I tried to attach to running application, but debbuger immediately quits > > > >> > >> I don't know how the locking is going to behave, though. > > > > In principle, it would suit me a running the client and server parts > of my application in two different instances of QtCreator. But I can > not open the same project in two different instances of QtCreator. > > Why do they have to be in the same project? Use 2 projects and 2 > instances. > > - Orgad > Because in project I have a client, a server and shared libraries. I would like to see it as a single solution in the development and debugging. But judging from your question, I have to invent a surrogate solution for debugging. -- Pavel -------------- next part -------------- An HTML attachment was scrubbed... URL: From Eike.Ziller at theqtcompany.com Mon Mar 7 08:36:52 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Mon, 7 Mar 2016 07:36:52 +0000 Subject: [Qt-creator] Clang Code Model: add -Wno-padded In-Reply-To: <56D9B79C.9080109@gmail.com> References: <56D882BE.4060006@gmail.com> <56D8941B.7020801@gmail.com> <56D8AB0E.7020505@gmail.com> <56D9B79C.9080109@gmail.com> Message-ID: > On Mar 4, 2016, at 5:28 PM, Andrzej Telszewski wrote: > > On 04/03/16 01:05, Bubke Marco wrote: >> Can you file a bug report? > > But is it really a bug? bugreports.qt.io is used for all kinds of issues, be it bugs, suggestions, feature requests, ... Br, Eike > Have a look at the attached image. > > If I add -Wno-padded, then the warning disappears. > > In my opinion, adding -Wno-padded to the default flags would be sufficient. > >> >> -- >> Sent from cellphone, sorry for the typos >> >> >> >> On March 3, 2016 22:22:26 Andrzej Telszewski wrote: >> >>> On 03/03/16 21:55, Bubke Marco wrote: >>>> You don't have to silence the warning, it is a warning not a error. And you can change that in the settings. >>>> >>>> I think it quite informative that >>>> >>>> bool x, y, z; >>>> int w; >>>> char c; >>>> >>>> Is not optimal. If you don't like that disable it in the settings. >>> >>> I do think that having this kind of information is useful, but in my >>> opinion -Wno-padded should be default, and turning on the warning should >>> be left to those who need it. I personally added -Wno-padded to the >>> clang model settings. >>> >>> Say you have a struct like that: >>> struct nopad >>> { >>> bool flag1; >>> bool flag2; >>> }; >>> >>> It is perfectly fine structure and you don't have to pad it, yet you get >>> the warning. I would understand, if the warning told you to re-arrange >>> the members, but it simply warns you about what is the default compiler >>> behavior, that it will add 2 or 6 bytes of padding. This has been the >>> default compiler behavior for ages, so I don't see the point of giving >>> the warning by default;) In addition, the padding depends on the >>> architecture you're compiling for, so you would have to take it into >>> account. >>> >>> I'm not against the warning, I'm just saying it shouldn't be enabled by >>> default. >>> >>> My real structure is like that: >>> typedef struct >>> { >>> string name; >>> bool can_span; >>> } info_key_t; >>> >>> There is nothing sensible I can change to make the warning disappear. >>> >>>> >>>> -- >>>> Sent from cellphone, sorry for the typos >>>> >>>> >>>> On March 3, 2016 20:44:28 Andrzej Telszewski wrote: >>>> >>>>> On 03/03/16 20:23, Bubke Marco wrote: >>>>>> Hi, we will add more options in 4.0 to the code model, so you can choose different sets. >>>>>> >>>>>> Anyway I find the padding warning quite useful because it shows that your data structures are sub optimal. >>>>>> >>>>> >>>>> Nope, it does not tell you it's suboptimal. It tells you that compiler >>>>> will add the additional padding/alignment bytes. >>>>> >>>>> Having structure with say 3 bool's and nothing more is perfectly fine. >>>>> In such a case, to silence the warning, I would have to add some dummy >>>>> fields, which most of the time is not necessary. >>>>> >>>>>> -- >>>>>> Sent from cellphone, sorry for the typos >>>>>> >>>>>> >>>>>> >>>>>> On March 3, 2016 19:30:31 Andrzej Telszewski wrote: >>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> Qt Creator: 3.6.0 >>>>>>> >>>>>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>>>>> I don't believe that warning about padding should be the default behavior. >>>>>>> >>>>>>> -- >>>>>>> Best regards, >>>>>>> Andrzej Telszewski >>>>>>> _______________________________________________ >>>>>>> Qt-creator mailing list >>>>>>> Qt-creator at qt-project.org >>>>>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>>>> >>>>> >>>>> >>>>> -- >>>>> Best regards, >>>>> Andrzej Telszewski >>>> >>>> -- >>>> Sent from cellphone, sorry for the typos >>>> _______________________________________________ >>>> Qt-creator mailing list >>>> Qt-creator at qt-project.org >>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>> >>> >>> >>> -- >>> Pozdrawiam, >>> Best regards, >>> Andrzej Telszewski >> . >> > > > -- > Pozdrawiam, > Best regards, > Andrzej Telszewski > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Alessandro.Portale at theqtcompany.com Mon Mar 7 12:30:28 2016 From: Alessandro.Portale at theqtcompany.com (Portale Alessandro) Date: Mon, 7 Mar 2016 11:30:28 +0000 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: <56D9D733.5080608@gmail.com> References: <56D9D733.5080608@gmail.com> Message-ID: From: Qt-creator [mailto:qt-creator-bounces+alessandro.portale=theqtcompany.com at qt-project.org] On Behalf Of Mike Jackson Sent: Friday, March 4, 2016 7:43 PM To: qt-creator at qt-project.org Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go I just compiled QtCreator 4.0 and what a surprise. It seems as if the QtCreator team seems to have followed everyone else in UI design lately and tossed out 30 years of Human Computer Interaction design. Namely, you tossed colorful icons to go with a flat mono-chrome color palette. I do not understand where this comes from? We programmers have been fighting and waiting for years to get able to have high res icons that can convey information quickly and when we get that capability all the UI designers seem to have just given up on the concept. Color conveys information quickly. It lets the human brain decipher information with less effort than a monochrome palette. Numerous experts in the field have written at length about this issue and where the UI design is going and just how wrong it is. I urge the QtCreator team to at least include a theme that retains the colorful icons from QtCreator 3.x in the 4.0 release. It is not too late. Let me put forth a very direct example. If you drive a white car and you park that car in a parking lot with 1000 other white cars of the same type (say 4 door passenger style cars), when you come back out to get in your car, unless you parked your car in a very specific spot (first or last) your brain needs to work much hard to remember exactly which car is yours. Now imagine your car is red, in that same lot of 1000 white cars. Finding your car is instantaneous. Hi Michael, Thank you for the feedback. Do your comments refer to the state before or after this change: https://codereview.qt-project.org/147723/ ? Best Regards, Alessandro -------------- next part -------------- An HTML attachment was scrubbed... URL: From adrozdoff at gmail.com Mon Mar 7 14:44:26 2016 From: adrozdoff at gmail.com (Alexander Drozdov) Date: Mon, 7 Mar 2016 23:44:26 +1000 Subject: [Qt-creator] CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched] Message-ID: 2016-02-29 19:31 GMT+10:00 Hunger Tobias : > Hi Alexander, > Hi Tobias, sorry for delay > On Mo, 2016-02-29 at 18:44 +1000, Alexander Drozdov wrote: > > Improved CMake support? What about scope of this improvements? > > 4.0 will mostly have workflow improvements for CMake. > > There have been no improvements to the parser, etc. that extracts data from > CMake for our code model. Some bugs got fixed along the way. E.g. syntax > highlighting should work a bit better, but nothing major. > Thanks for info! > > > Currently I > > see, that two of the four features that I support at the > > CMakeProjectManager2 (fork on github) already implemented: > > 1) Handling Kit compiler. But only C++, not C or other. But it simple > > configurable at the Kit configuration page now. > > QtCreator does only know one toolchain for each Kit and assumes that to be > a C++ > one. So there simply is not enough information to set up the C compiler > with the > information available. We will need to manage toolchains and programming > languages they support for that and I currently have no plans to go that > far, > even though that information is also useful for other build systems. > > You can add any cmake settings you need to the Kit's cmake configuration > though > and those settings will then be applied automatically everywhere that kit > is > used. > I know and understand. In my solution I simple predict C compiler based on some info: g++ -> gcc, clang++ -> clang, cl.exe -> cl.exe :-) with same prefixes and suffixes Look (core removed after dialog reworks): https://github.com/h4tr3d/cmakeprojectmanager2/commit/268b1cb93ea3fb40c7bdcaead51452c764dc95bb#diff-8809e7f792412acf4605265b6ad2c20fR790 > > > 2) Pointing Build Type to the CMake > > That was in 3.6 already if I remember correctly:-) > Yes! And it good thing! > > > Next features I need, but it will be nevenr implement: > > 1) Good management for the CMake toolchain files. Upstream changes so > > fundamental, but I already port this changes to the new codebase. But > > pointing compiler via CMAKE_CXX_COMPILER, not via CXX/CC evn variables > > breaks toolchain files. I solve it via forgoting Kit configuration in > case, > > when user select Toolchain file. > > I have never used those myself: What exactly do those do? As I understood > it, > they are just a bunch of settings that are fed to cmake to pre-fill its > cash. Is > that correct? > No. Toolchain file in simple way look like settings. But in some complex cases it can call CMake functions, search complilers via find_program() and so so. To somplify: Toolchain file is a file that preloaded by the cmake before CMakeLists.txt. > > If that is the case: Can the CMake Configuration in the Kits cover that? Not in all cases. Also, in some cases you can want simple builds current project with current Kit but with suggested Toolchain. Toolchain can be specific for single project. Create Kits for every toolchains not a good idea. IMHO. Maybe we could add an "import from toolchain file" to the CMake > Configuration dialog? > Not a good idea. See above. > > > 2) Adding files via context menu and show all files in source Tree: it > > allows to add new file to the target directory, see it and after simple > add > > to the needed target at the CMakeLists.txt > > The only chance I ever see that happening is for cmake-daemon ( > https://steveire > .wordpress.com/2016/01/24/cmake-daemon-for-user-tools/ ) to get accepted > upstream. > I understand. I try to parse cmake (adopt tokenizer to build with plugin) but it to complex to process CMakeLists.txt and in sub-CMakeLists.txt. So currently I use next scenario: 1. For code module I use files provided by the CbpParser 2. For source tree I display all files. Yes, it displayes files like README, LICENSE and some sources that skipped from the compilation (and from the Model). But it simplify navigation and adding/removing/renaming 3. Enable items Add/Rename/Remove file 4. In handlers of the given signals I ask BuildDirManager to do forceReparse() This scenarion works very well for FILE GLOBBING and AUX SOURCE DIR. And yes, this scenarion requires addition CMake execution when files described in CMakeLists.txt without globbing. And yes, this algorithm requires additional work: 1. Scan sources tree 2. Sorts tree and cmake tree 3. Search for added files (ProjectExplorer::compareSortedLists) 4. Substract and clear source dir tree (qDeleteAll + ProjectExplorer::subtractSortedList) 5. Add it to the cmake tree 6. Pass new cmake tree list to the buildTree() Also, this behaviour same to the KDevelop, CLion. But it can be done optional in simple way. One question: what about addition items in the Project tree nodes context menus: is it possible to add sone custom item? Like "Display all files". Look: * https://github.com/h4tr3d/cmakeprojectmanager2/commit/61df428d14c7bb785b50f26a90517aecb701a9eb - reading source tree > > > New changes, that likes me: > > 1) Integrated CMake dialog to the Project page and transparency CMake > > running. > > 2) Table editor for the CMake options. Small qestion according this > > improvement: will be added some set of known options and ability to > select > > it via integrated combo box? will be editor delegates according to the > > option type added, like: if option type is a PATH use File Open dialog? > > The editor already uses the type hints to some extend: The bool ones are > tickboxes that tackles between "ON" and "OFF". > > A combobox to select between a set of different options is not going to > happen, > simply because cmake has no way (that I found) to tell Qt Creator about the > allowed values. Ok. Not a problem. But, I found that it is impossible to add some custom value: there is Edit button but there is no Add button. It works as planed? > Using path-chooser to pick a path/directory is definitely an > option though. > > > So, roadmap for the CMake Project Manager plugin is interesting. > > My roadmap is simple: Make what is there stable for 4.0. > Good :-) > > Next step is to help to polish cmake-daemon up to a state where it can be > pushed > for review to the cmake project and use that when/if it makes it upstream. > That > changeset is a game-changer (not just for Qt Creator!), but unfortunately > my > impression is that the interest from the cmake side is low to non-existent. > Heh. We wants good libcmake for a long time. Seems that cmake daemon possible only as a separate project :-( > > Any help with cmake-daemon would be greatly appreciated by Steven, My work needs a lot of work time. My family requires a lot of personal time. My personal projects and other open source activity requires a lot time too but has so low priority :-) So, I want to do wold better, but I need more than 24 hours per day or too much my clones :-D > any help with the cmake plugin is greatly appreciated by me:-) > What kind of the help you needs? > > Finally, please test the 4.0 branch and report bugs, especially if you use > CMake > a lot. > > Oh yest. A lot. Currently some observation: 1) Kit settings after Apply pressing or after Ok and dialog reopening drops CMake generation selection to the Codeblocks - Ninja and marks Kit as modified. If you select Codeblocks - Makefiles Kit modification markers clears. If there is no reported issue and it is not known issue, I can fill report. 2) With passing CMAKE_CXX_COMPILER via cmake paramters we can loops cmake and QtC project opening in some cases. Simple project that demostrate it is attched. Note, there is 3 (known for me) ways to setup compiler to the QtC: a) via Env variable CXX (CC for C, other variables uses for other langs) b) via Toolchain file c) via -DCMAKE_CXX_COMPILER items declared in the increasing priority order. Current implementation (4.0/master - I use master, it simple for me to maintain Ubuntu PPA with master QtC) uses c) that overrides all other settings. IMO, case a) must be used instead: Kit should simple notify cmake about compiler, not strongly set it. It also can breaks some CMake project-specific magic. But I am not sure: is setting compiler via Env variable is a portable way? But CMake suggests it itself in error outputs: ~~~ Tell CMake where to find the compiler by setting either the environment variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path to the compiler, or to the compiler name if it is in the PATH. ~~~ 3) User needs an ability to force cache clean before next cmake run. It requires in some cases. Look my solution: a) https://github.com/h4tr3d/cmakeprojectmanager2/commit/bb5f74b01c63e092f90a3e4c49c4aaec2f8abea4 b) https://github.com/h4tr3d/cmakeprojectmanager2/commit/a8d6431b2649c36d02c05332c5cafe8faa4539b5 c) https://github.com/h4tr3d/cmakeprojectmanager2/commit/5308efab9d9359c671e35ea65e82369b4040b3f9 d) https://github.com/h4tr3d/cmakeprojectmanager2/commit/38139d6ed97f1fd6adecf92aaa4a55d6c620f513 (adopts changes above to the current master) 4) If you agree with showing all files in project tree and adding/removing/renaming files as I describe above, I can submit merge request tot the gerrit (and checks it work after last accounts merging :-)) Some yesrs ago this changes was discarded. -- WBR, Alexander Drozdov http://htrd.su -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cmake-loop.tar.xz Type: application/x-xz Size: 808 bytes Desc: not available URL: From imikejackson at gmail.com Mon Mar 7 15:15:04 2016 From: imikejackson at gmail.com (Mike Jackson) Date: Mon, 07 Mar 2016 09:15:04 -0500 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: References: <56D9D733.5080608@gmail.com> Message-ID: <56DD8CE8.6000004@gmail.com> Looks like before that change set. I'll give master another go and see what it looks like now. -- Mike Jackson > Portale Alessandro > March 7, 2016 at 6:30 AM > > *From:*Qt-creator > [mailto:qt-creator-bounces+alessandro.portale=theqtcompany.com at qt-project.org] > *On Behalf Of *Mike Jackson > *Sent:* Friday, March 4, 2016 7:43 PM > *To:* qt-creator at qt-project.org > *Subject:* [Qt-creator] QtCreator 4.0: Where did my color icons go > > > I just compiled QtCreator 4.0 and what a surprise. It seems as if the > QtCreator team seems to have followed everyone else in UI design > lately and tossed out 30 years of Human Computer Interaction design. > Namely, you tossed colorful icons to go with a flat mono-chrome color > palette. I do not understand where this comes from? We programmers > have been fighting and waiting for years to get able to have high res > icons that can convey information quickly and when we get that > capability all the UI designers seem to have just given up on the > concept. Color conveys information quickly. It lets the human brain > decipher information with less effort than a monochrome palette. > Numerous experts in the field have written at length about this issue > and where the UI design is going and just how wrong it is. I urge the > QtCreator team to at least include a theme that retains the colorful > icons from QtCreator 3.x in the 4.0 release. It is not too late. > > > Let me put forth a very direct example. If you drive a white car and > you park that car in a parking lot with 1000 other white cars of the > same type (say 4 door passenger style cars), when you come back out to > get in your car, unless you parked your car in a very specific spot > (first or last) your brain needs to work much hard to remember exactly > which car is yours. Now imagine your car is red, in that same lot of > 1000 white cars. Finding your car is instantaneous. > > Hi Michael, > > Thank you for the feedback. Do your comments refer to the state before > or after this change: https://codereview.qt-project.org/147723/? > > Best Regards, > > Alessandro > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lykurg at gmail.com Mon Mar 7 17:42:37 2016 From: lykurg at gmail.com (Lorenz Haas) Date: Mon, 7 Mar 2016 17:42:37 +0100 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: References: <56D9D733.5080608@gmail.com> Message-ID: <56DDAF7D.3020504@gmail.com> Hi Alessandro, are you going to implement the entire flat style UI Diana proposed here http://blog.qt.io/blog/2015/02/09/qt-creator-flat-style-ui-design-new-light-theme for 4.0? Cheers, Lorenz From ariel at edis.mx Mon Mar 7 18:02:36 2016 From: ariel at edis.mx (Ariel Molina) Date: Mon, 7 Mar 2016 11:02:36 -0600 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: <56DDAF7D.3020504@gmail.com> References: <56D9D733.5080608@gmail.com> <56DDAF7D.3020504@gmail.com> Message-ID: Please do it, that's a nice theme! On Mon, Mar 7, 2016 at 10:42 AM, Lorenz Haas wrote: > Hi Alessandro, > > are you going to implement the entire flat style UI Diana proposed here > > http://blog.qt.io/blog/2015/02/09/qt-creator-flat-style-ui-design-new-light-theme > for 4.0? > > Cheers, > Lorenz > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Ariel Molina R. Oficina: +52 (222) 3723196 Movil: +521 2226 758874 http://edis.mx -------------- next part -------------- An HTML attachment was scrubbed... URL: From jhihn at gmx.com Mon Mar 7 20:44:08 2016 From: jhihn at gmx.com (Jason H) Date: Mon, 7 Mar 2016 20:44:08 +0100 Subject: [Qt-creator] Sug: Refactor text Message-ID: It would be cool if we could right click on a string "this" and refactor: translate or refactor:translate plural, where in it gets changed from "this" to: qsTr("this") or qsTr("this", "", n) Respectively. Alternative "refactorings" could be: All uppercase, Title Case, or all lower case. From Tobias.Hunger at theqtcompany.com Tue Mar 8 10:39:31 2016 From: Tobias.Hunger at theqtcompany.com (Hunger Tobias) Date: Tue, 8 Mar 2016 09:39:31 +0000 Subject: [Qt-creator] Sug: Refactor text In-Reply-To: References: Message-ID: <1457429970.1068.4.camel@theqtcompany.com> On Mo, 2016-03-07 at 20:44 +0100, Jason H wrote: > It would be cool if we could right click on a string "this" and refactor: > translate or refactor:translate plural, where in it gets changed from "this" > to: > qsTr("this")  or > qsTr("this", "", n) > Respectively. > > Alternative "refactorings" could be: All uppercase, Title Case, or all lower > case.  Please report suggestions, feature requests and bug reports at https://bugreport s.qt.io/ Reports stay around in the bug tracker forever... here they get lost as soon as enough mails have arrived to push your suggestions down far enough to require scrolling;-) Best Regards, Tobias -- Tobias Hunger, Senior Software Engineer | The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Tobias.Hunger at theqtcompany.com Tue Mar 8 11:01:13 2016 From: Tobias.Hunger at theqtcompany.com (Hunger Tobias) Date: Tue, 8 Mar 2016 10:01:13 +0000 Subject: [Qt-creator] CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched] In-Reply-To: References: Message-ID: <1457431272.1068.23.camel@theqtcompany.com> Hi Alexander, On Mo, 2016-03-07 at 23:44 +1000, Alexander Drozdov wrote: > 2016-02-29 19:31 GMT+10:00 Hunger Tobias : > Hi Tobias, sorry for delay No worry:-) > >  Currently I > > > see, that two of the four features that I support at the > > > CMakeProjectManager2 (fork on github) already implemented: > > > 1) Handling Kit compiler. But only C++, not C or other. But it simple > > > configurable at the Kit configuration page now. > > QtCreator does only know one toolchain for each Kit and assumes that to be > > a C++ > > one. So there simply is not enough information to set up the C compiler > > with the > > information available. We will need to manage toolchains and programming > > languages they support for that and I currently have no plans to go that > > far, > > even though that information is also useful for other build systems. > > > > You can add any cmake settings you need to the Kit's cmake configuration > > though > > and those settings will then be applied automatically everywhere that kit > > is > > used. > > > I know and understand. In my solution I simple predict C compiler based on > some info: g++ -> gcc, clang++ -> clang, cl.exe -> cl.exe :-) with same > prefixes and suffixes > Look (core removed after dialog reworks): > https://github.com/h4tr3d/cmakeprojectmanager2/commit/268b1cb93ea3fb40c7bdcaea > d51452c764dc95bb#diff-8809e7f792412acf4605265b6ad2c20fR790 I would prefer to keep this as is: It is pretty straight forward for the users (or an admin/packager) to provide custom cmake config flags to be set by a kit. So this problem can be addressed at that level (till Qt Creator starts to manage several toolchains per kit at some point -- if we ever go there). > No. Toolchain file in simple way look like settings. But in some complex > cases it can call CMake functions, search complilers via find_program() and > so so. > > To somplify: Toolchain file is a file that preloaded by the cmake before > CMakeLists.txt. I would argue that they still are just a bunch of settings... at least after they are executed. Since kits are host specific, I still think we can get away with the current approach: Just put whatever concrete information you need into the Kit's settings and you are done. Yes, that fails to detect compilers and stuff for you, but you know where those are anyway. > If that is the case: Can the CMake Configuration in the Kits cover that? > > Not in all cases. Also, in some cases you can want simple builds current > project with current Kit but with suggested Toolchain. Toolchain can be > specific for single project. Create Kits for every toolchains not a good > idea. IMHO. IMHO it is:-) How else would you switch between different configurations easily? > Maybe we could add an "import from toolchain file" to the CMake > > > > Configuration dialog? > > > Not a good idea. See above. OK. > 2) Adding files via context menu and show all files in source Tree: it > > > allows to add new file to the target directory, see it and after simple > > add > > > > > > to the needed target at the CMakeLists.txt > > The only chance I ever see that happening is for cmake-daemon ( > > https://steveire > > .wordpress.com/2016/01/24/cmake-daemon-for-user-tools/ ) to get accepted > > upstream. > > > I understand. I try to parse cmake (adopt tokenizer to build with plugin) > but it to complex to process CMakeLists.txt and in sub-CMakeLists.txt. I am not brave enough to go there:-) > So currently I use next scenario: > 1. For code module I use files provided by the CbpParser > 2. For source tree I display all files. Yes, it displayes files like > README, LICENSE and some sources that skipped from the compilation (and > from the Model). But it simplify navigation and adding/removing/renaming I have been thinking on doing that, too. I hate the idea of just scanning the file system, but I think it is the best that cmake will let us do. > 3. Enable items Add/Rename/Remove file I don't plan on that yet. > 4. In handlers of the given signals I ask BuildDirManager to do > forceReparse() > > This scenarion works very well for FILE GLOBBING and AUX SOURCE DIR. And > yes, this scenarion requires addition CMake execution when files described > in CMakeLists.txt without globbing. Yes, I see that it will work there. But that is not good enough for me, simply because I will get tons of bug reports about "Add file" not working from lots of people that do not use such a setup. There are just too many cmake users out there for me to risk that:-) > And yes, this algorithm requires additional work: > 1. Scan sources tree > 2. Sorts tree and cmake tree > 3. Search for added files (ProjectExplorer::compareSortedLists) > 4. Substract and clear source dir tree (qDeleteAll + > ProjectExplorer::subtractSortedList) > 5. Add it to the cmake tree > 6. Pass new cmake tree list to the buildTree() > > Also, this behaviour same to the KDevelop, CLion. But it can be done > optional in simple way. > > One question: what about addition items in the Project tree nodes context > menus: is it possible to add sone custom item? Like "Display all files". Not in a straight forward way... The tree is pretty static as is the transformation from the full tree to the simplified tree. > Look: > * > https://github.com/h4tr3d/cmakeprojectmanager2/commit/61df428d14c7bb785b50f26a > 90517aecb701a9eb > - reading source tree Could you consider to push the scanning logic upstream? > > New changes, that likes me: > > > 1) Integrated CMake dialog to the Project page and transparency CMake > > > running. > > > 2) Table editor for the CMake options. Small qestion according this > > > improvement: will be added some set of known options and ability to > > select > > > > > > it via integrated combo box? will be editor delegates according to the > > > option type added, like: if option type is a PATH use File Open dialog? > > The editor already uses the type hints to some extend: The bool ones are > > tickboxes that tackles between "ON" and "OFF". > > > > A combobox to select between a set of different options is not going to > > happen, > > simply because cmake has no way (that I found) to tell Qt Creator about the > > allowed values. > > Ok. Not a problem. But, I found that it is impossible to add some custom > value: there is Edit button but there is no Add button. It works as planed? I wanted to implement adding and unsetting values, but then I thought that this will probably not be necessary: CMake puts all values it cares for into the cache, doesn't it? If that is true, then anything you add will be ignored anyway -- so why offer that in the first place? I am still unsure about the unsetting of values though. > > Next step is to help to polish cmake-daemon up to a state where it can be > > pushed > > for review to the cmake project and use that when/if it makes it upstream. > > That > > changeset is a game-changer (not just for Qt Creator!), but unfortunately > > my > > impression is that the interest from the cmake side is low to non-existent. > > > Heh. We wants good libcmake for a long time. Seems that cmake daemon > possible only as a separate project :-( If it stays separate, then I won't use it. I do not have the time to maintain a cmake branch! I am very willing to help getting the code upstream and to get it into a state where it is useful to a real-live project (currently it is suitable for some demos, but not production ready), but I do not want to commit to more than that. > > Any help with cmake-daemon would be greatly appreciated by Steven, > > My work needs a lot of work time. My family requires a lot of personal > time. My personal projects and other open source activity requires a lot > time too but has so low priority :-) > So, I want to do wold better, but I need more than 24 hours per day or too > much my clones :-D Yes, same here:-) I want to hack on an IDE, not some random build system! > > any help with the cmake plugin is greatly appreciated by me:-) > > > What kind of the help you needs? Bugreports, feature requests, code, documentation updates, everything:-) > > Finally, please test the 4.0 branch and report bugs, especially if you use > > CMake > > a lot. > > > > Oh yest. A lot. Currently some observation: Could you please file such issues in the bugtracker at https://bugreports.qt.io/  ? They are much simpler to track there! I will try to fix the stuff you reported by mail asap. No need to put those into the tracker:-) > 2) With passing CMAKE_CXX_COMPILER via cmake paramters we can loops cmake > and QtC project opening in some cases. Simple project that demostrate it is > attched. Note, there is 3 (known for me) ways to setup compiler to the QtC: > a) via Env variable CXX (CC for C, other variables uses for other langs) > b) via Toolchain file > c) via -DCMAKE_CXX_COMPILER > items declared in the increasing priority order. Current implementation > (4.0/master - I use master, it simple for me to maintain Ubuntu PPA with > master QtC) uses c) that overrides all other settings. IMO, case a) must be > used instead: Kit should simple notify cmake about compiler, not strongly > set it. It also can breaks some CMake project-specific magic. > > But I am not sure: is setting compiler via Env variable is a portable way? > But CMake suggests it itself  in error outputs: > ~~~ >   Tell CMake where to find the compiler by setting either the environment >   variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full > path >   to the compiler, or to the compiler name if it is in the PATH. > ~~~ I still prefer to just use -DCMAKE_CXX_COMPILER: We want to force it, as the code model needs to have the correct compiler to be fully useful. > 3) User needs an ability to force cache clean before next cmake run. It > requires in some cases. Look my solution: > a) > https://github.com/h4tr3d/cmakeprojectmanager2/commit/bb5f74b01c63e092f90a3e4c > 49c4aaec2f8abea4 > b) > https://github.com/h4tr3d/cmakeprojectmanager2/commit/a8d6431b2649c36d02c05332 > c5cafe8faa4539b5 > c) > https://github.com/h4tr3d/cmakeprojectmanager2/commit/5308efab9d9359c671e35ea6 > 5e82369b4040b3f9 > d) > https://github.com/h4tr3d/cmakeprojectmanager2/commit/38139d6ed97f1fd6adecf92a > aa4a55d6c620f513 > (adopts changes above to the current master) Could you push that into codereview.qt-project.org? That makes it way safer for me to reuse some/all the code. > 4) If you agree with showing all files in project tree and > adding/removing/renaming files as I describe above, I can submit merge > request tot the gerrit (and checks it work after last accounts merging :-)) > Some yesrs ago this changes was discarded. I would very much like the file scanning stuff, but the adding/removing itself will probably cause way to many bug reports for me to feel comfortable to implement that at this time. Best Regards and thanks for taking the time to report your feedback! Tobias -- Tobias Hunger, Senior Software Engineer | The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From adrozdoff at gmail.com Tue Mar 8 15:48:29 2016 From: adrozdoff at gmail.com (Alexander Drozdov) Date: Wed, 9 Mar 2016 00:48:29 +1000 Subject: [Qt-creator] CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched] In-Reply-To: <1457431272.1068.23.camel@theqtcompany.com> References: <1457431272.1068.23.camel@theqtcompany.com> Message-ID: 2016-03-08 20:01 GMT+10:00 Hunger Tobias : > Hi Alexander, > Hi again! [skip] > > > > > I know and understand. In my solution I simple predict C compiler based > on > > some info: g++ -> gcc, clang++ -> clang, cl.exe -> cl.exe :-) with same > > prefixes and suffixes > > Look (core removed after dialog reworks): > > > https://github.com/h4tr3d/cmakeprojectmanager2/commit/268b1cb93ea3fb40c7bdcaea > > d51452c764dc95bb#diff-8809e7f792412acf4605265b6ad2c20fR790 > > I would prefer to keep this as is: It is pretty straight forward for the > users > (or an admin/packager) to provide custom cmake config flags to be set by a > kit. > So this problem can be addressed at that level (till Qt Creator starts to > manage > several toolchains per kit at some point -- if we ever go there). > I think, that compilers for the langs, supported by the QtC should be detected too. Otherwise: we can create C project via wizard but can't point C compiler. Not a good (is it problem solved for Qbs, qmake and how?). But this question must be discussed ouside of the CMake-only context. It more deeply. > > > > > No. Toolchain file in simple way look like settings. But in some complex > > cases it can call CMake functions, search complilers via find_program() > and > > so so. > > > > To somplify: Toolchain file is a file that preloaded by the cmake before > > CMakeLists.txt. > > I would argue that they still are just a bunch of settings... at least > after > they are executed. Since kits are host specific, I still think we can get > away > with the current approach: Just put whatever concrete information you need > into > the Kit's settings and you are done. Yes, that fails to detect compilers > and > stuff for you, but you know where those are anyway. > > > If that is the case: Can the CMake Configuration in the Kits cover that? > > > > Not in all cases. Also, in some cases you can want simple builds current > > project with current Kit but with suggested Toolchain. Toolchain can be > > specific for single project. Create Kits for every toolchains not a good > > idea. IMHO. > > IMHO it is:-) How else would you switch between different configurations > easily? > Heh, two compilers (gcc 4.8, gcc 5.3) and two Qt versions (4.8, 5.5) requires to create four Kit configurations. Adding clang increases possible Kit configurations up to six. Increasing Kit settings increases possible Kit configurations dramatically. CMake toolchain can use same compiler but provide some other presets. In general Toolchain file can be used not only for cross-building. In common: it can be interpreted as a per-build project configuration. Complex and power configuration, not only VARIABLE=value pairs. [skip] > > > I understand. I try to parse cmake (adopt tokenizer to build with plugin) > > but it to complex to process CMakeLists.txt and in sub-CMakeLists.txt. > > I am not brave enough to go there:-) > I too :-) Simply tried. But KDevelop and CLion guys more-less solve this problem. > > > 4. In handlers of the given signals I ask BuildDirManager to do > > forceReparse() > > > > This scenarion works very well for FILE GLOBBING and AUX SOURCE DIR. And > > yes, this scenarion requires addition CMake execution when files > described > > in CMakeLists.txt without globbing. > > Yes, I see that it will work there. But that is not good enough for me, > simply > because I will get tons of bug reports about "Add file" not working from > lots of > people that do not use such a setup. There are just too many cmake users > out > there for me to risk that:-) > Hmm. On file adding cmake reparsing forces. We can checks that added files not goes to the any target and output notice to the top of the editor (like in past notice to re-run cmake outputs). Like: "File added but not assigned to any target, please edit CMakeLists.txt manually". Yes only source-code files should be filtered :-) [skip] > > Not in a straight forward way... The tree is pretty static as is the > transformation from the full tree to the simplified tree. > Ok, thanks. > > > Look: > > * > > > https://github.com/h4tr3d/cmakeprojectmanager2/commit/61df428d14c7bb785b50f26a > > 90517aecb701a9eb > > - reading source tree > > Could you consider to push the scanning logic upstream? > No problem. Simple give me some amount of time :-) Also, could you please provide Best Qt-friedly way to scan directory tree? If recursion can be omited - it will be good. My current solution uses QDir::entryInfoList() + recursion. [skip] > > > A combobox to select between a set of different options is not going to > > > happen, > > > simply because cmake has no way (that I found) to tell Qt Creator > about the > > > allowed values. > > > > Ok. Not a problem. But, I found that it is impossible to add some custom > > value: there is Edit button but there is no Add button. It works as > planed? > > I wanted to implement adding and unsetting values, but then I thought that > this > will probably not be necessary: CMake puts all values it cares for into the > cache, doesn't it? If that is true, then anything you add will be ignored > anyway > -- so why offer that in the first place? > I am not ready to answer. > > I am still unsure about the unsetting of values though. > I will fill feature request for next thing: outputs description for option if present (new table colum is needed). What do you think about it? > > What kind of the help you needs? > Bugreports, feature requests, code, documentation updates, everything:-) > Ooookay! > > > > Finally, please test the 4.0 branch and report bugs, especially if you > use > > > CMake > > > a lot. > > > > > > Oh yest. A lot. Currently some observation: > > Could you please file such issues in the bugtracker at > https://bugreports.qt.io/ > ? They are much simpler to track there! > I work with master version of the QtC (yes I am crazy. a little), some features can be incomplete yet or contains known bugs that will be fixed by the next commit. What do you think about next steps: 1. found issue 2. search on bug tracker 3. wait for a week 4. try agait 5. still present? search on bug tracker 6. not found? report it. Or I can mail it and fill report after feedback? (currently only two most interesting places of the QtC for me: Baremetal and CMake) > I still prefer to just use -DCMAKE_CXX_COMPILER: We want to force it, as > the > code model needs to have the correct compiler to be fully useful. > As I understand Code model simple uses Kit compiler, am I right? So, if user select Kit but overrides Toolchain and (possible) Compiler it is expected behaviour that QtC Model continue to work with current Kit-Specific compiler, now with Toolchain-specific. It also helps to avoid cmake loops. > > > 3) User needs an ability to force cache clean before next cmake run. It > > requires in some cases. Look my solution: > [skip] > Could you push that into codereview.qt-project.org? That makes it way > safer for > me to reuse some/all the code. > As for scanning: give me time :-) > > > 4) If you agree with showing all files in project tree and > > adding/removing/renaming files as I describe above, I can submit merge > > request tot the gerrit (and checks it work after last accounts merging > :-)) > > Some yesrs ago this changes was discarded. > > I would very much like the file scanning stuff, but the adding/removing > itself > will probably cause way to many bug reports for me to feel comfortable to > implement that at this time. > Ok. I will submit "dir scanning" as a complete request and "adding/renaming/removing" after some time as a separate patch: to discuss via review board. I think that adding files via context menu very comfortable and bug reports can be prevented by the good user iteraction. > Best Regards and thanks for taking the time to report your feedback! > You are welcome! -- WBR, Alexander Drozdov http://htrd.su -------------- next part -------------- An HTML attachment was scrubbed... URL: From christian.kandeler at theqtcompany.com Tue Mar 8 15:59:19 2016 From: christian.kandeler at theqtcompany.com (Christian Kandeler) Date: Tue, 8 Mar 2016 15:59:19 +0100 Subject: [Qt-creator] CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched] In-Reply-To: References: <1457431272.1068.23.camel@theqtcompany.com> Message-ID: <56DEE8C7.8090800@theqtcompany.com> On 03/08/2016 03:48 PM, Alexander Drozdov wrote: > I think, that compilers for the langs, supported by the QtC should be > detected too. Otherwise: we can create C project via wizard but can't > point C compiler. Not a good (is it problem solved for Qbs, qmake and > how?). It's an unsolved problem. Basically Qt Creator assumes single-language projects. Christian From Alessandro.Portale at theqtcompany.com Tue Mar 8 18:22:43 2016 From: Alessandro.Portale at theqtcompany.com (Portale Alessandro) Date: Tue, 8 Mar 2016 17:22:43 +0000 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: References: <56D9D733.5080608@gmail.com> <56DDAF7D.3020504@gmail.com>, Message-ID: Hi, Yes, implementing Diana's theme for 4.0 is the goal. At least the darker variant, and the light one afterwards. The original theme with gradients and colored mode icons as-well as the "dark" theme are supposed to stay. Best Regards, Alessandro ________________________________ From: Qt-creator on behalf of Ariel Molina Please do it, that's a nice theme! On Mon, Mar 7, 2016 at 10:42 AM, Lorenz Haas > wrote: Hi Alessandro, are you going to implement the entire flat style UI Diana proposed here http://blog.qt.io/blog/2015/02/09/qt-creator-flat-style-ui-design-new-light-theme for 4.0? Cheers, Lorenz -------------- next part -------------- An HTML attachment was scrubbed... URL: From jhihn at gmx.com Tue Mar 8 18:28:34 2016 From: jhihn at gmx.com (Jason H) Date: Tue, 8 Mar 2016 18:28:34 +0100 Subject: [Qt-creator] Sug: Refactor text In-Reply-To: <1457429970.1068.4.camel@theqtcompany.com> References: , <1457429970.1068.4.camel@theqtcompany.com> Message-ID: > On Mo, 2016-03-07 at 20:44 +0100, Jason H wrote: > > It would be cool if we could right click on a string "this" and refactor: > > translate or refactor:translate plural, where in it gets changed from "this" > > to: > > qsTr("this")  or > > qsTr("this", "", n) > > Respectively. > > > > Alternative "refactorings" could be: All uppercase, Title Case, or all lower > > case.  > > Please report suggestions, feature requests and bug reports at https://bugreport > s.qt.io/ > > Reports stay around in the bug tracker forever... here they get lost as soon as > enough mails have arrived to push your suggestions down far enough to require > scrolling;-) That's fine. Sometimes I like to post here first to get comments, ranging from: 1. It's there, you just aren't using it correctly. 2. That's brilliant! :-) 3. That's stupid. :-( 4. Me too. (In no particular order) From lykurg at gmail.com Tue Mar 8 19:15:17 2016 From: lykurg at gmail.com (Lorenz Haas) Date: Tue, 8 Mar 2016 19:15:17 +0100 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: References: <56D9D733.5080608@gmail.com> <56DDAF7D.3020504@gmail.com> Message-ID: <56DF16B5.8030705@gmail.com> > Yes, implementing Diana's theme for 4.0 is the goal. At least the darker > variant, and the light one afterwards. Nice! I've already started to play with the variable in the theme since I like the flat buttons on the right :) Thanks for implementing it. Lorenz From jhihn at gmx.com Tue Mar 8 23:10:32 2016 From: jhihn at gmx.com (Jason H) Date: Tue, 8 Mar 2016 23:10:32 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes Message-ID: A while back, I started a discussion about Creator's quoting policy. There was a lot of support, and a few people were concerned about changing the behavior that has been there since the beginning. However, I'm again tired of fighting with creator inserting "" when I type " and deleting both when I backspace over one. This behavior only occurs in QtCreator, and no other editor _ever_, and it needs to go. It is a "clever trick" that costs more than it is clever. At least for me, YMMV. Yes, there is a checkbox, but that applies to brackets and braces as well. Fundamentally, I want to changes to the behavior: 1. If the character after the cursor is not whitespace, do NOT double quote ("") 2. If backspacing over a double quote, do not remove both double quotes. For the love of all that is binary!!! From ariel at edis.mx Tue Mar 8 23:15:30 2016 From: ariel at edis.mx (Ariel Molina) Date: Tue, 8 Mar 2016 16:15:30 -0600 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: I support Jason, it's annoying, but somehow i got used to the pattern of post-quote elimination which is suboptimal to say the least. Ariel On Tue, Mar 8, 2016 at 4:10 PM, Jason H wrote: > A while back, I started a discussion about Creator's quoting policy. > There was a lot of support, and a few people were concerned about changing > the behavior that has been there since the beginning. > > However, I'm again tired of fighting with creator inserting "" when I type > " and deleting both when I backspace over one. This behavior only occurs in > QtCreator, and no other editor _ever_, and it needs to go. It is a "clever > trick" that costs more than it is clever. At least for me, YMMV. > > Yes, there is a checkbox, but that applies to brackets and braces as well. > > Fundamentally, I want to changes to the behavior: > 1. If the character after the cursor is not whitespace, do NOT double > quote ("") > 2. If backspacing over a double quote, do not remove both double quotes. > > For the love of all that is binary!!! > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Ariel Molina R. Oficina: +52 (222) 3723196 Movil: +521 2226 758874 http://edis.mx -------------- next part -------------- An HTML attachment was scrubbed... URL: From jochen_becher at gmx.de Tue Mar 8 23:27:38 2016 From: jochen_becher at gmx.de (Jochen Becher) Date: Tue, 08 Mar 2016 23:27:38 +0100 Subject: [Qt-creator] Linguist Message-ID: <1457476058.19124.1.camel@gmx.de> Hi, is there any good reason why linguist is not configured as a standard external tool? I'd like to add it. Regards, Jochen From realnc at gmail.com Wed Mar 9 03:49:53 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Wed, 9 Mar 2016 04:49:53 +0200 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: On 09/03/16 00:10, Jason H wrote: > A while back, I started a discussion about Creator's quoting policy. There was a lot of support, and a few people were concerned about changing the behavior that has been there since the beginning. > > However, I'm again tired of fighting with creator inserting "" when I type " and deleting both when I backspace over one. This behavior only occurs in QtCreator, and no other editor _ever_, and it needs to go. It is a "clever trick" that costs more than it is clever. At least for me, YMMV. > > Yes, there is a checkbox, but that applies to brackets and braces as well. > > Fundamentally, I want to changes to the behavior: > 1. If the character after the cursor is not whitespace, do NOT double quote ("") > 2. If backspacing over a double quote, do not remove both double quotes. > > For the love of all that is binary!!! Wait, it's not a bug? They actually want that behavior? lol :-P From apaku at gmx.de Wed Mar 9 09:14:13 2016 From: apaku at gmx.de (Andreas Pakulat) Date: Wed, 9 Mar 2016 09:14:13 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: Hi, On Tue, Mar 8, 2016 at 11:10 PM, Jason H wrote: > A while back, I started a discussion about Creator's quoting policy. > There was a lot of support, and a few people were concerned about changing > the behavior that has been there since the beginning. > > However, I'm again tired of fighting with creator inserting "" when I type > " and deleting both when I backspace over one. This behavior only occurs in > QtCreator, and no other editor _ever_, and it needs to go. It is a "clever > trick" that costs more than it is clever. At least for me, YMMV. > Thats actually not true, all kate-based editors/IDE's used to have the same behavior. I haven't used those for a few years now so it may have been changed. In addition Eclipse' Java editor also has almost the same behavior. That being said, QtC does have a few bugs that I believe Eclipse may not have. For example, hitting " twice in a row does not insert 3 quotes, but rather inserts a double-quote and then moves my cursor after the second quote. And deleting the second quote will indeed not delete both, thats where Eclipse deviates from QtC's behavior. Personally I like the automatic closing-quote injection for the same reason that I like the closing-brace injection: I can just concern myself with the actual content. And while I recall this behavior getting in my way in QtC, I can't recall having that problem in Eclipse (or maybe my brain is just more selective than I'd like it to be). > Fundamentally, I want to changes to the behavior: > 1. If the character after the cursor is not whitespace, do NOT double > quote ("") > Whats the reasoning behind this, i.e. why is a whitespace a good indicator that you want a closing quote added? That would work quite badly if completion added a function call with spaces inside the parenthesis or when the user added those already. > 2. If backspacing over a double quote, do not remove both double quotes. > I can agree to that, it saves you at most 1 keystroke but often times causes more pain, for example if there happens to be 3 quotes in a row (for whatever reason). Andreas -------------- next part -------------- An HTML attachment was scrubbed... URL: From Eike.Ziller at theqtcompany.com Wed Mar 9 09:21:28 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Wed, 9 Mar 2016 08:21:28 +0000 Subject: [Qt-creator] Linguist In-Reply-To: <1457476058.19124.1.camel@gmx.de> References: <1457476058.19124.1.camel@gmx.de> Message-ID: <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> > On Mar 8, 2016, at 11:27 PM, Jochen Becher wrote: > > Hi, > > is there any good reason why linguist is not configured as a standard > external tool? I'd like to add it. You can open a .ts file in Qt Linguist with “Open with... > Qt Linguist” (e.g. from the project tree, file system view, or from the document menu next to the file name above the editor (when it is already opened in an editor). Br, Eike > > Regards, Jochen > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From adrozdoff at gmail.com Wed Mar 9 12:24:17 2016 From: adrozdoff at gmail.com (Alexander Drozdov) Date: Wed, 9 Mar 2016 21:24:17 +1000 Subject: [Qt-creator] CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched] Message-ID: 2016-03-08 20:01 GMT+10:00 Hunger Tobias : > > > 4) If you agree with showing all files in project tree and > > adding/removing/renaming files as I describe above, I can submit merge > > request tot the gerrit (and checks it work after last accounts merging > :-)) > > Some yesrs ago this changes was discarded. > > I would very much like the file scanning stuff, but the adding/removing > itself > will probably cause way to many bug reports for me to feel comfortable to > implement that at this time. > > I tried to login to the CodeReview via new signle token... But login process fails with error and creates new account that can't be assigned with all my mails :( My old user name on CodeReview is a monah_tuk, mails: adrozdoff at gmail.com and hatred at inbox.ru. New accout ID is: 1003847. Is it possible to assign old account (https://codereview.qt-project.org/#/q/owner:%22monah_tuk%22,n,z) with new single login system? PS added list: by my mistake was sended only to Tobias -- WBR, Alexander Drozdov http://htrd.su -------------- next part -------------- An HTML attachment was scrubbed... URL: From oswald.buddenhagen at theqtcompany.com Wed Mar 9 12:58:09 2016 From: oswald.buddenhagen at theqtcompany.com (Oswald Buddenhagen) Date: Wed, 9 Mar 2016 12:58:09 +0100 Subject: [Qt-creator] fixing qt login (Re: CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched]) In-Reply-To: References: Message-ID: <20160309115809.GA22632@troll08.it.local> On Wed, Mar 09, 2016 at 09:24:17PM +1000, Alexander Drozdov wrote: > I tried to login to the CodeReview via new signle token... But login > process fails with error and creates new account that can't be assigned > with all my mails :( > > My old user name on CodeReview is a monah_tuk, mails: adrozdoff at gmail.com > and hatred at inbox.ru. New accout ID is: 1003847. Is it possible to assign > old account (https://codereview.qt-project.org/#/q/owner:%22monah_tuk%22,n,z) > with new single login system? > your old account should now work with the qt login, and the new account ceased to exist. your secondary jira account (which was named like your second email address) was inactivated, and the issues reported from it were moved to the other account. comments made from it cannot be re-assigned, though. that sort-of implies that some of your watches were de-facto lost. From adrozdoff at gmail.com Wed Mar 9 13:44:38 2016 From: adrozdoff at gmail.com (Alexander Drozdov) Date: Wed, 9 Mar 2016 22:44:38 +1000 Subject: [Qt-creator] fixing qt login (Re: CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched]) In-Reply-To: <20160309115809.GA22632@troll08.it.local> References: <20160309115809.GA22632@troll08.it.local> Message-ID: 2016-03-09 21:58 GMT+10:00 Oswald Buddenhagen < oswald.buddenhagen at theqtcompany.com>: > your old account should now work with the qt login, and the new account > ceased to exist. > > your secondary jira account (which was named like your second email > address) was inactivated, and the issues reported from it were moved to > the other account. comments made from it cannot be re-assigned, though. > that sort-of implies that some of your watches were de-facto lost. > Oswald, thanks a lot! -- WBR, Alexander Drozdov http://htrd.su -------------- next part -------------- An HTML attachment was scrubbed... URL: From nikolai.kosjar at theqtcompany.com Wed Mar 9 14:09:41 2016 From: nikolai.kosjar at theqtcompany.com (Nikolai Kosjar) Date: Wed, 9 Mar 2016 14:09:41 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: <56E02095.2040506@theqtcompany.com> Hi! Let me summarize for myself and others the ways we can add quotes to some text. Default settings are assumed, the relevant one is: [x] Automatically insert matching characters. a) Insert quote before text: Type: "foo This results in: "foo" The closing quote was added after typing the initial ". Looks fine. b) Insert quotes afterwards by using selection. 1. Type: foo bar 2. Select the typed text 3. Hit " This results in: "foo bar" Looks fine. c) Insert quotes afterwards by starting with the closing quote You have: foo bar Type: " This results in: foo bar"" Ops 1: The second " was not intended. Hit backspace. Ops 2: Both quotes are removed and you are the beginning again. Note that if you hit " and another one is added, your cursor is then between the quotes: "|". Hitting backspace here removes then both quotes. If the cursor is behind both quotes (e.g. you've moved it there manually), only one quote is removed. This also applies to () and [], but not to {} (thinking of lambdas we probably should include that one, too). Although the option in the settings is named "Automatically insert matching characters" it seems to also handle the removal of the pairs, at least when the cursor is in-between. We probably should change this if we do not introduce further options. I agree that we should address the "Ops 2" from above, that is not removing both quotes if the cursor is in-between them. This will make it inconsistent when looking at () and [], but I think that's justified in this case. The problem here is the single quote by itself - you can't tell whether it's an opening or closing one without further information (typing ')' does not add you the opening one automatically). Regarding "Ops 1", I don't think we can reliably detect whether the users inserts an opening or closing quote. Your heuristic is not enough: On 03/08/2016 11:10 PM, Jason H wrote: > Fundamentally, I want to changes to the behavior: > 1. If the character after the cursor is not whitespace, do NOT double quote ("") Would break case a) from above for e.g. function arguments: foo(|) foo(|,1) > 2. If backspacing over a double quote, do not remove both double quotes. Agree, see above. Nikolai From Eike.Ziller at theqtcompany.com Wed Mar 9 15:03:23 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Wed, 9 Mar 2016 14:03:23 +0000 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: <56E02095.2040506@theqtcompany.com> References: <56E02095.2040506@theqtcompany.com> Message-ID: <78C89B76-0FD1-4B06-B928-C22B8398B0F7@theqtcompany.com> > On Mar 9, 2016, at 2:09 PM, Nikolai Kosjar wrote: > > Hi! > > Let me summarize for myself and others the ways we can add quotes to some text. Default settings are assumed, the relevant one is: > > [x] Automatically insert matching characters. > > > a) Insert quote before text: > Type: "foo > This results in: "foo" > The closing quote was added after typing the initial ". > Looks fine. > > b) Insert quotes afterwards by using selection. > 1. Type: foo bar > 2. Select the typed text > 3. Hit " > This results in: "foo bar" > Looks fine. Actually this is often wrong. E.g. select “foo” and pressing “ will add quotes around the already quoted text. Also in many other cases I just want to replace the selected text when pressing a key, even if it is a " > c) Insert quotes afterwards by starting with the closing quote > You have: foo bar > Type: " > This results in: foo bar"" > Ops 1: The second " was not intended. > Hit backspace. > Ops 2: Both quotes are removed and you are the beginning again. > > > Note that if you hit " and another one is added, your cursor is then between the quotes: "|". Hitting backspace here removes then both quotes. If the cursor is behind both quotes (e.g. you've moved it there manually), only one quote is removed. This also applies to () and [], but not to {} (thinking of lambdas we probably should include that one, too). > > Although the option in the settings is named "Automatically insert matching characters" it seems to also handle the removal of the pairs, at least when the cursor is in-between. We probably should change this if we do not introduce further options. > > I agree that we should address the "Ops 2" from above, that is not removing both quotes if the cursor is in-between them. This will make it inconsistent when looking at () and [], but I think that's justified in this case. The problem here is the single quote by itself - you can't tell whether it's an opening or closing one without further information (typing ')' does not add you the opening one automatically). > > Regarding "Ops 1", I don't think we can reliably detect whether the users inserts an opening or closing quote. Your heuristic is not enough: > > On 03/08/2016 11:10 PM, Jason H wrote: >> Fundamentally, I want to changes to the behavior: >> 1. If the character after the cursor is not whitespace, do NOT double quote ("") > > Would break case a) from above for e.g. function arguments: > > foo(|) > foo(|,1) > >> 2. If backspacing over a double quote, do not remove both double quotes. > > Agree, see above. > > > Nikolai > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From nikolai.kosjar at theqtcompany.com Wed Mar 9 16:10:47 2016 From: nikolai.kosjar at theqtcompany.com (Nikolai Kosjar) Date: Wed, 9 Mar 2016 16:10:47 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: <56E02095.2040506@theqtcompany.com> References: <56E02095.2040506@theqtcompany.com> Message-ID: <56E03CF7.1010302@theqtcompany.com> >> 2. If backspacing over a double quote, do not remove both double quotes. > > Agree, see above. https://codereview.qt-project.org/#/c/151822/ From thetypz at gmail.com Wed Mar 9 16:35:37 2016 From: thetypz at gmail.com (Typz) Date: Wed, 9 Mar 2016 16:35:37 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: <56E03CF7.1010302@theqtcompany.com> References: <56E02095.2040506@theqtcompany.com> <56E03CF7.1010302@theqtcompany.com> Message-ID: Not sure this patch really does the expected behavior: with this patch, backspace will remove the *opening* quote, and the cursor will be before the quote instead of after it... It seems to me that the backspace should really just "undo" the auto-close quote in this case, e.g. remove the closing quote. -- Francois On Wed, Mar 9, 2016 at 4:10 PM, Nikolai Kosjar < nikolai.kosjar at theqtcompany.com> wrote: > 2. If backspacing over a double quote, do not remove both double quotes. >>> >> >> Agree, see above. >> > > https://codereview.qt-project.org/#/c/151822/ > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -------------- next part -------------- An HTML attachment was scrubbed... URL: From chighland at gmail.com Wed Mar 9 16:45:56 2016 From: chighland at gmail.com (Coda Highland) Date: Wed, 9 Mar 2016 07:45:56 -0800 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: On Wed, Mar 9, 2016 at 12:14 AM, Andreas Pakulat wrote: > Hi, > > On Tue, Mar 8, 2016 at 11:10 PM, Jason H wrote: >> >> A while back, I started a discussion about Creator's quoting policy. >> There was a lot of support, and a few people were concerned about changing >> the behavior that has been there since the beginning. >> >> However, I'm again tired of fighting with creator inserting "" when I type >> " and deleting both when I backspace over one. This behavior only occurs in >> QtCreator, and no other editor _ever_, and it needs to go. It is a "clever >> trick" that costs more than it is clever. At least for me, YMMV. > > > Thats actually not true, all kate-based editors/IDE's used to have the same > behavior. I haven't used those for a few years now so it may have been > changed. In addition Eclipse' Java editor also has almost the same behavior. > > That being said, QtC does have a few bugs that I believe Eclipse may not > have. For example, hitting " twice in a row does not insert 3 quotes, but > rather inserts a double-quote and then moves my cursor after the second > quote. This isn't a bug. MOST auto-pairing functions I've used have the behavior that it keeps track of the character it auto-inserted and if you type it in that position it assumes you were typing as if the auto-inserted one wasn't there. /s/ Adam From imikejackson at gmail.com Wed Mar 9 16:58:46 2016 From: imikejackson at gmail.com (Mike Jackson) Date: Wed, 09 Mar 2016 10:58:46 -0500 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: <56E02095.2040506@theqtcompany.com> <56E03CF7.1010302@theqtcompany.com> Message-ID: <56E04836.4000501@gmail.com> I agree. If I type a ' character I get 2 of them (''), then if I hit a backspace I delete the first one, but not the second one. How is this saving me typing? -- Mike Jackson [mike.jackson at bluequartz.net] > Typz > March 9, 2016 at 10:35 AM > Not sure this patch really does the expected behavior: with this > patch, backspace will remove the *opening* quote, and the cursor will > be before the quote instead of after it... It seems to me that the > backspace should really just "undo" the auto-close quote in this case, > e.g. remove the closing quote. > > -- > Francois > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > Nikolai Kosjar > March 9, 2016 at 10:10 AM > > https://codereview.qt-project.org/#/c/151822/ > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > Nikolai Kosjar > March 9, 2016 at 8:09 AM > Hi! > > Let me summarize for myself and others the ways we can add quotes to > some text. Default settings are assumed, the relevant one is: > > [x] Automatically insert matching characters. > > > a) Insert quote before text: > Type: "foo > This results in: "foo" > The closing quote was added after typing the initial ". > Looks fine. > > b) Insert quotes afterwards by using selection. > 1. Type: foo bar > 2. Select the typed text > 3. Hit " > This results in: "foo bar" > Looks fine. > > c) Insert quotes afterwards by starting with the closing quote > You have: foo bar > Type: " > This results in: foo bar"" > Ops 1: The second " was not intended. > Hit backspace. > Ops 2: Both quotes are removed and you are the beginning again. > > > Note that if you hit " and another one is added, your cursor is then > between the quotes: "|". Hitting backspace here removes then both > quotes. If the cursor is behind both quotes (e.g. you've moved it > there manually), only one quote is removed. This also applies to () > and [], but not to {} (thinking of lambdas we probably should include > that one, too). > > Although the option in the settings is named "Automatically insert > matching characters" it seems to also handle the removal of the pairs, > at least when the cursor is in-between. We probably should change this > if we do not introduce further options. > > I agree that we should address the "Ops 2" from above, that is not > removing both quotes if the cursor is in-between them. This will make > it inconsistent when looking at () and [], but I think that's > justified in this case. The problem here is the single quote by itself > - you can't tell whether it's an opening or closing one without > further information (typing ')' does not add you the opening one > automatically). > > Regarding "Ops 1", I don't think we can reliably detect whether the > users inserts an opening or closing quote. Your heuristic is not enough: > > On 03/08/2016 11:10 PM, Jason H wrote: >> Fundamentally, I want to changes to the behavior: >> 1. If the character after the cursor is not whitespace, do NOT double >> quote ("") > > Would break case a) from above for e.g. function arguments: > > foo(|) > foo(|,1) > >> 2. If backspacing over a double quote, do not remove both double quotes. > > Agree, see above. > > > Nikolai > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > Jason H > March 8, 2016 at 5:10 PM > A while back, I started a discussion about Creator's quoting policy. > There was a lot of support, and a few people were concerned about > changing the behavior that has been there since the beginning. > > However, I'm again tired of fighting with creator inserting "" when I > type " and deleting both when I backspace over one. This behavior only > occurs in QtCreator, and no other editor _ever_, and it needs to go. > It is a "clever trick" that costs more than it is clever. At least for > me, YMMV. > > Yes, there is a checkbox, but that applies to brackets and braces as well. > > Fundamentally, I want to changes to the behavior: > 1. If the character after the cursor is not whitespace, do NOT double > quote ("") > 2. If backspacing over a double quote, do not remove both double quotes. > > For the love of all that is binary!!! > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -------------- next part -------------- An HTML attachment was scrubbed... URL: From cristian.adam at gmail.com Wed Mar 9 17:17:25 2016 From: cristian.adam at gmail.com (Cristian Adam) Date: Wed, 9 Mar 2016 17:17:25 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: On Tue, Mar 8, 2016 at 11:10 PM, Jason H wrote: > A while back, I started a discussion about Creator's quoting policy. > There was a lot of support, and a few people were concerned about changing > the behavior that has been there since the beginning. > > However, I'm again tired of fighting with creator inserting "" when I type > " and deleting both when I backspace over one. This behavior only occurs in > QtCreator, and no other editor _ever_, and it needs to go. It is a "clever > trick" that costs more than it is clever. At least for me, YMMV. > > Yes, there is a checkbox, but that applies to brackets and braces as well. > > Fundamentally, I want to changes to the behavior: > 1. If the character after the cursor is not whitespace, do NOT double > quote ("") > 2. If backspacing over a double quote, do not remove both double quotes. > > For the love of all that is binary!!! > > It's way saner to disable this feature! ("Automatically insert matching characters") This is the first thing that I configure when I install a fresh Qt Creator (or any other text editor) on a new computer. We software developers write text for a living, it is very wise to invest in touch typing skills. I like the WYTIWYS (what you type is what you see) concept. Cheers, Cristian. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ariel at edis.mx Wed Mar 9 17:23:27 2016 From: ariel at edis.mx (Ariel Molina) Date: Wed, 9 Mar 2016 10:23:27 -0600 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: Message-ID: This behavior is confusing, there are some bugs filed against it. See: https://bugreports.qt.io/browse/QTCREATORBUG-14111 https://bugreports.qt.io/browse/QTCREATORBUG-2155 Ariel On Wed, Mar 9, 2016 at 9:45 AM, Coda Highland wrote: > On Wed, Mar 9, 2016 at 12:14 AM, Andreas Pakulat wrote: > > Hi, > > > > On Tue, Mar 8, 2016 at 11:10 PM, Jason H wrote: > >> > >> A while back, I started a discussion about Creator's quoting policy. > >> There was a lot of support, and a few people were concerned about > changing > >> the behavior that has been there since the beginning. > >> > >> However, I'm again tired of fighting with creator inserting "" when I > type > >> " and deleting both when I backspace over one. This behavior only > occurs in > >> QtCreator, and no other editor _ever_, and it needs to go. It is a > "clever > >> trick" that costs more than it is clever. At least for me, YMMV. > > > > > > Thats actually not true, all kate-based editors/IDE's used to have the > same > > behavior. I haven't used those for a few years now so it may have been > > changed. In addition Eclipse' Java editor also has almost the same > behavior. > > > > That being said, QtC does have a few bugs that I believe Eclipse may not > > have. For example, hitting " twice in a row does not insert 3 quotes, but > > rather inserts a double-quote and then moves my cursor after the second > > quote. > > This isn't a bug. MOST auto-pairing functions I've used have the > behavior that it keeps track of the character it auto-inserted and if > you type it in that position it assumes you were typing as if the > auto-inserted one wasn't there. > > /s/ Adam > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > http://edis.mx -------------- next part -------------- An HTML attachment was scrubbed... URL: From jhihn at gmx.com Wed Mar 9 18:27:08 2016 From: jhihn at gmx.com (Jason H) Date: Wed, 9 Mar 2016 18:27:08 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: <78C89B76-0FD1-4B06-B928-C22B8398B0F7@theqtcompany.com> References: <56E02095.2040506@theqtcompany.com>, <78C89B76-0FD1-4B06-B928-C22B8398B0F7@theqtcompany.com> Message-ID: It seems we are unanimous on the don't-remove-both-quotes-on-backspace. I am open to sme auto-quoting behavior, so let's explore this. As Eike starts to do: > > Would break case a) from above for e.g. function arguments: > > > > foo(|) > > foo(|,1) someFunc(" - ok to insert " and move cursor 1 space back: = someFunc("|" this is sometext" - NOT ok to insert " and move cursor 1 space back: = sometext"|" <- this won't ever happen in source, unless it's some kind of text decorator What I meant to do above is start with a ", however I missed it, and the auto insert behavior just makes it worse. It might be cool if it was really intelligent and figured out that due to EOL or whitespace to the right that it was the end quote and go find the best place for the open quote: this is sometext" - BOL = "this is sometext"| someFunc(this is sometext" - back up to (, advance to first non-whitespace, insert " = someFunc("this is sometext"| someFunc2(a, this is someText" - back up to ,, advance to first non-whitespace, insert " = someFunc(a, "this is sometext"| However that logic would fail with: // someFunc(QString longstring) someFunc( No. As I said, the other" -> someFunc( No, As I said, "the other" (which isn't a terrible guess. but without some rich AST, we can't do better) At which point, I'd argue that we're hurting as much as helping. (50%/50% on if we are right) If there is an AST, we can work backwards from the EOL " grabbing tokens that aren't symbols defined in the code, until we arrive at: someFunc( "No. As I said, the other" Absent the AST logic, I'd say do nothing. From shap at eros-os.org Wed Mar 9 19:07:45 2016 From: shap at eros-os.org (Jonathan S. Shapiro) Date: Wed, 9 Mar 2016 10:07:45 -0800 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: <56E02095.2040506@theqtcompany.com> <78C89B76-0FD1-4B06-B928-C22B8398B0F7@theqtcompany.com> Message-ID: I don't care one way or the other how this issue gets resolved, but I did want to offer an observation. Because it is surprising, it is helpful if modal behavior is as consistent as possible. In this case, the behavior involves how "backspace" behaves when the input looks like: "|" where '|' represents the position of the caret. There are several small suggestions that I wanted to make: 1. The behavior of backspace in the "inside double-quote" case should match the behavior in the "inside single quote" case. 2. It should perhaps *also* match the behavior in the "inside curly braces", "inside square braces" and "inside parentheses" cases. This is the point that I suspect will want some discussion. 3. I don't know if we have enough information in the editor to do so, but we should think about whether the modal backspace behavior should only occur between an opening and a closing pair. That is, we should think about what happens in the following case: "some"|"text" And we should also consider the cases: ""|" ""|"" Unless I am missing something in my thoughts, the modal backspace behavior does the right thing in these cases. It deletes the "wrong" quote (that is: the one *after* the caret), but the end result has the desured number of double quotes and the same caret position that the "normal" backspace would have produced. We also need to consider cases similar to: ab(|)xz What should be the behavior of backspace when the bracketing open and close characters do not "match". I actually think it is reasonable in this case to delete the closing bracket. It seems reasonable to me for [], (), and {}. Initially surprising, but something the use will quickly get used to. That is: it does not appear to me that any special heuristic "magic" is needed to deal with any of this. But I may be wrong, and it would be good to think these cases through to be sure. Good day to all, Jonathan Shapiro -------------- next part -------------- An HTML attachment was scrubbed... URL: From chighland at gmail.com Wed Mar 9 19:49:27 2016 From: chighland at gmail.com (Coda Highland) Date: Wed, 9 Mar 2016 10:49:27 -0800 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: <56E02095.2040506@theqtcompany.com> <78C89B76-0FD1-4B06-B928-C22B8398B0F7@theqtcompany.com> Message-ID: I think one thing that would make a big difference is if the behavior were keyed off of more than JUST the current state of the document and the position of the caret. As I alluded to upthread, the other editors I've used that have automatic brace/quote insertion only do magic on automatically-inserted brace/quote characters, and they highlight that character differently to indicate that its behavior is in fact magical -- if you type that character, it'll be overwritten, and if you backspace its mate, it'll be deleted. In all such cases, the magic goes away once you've navigated away from it. (The definition of "navigated away from" isn't precisely clear, but the visual highlighting at least makes the behavior obvious.) If you position your cursor next to a quote or brace that was already in the document, no magic should occur whatsoever. You should be able to edit the document without unexpected behaviors. /s/ Adam On Wed, Mar 9, 2016 at 10:07 AM, Jonathan S. Shapiro wrote: > I don't care one way or the other how this issue gets resolved, but I did > want to offer an observation. > > Because it is surprising, it is helpful if modal behavior is as consistent > as possible. In this case, the behavior involves how "backspace" behaves > when the input looks like: > > "|" > > where '|' represents the position of the caret. There are several small > suggestions that I wanted to make: > > 1. The behavior of backspace in the "inside double-quote" case should match > the behavior in the "inside single quote" case. > > 2. It should perhaps *also* match the behavior in the "inside curly braces", > "inside square braces" and "inside parentheses" cases. This is the point > that I suspect will want some discussion. > > 3. I don't know if we have enough information in the editor to do so, but we > should think about whether the modal backspace behavior should only occur > between an opening and a closing pair. That is, we should think about what > happens in the following case: > > "some"|"text" > > And we should also consider the cases: > > ""|" > ""|"" > > Unless I am missing something in my thoughts, the modal backspace behavior > does the right thing in these cases. It deletes the "wrong" quote (that is: > the one *after* the caret), but the end result has the desured number of > double quotes and the same caret position that the "normal" backspace would > have produced. > > We also need to consider cases similar to: > > ab(|)xz > > What should be the behavior of backspace when the bracketing open and close > characters do not "match". I actually think it is reasonable in this case to > delete the closing bracket. It seems reasonable to me for [], (), and {}. > Initially surprising, but something the use will quickly get used to. > > That is: it does not appear to me that any special heuristic "magic" is > needed to deal with any of this. But I may be wrong, and it would be good to > think these cases through to be sure. > > > Good day to all, > > > Jonathan Shapiro > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > From jhihn at gmx.com Wed Mar 9 20:55:43 2016 From: jhihn at gmx.com (Jason H) Date: Wed, 9 Mar 2016 20:55:43 +0100 Subject: [Qt-creator] Linguist In-Reply-To: <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> References: <1457476058.19124.1.camel@gmx.de>, <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: > > Hi, > > > > is there any good reason why linguist is not configured as a standard > > external tool? I'd like to add it. > > You can open a .ts file in Qt Linguist with “Open with... > Qt Linguist” (e.g. from the project tree, file system view, or from the document menu next to the file name above the editor (when it is already opened in an editor). I think his problem (and mine) is that it should be under Tools > External > Linguist. Right now your only two options are lupdate and lrelease. The GUI should be invokable from that last menu. From apoenitz at t-online.de Wed Mar 9 21:27:54 2016 From: apoenitz at t-online.de (=?iso-8859-1?Q?Andr=E9_P=F6nitz?=) Date: Wed, 9 Mar 2016 21:27:54 +0100 Subject: [Qt-creator] It's time to fix double redundant quotes In-Reply-To: References: <56E02095.2040506@theqtcompany.com> <78C89B76-0FD1-4B06-B928-C22B8398B0F7@theqtcompany.com> Message-ID: <20160309202754.GA2548@klara.mpi.htwm.de> On Wed, Mar 09, 2016 at 10:49:27AM -0800, Coda Highland wrote: > I think one thing that would make a big difference is if the behavior > were keyed off of more than JUST the current state of the document and > the position of the caret. > > As I alluded to upthread, the other editors I've used that have > automatic brace/quote insertion only do magic on > automatically-inserted brace/quote characters, and they highlight that > character differently to indicate that its behavior is in fact magical > -- if you type that character, it'll be overwritten, and if you > backspace its mate, it'll be deleted. > > In all such cases, the magic goes away once you've navigated away from > it. (The definition of "navigated away from" isn't precisely clear, > but the visual highlighting at least makes the behavior obvious.) > > If you position your cursor next to a quote or brace that was already > in the document, no magic should occur whatsoever. You should be able > to edit the document without unexpected behaviors. Now that you spell it out I think that "losing magic after navigation" is indeed the important difference between current and desirable behaviour. Andre' From adrozdoff at gmail.com Thu Mar 10 06:58:20 2016 From: adrozdoff at gmail.com (Alexander Drozdov) Date: Thu, 10 Mar 2016 15:58:20 +1000 Subject: [Qt-creator] CMake in QtC 4.0 [ex Qt Creator 4.0 soft-branched] In-Reply-To: <1457431272.1068.23.camel@theqtcompany.com> References: <1457431272.1068.23.camel@theqtcompany.com> Message-ID: 2016-03-08 20:01 GMT+10:00 Hunger Tobias : > > > 3) User needs an ability to force cache clean before next cmake run. It > > requires in some cases. Look my solution: > [skip] > Could you push that into codereview.qt-project.org? That makes it way > safer for > me to reuse some/all the code. > Review 1: https://codereview.qt-project.org/#/c/151873/ > > > 4) If you agree with showing all files in project tree and > > adding/removing/renaming files as I describe above, I can submit merge > > request tot the gerrit (and checks it work after last accounts merging > :-)) > > Some yesrs ago this changes was discarded. > > I would very much like the file scanning stuff, but the adding/removing > itself > will probably cause way to many bug reports for me to feel comfortable to > implement that at this time. > > Review 2 (scanning only): https://codereview.qt-project.org/#/c/151872/ -- WBR, Alexander Drozdov http://htrd.su -------------- next part -------------- An HTML attachment was scrubbed... URL: From mathias at taschenorakel.de Thu Mar 10 08:36:16 2016 From: mathias at taschenorakel.de (Mathias Hasselmann) Date: Thu, 10 Mar 2016 08:36:16 +0100 Subject: [Qt-creator] QtCreator 4.0: Where did my color icons go In-Reply-To: <56D9D733.5080608@gmail.com> References: <56D9D733.5080608@gmail.com> Message-ID: <56E123F0.7060008@taschenorakel.de> I just peeked over the fence. Is this about the dark frame theme? It's incomplete and doesn't blend well with Ubuntu's curvy Ambience theme, but actually I liked the look: The old icons feel very outdated after that short peeking, and I can see how such adoptions are needed to fit into today's Windows or OSX. Switching back for now because of the issues, but suddenly my desktop feels very oldish. Keep on the good work and don't let the naysayers discourage you. Ciao, Mathias Am 04.03.2016 um 19:42 schrieb Mike Jackson: > > I just compiled QtCreator 4.0 and what a surprise. It seems as if the > QtCreator team seems to have followed everyone else in UI design lately > and tossed out 30 years of Human Computer Interaction design. Namely, > you tossed colorful icons to go with a flat mono-chrome color palette. I > do not understand where this comes from? We programmers have been > fighting and waiting for years to get able to have high res icons that > can convey information quickly and when we get that capability all the > UI designers seem to have just given up on the concept. Color conveys > information quickly. It lets the human brain decipher information with > less effort than a monochrome palette. Numerous experts in the field > have written at length about this issue and where the UI design is going > and just how wrong it is. I urge the QtCreator team to at least include > a theme that retains the colorful icons from QtCreator 3.x in the 4.0 > release. It is not too late. > > > Let me put forth a very direct example. If you drive a white car and you > park that car in a parking lot with 1000 other white cars of the same > type (say 4 door passenger style cars), when you come back out to get in > your car, unless you parked your car in a very specific spot (first or > last) your brain needs to work much hard to remember exactly which car > is yours. Now imagine your car is red, in that same lot of 1000 white > cars. Finding your car is instantaneous. > > Thanks for listening. > -- > Mike Jackson > > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > From frederic.marchal at wowtechnology.com Thu Mar 10 19:09:11 2016 From: frederic.marchal at wowtechnology.com (=?UTF-8?B?RnLDqWTDqXJpYyBNYXJjaGFs?=) Date: Thu, 10 Mar 2016 19:09:11 +0100 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: 2016-03-09 20:55 GMT+01:00 Jason H : > >> > Hi, >> > >> > is there any good reason why linguist is not configured as a standard >> > external tool? I'd like to add it. >> >> You can open a .ts file in Qt Linguist with “Open with... > Qt Linguist” (e.g. from the project tree, file system view, or from the document menu next to the file name above the editor (when it is already opened in an editor). > > I think his problem (and mine) is that it should be under Tools > External > Linguist. Right now your only two options are lupdate and lrelease. The GUI should be invokable from that last menu. You can add it yourself. Click on Tools / External / Configure... Add a "Linguist" entry running %{CurrentProject:QT_INSTALL_BINS}\linguist.exe on Windows or %{CurrentProject:QT_INSTALL_BINS}/linguist.exe on Linux. If the ts files are in the project directory, you may also want to set the "Working directory" to %{CurrentProject:Path}. With this trick, Linguist will directly show the directory with the ts files when you click on File / Open. Frederic From chighland at gmail.com Thu Mar 10 19:22:10 2016 From: chighland at gmail.com (Coda Highland) Date: Thu, 10 Mar 2016 10:22:10 -0800 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: On Thu, Mar 10, 2016 at 10:09 AM, Frédéric Marchal wrote: > 2016-03-09 20:55 GMT+01:00 Jason H : >> >>> > Hi, >>> > >>> > is there any good reason why linguist is not configured as a standard >>> > external tool? I'd like to add it. >>> >>> You can open a .ts file in Qt Linguist with “Open with... > Qt Linguist” (e.g. from the project tree, file system view, or from the document menu next to the file name above the editor (when it is already opened in an editor). >> >> I think his problem (and mine) is that it should be under Tools > External > Linguist. Right now your only two options are lupdate and lrelease. The GUI should be invokable from that last menu. > > You can add it yourself. > > Click on Tools / External / Configure... > > Add a "Linguist" entry running > %{CurrentProject:QT_INSTALL_BINS}\linguist.exe on Windows or > %{CurrentProject:QT_INSTALL_BINS}/linguist.exe on Linux. > > If the ts files are in the project directory, you may also want to set > the "Working directory" to %{CurrentProject:Path}. With this trick, > Linguist will directly show the directory with the ts files when you > click on File / Open. > > Frederic > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator The question is why it's not there by default. /s/ Adam From jochen_becher at gmx.de Thu Mar 10 20:06:27 2016 From: jochen_becher at gmx.de (Jochen Becher) Date: Thu, 10 Mar 2016 20:06:27 +0100 Subject: [Qt-creator] Linguist In-Reply-To: <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: <1457636787.18028.2.camel@gmx.de> Oh, it didn't know this. Thank you. It is even better than using the Tools / External / Linguist menu. It's only a bit surprising that this menu does not contain an item to open Qt Linguist. Regards, Jochen Am Mittwoch, den 09.03.2016, 08:21 +0000 schrieb Ziller Eike: > > On Mar 8, 2016, at 11:27 PM, Jochen Becher > > wrote: > > > > Hi, > > > > is there any good reason why linguist is not configured as a > > standard > > external tool? I'd like to add it. > > You can open a .ts file in Qt Linguist with “Open with... > Qt > Linguist” (e.g. from the project tree, file system view, or from the > document menu next to the file name above the editor (when it is > already opened in an editor). > > Br, Eike > > > > > Regards, Jochen > > > > _______________________________________________ > > Qt-creator mailing list > > Qt-creator at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/qt-creator > > -- > Eike Ziller, Principle Software Engineer - The Qt Company GmbH > > The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin > Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja > Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht > Charlottenburg, HRB 144331 B > From Eike.Ziller at theqtcompany.com Fri Mar 11 09:13:52 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Fri, 11 Mar 2016 08:13:52 +0000 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: > On Mar 10, 2016, at 7:22 PM, Coda Highland wrote: > > On Thu, Mar 10, 2016 at 10:09 AM, Frédéric Marchal > wrote: >> 2016-03-09 20:55 GMT+01:00 Jason H : >>> >>>>> Hi, >>>>> >>>>> is there any good reason why linguist is not configured as a standard >>>>> external tool? I'd like to add it. >>>> >>>> You can open a .ts file in Qt Linguist with “Open with... > Qt Linguist” (e.g. from the project tree, file system view, or from the document menu next to the file name above the editor (when it is already opened in an editor). >>> >>> I think his problem (and mine) is that it should be under Tools > External > Linguist. Right now your only two options are lupdate and lrelease. The GUI should be invokable from that last menu. >> >> You can add it yourself. >> >> Click on Tools / External / Configure... >> >> Add a "Linguist" entry running >> %{CurrentProject:QT_INSTALL_BINS}\linguist.exe on Windows or >> %{CurrentProject:QT_INSTALL_BINS}/linguist.exe on Linux. >> >> If the ts files are in the project directory, you may also want to set >> the "Working directory" to %{CurrentProject:Path}. With this trick, >> Linguist will directly show the directory with the ts files when you >> click on File / Open. >> >> Frederic >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator > > The question is why it's not there by default. My question is more why the lrelease and lupdate items are in Tools > External, and not in the context menu for .pro files ;) But since I added them there..... I know that it was mostly as an example for the external tools, and how they can get access to state in Qt Creator (the current project in this case). Just opening Linguist is not there, because you can either open it with a .ts file directly, or use whatever OS means you have to open applications (e.g. Spotlight on OS X, Start menu search on Windows, and the search thingy in Ubuntu, all sound like a much quicker and more comfortable way than navigating through menus to Tools > External > Linguist > Open Linguist). I’m not against having it there though, so feel free to upload a change. Br, Eike -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From jhihn at gmx.com Fri Mar 11 17:56:08 2016 From: jhihn at gmx.com (Jason H) Date: Fri, 11 Mar 2016 17:56:08 +0100 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> , Message-ID: > > The question is why it's not there by default. > > My question is more why the lrelease and lupdate items are in Tools > External, and not in the context menu for .pro files ;) > But since I added them there..... I know that it was mostly as an example for the external tools, and how they can get access to state in Qt Creator (the current project in this case). > Just opening Linguist is not there, because you can either open it with a .ts file directly, or use whatever OS means you have to open applications (e.g. Spotlight on OS X, Start menu search on Windows, and the search thingy in Ubuntu, all sound like a much quicker and more comfortable way than navigating through menus to Tools > External > Linguist > Open Linguist). > I’m not against having it there though, so feel free to upload a change. My question is why we have to deal with lrelease at all? Seems like there should be a qmake auto build for .qm updates from .ts? On OSX, Linguist is not in my spotlight :-( Anyone else in the same boat? Also it takes time to launch Spotlight, type 'lin' and get the results, it's jarring experience. If I could launch that in Creator, that'd be so much better. (I'm on a recent mid-range Mac mini, no SSD for me!) Maybe Creator could get a "Translation" button in the left sidebar [Welcome/Edit/Design/Debug/TRANSLATE/Projects/Analyze/Help] and a native plug-in? Alternative is a "Translations" project view option [Projects/Open Documents/Bookmarks/.../Translations] which shows the normal projects pane file list but also shows the strings in the files,with the editor window the Translation pane from Linugust. Don't get me wrong, Linguist needs to be kept, and work stand-alone so we don't have to give it to translators. But being able to preview the QML screens with translations would be very cool. But given benefit-reward ratio, I'd just like to launch it from Creator. From ekke at ekkes-corner.org Fri Mar 11 18:09:32 2016 From: ekke at ekkes-corner.org (ekke) Date: Fri, 11 Mar 2016 18:09:32 +0100 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: <56E2FBCC.8030207@ekkes-corner.org> Am 11.03.16 um 17:56 schrieb Jason H: >>> The question is why it's not there by default. >> My question is more why the lrelease and lupdate items are in Tools > External, and not in the context menu for .pro files ;) >> But since I added them there..... I know that it was mostly as an example for the external tools, and how they can get access to state in Qt Creator (the current project in this case). >> Just opening Linguist is not there, because you can either open it with a .ts file directly, or use whatever OS means you have to open applications (e.g. Spotlight on OS X, Start menu search on Windows, and the search thingy in Ubuntu, all sound like a much quicker and more comfortable way than navigating through menus to Tools > External > Linguist > Open Linguist). >> I’m not against having it there though, so feel free to upload a change. > My question is why we have to deal with lrelease at all? Seems like there should be a qmake auto build for .qm updates from .ts? would be cool to get ts / qm files "by magic" coming from BlackBerry Cascades where Eclipse Momentics IDE only needs to add the languages to the project and then does this stuff "under the hood". ... was not easy to start with QtCreator workflow for translations for me see also http://forum.qt.io/topic/64123/localizing-qt-quick-apps > > On OSX, Linguist is not in my spotlight :-( Anyone else in the same boat? Also it takes time to launch Spotlight, type 'lin' and get the results, it's jarring experience. If I could launch that in Creator, that'd be so much better. (I'm on a recent mid-range Mac mini, no SSD for me!) > > Maybe Creator could get a "Translation" button in the left sidebar [Welcome/Edit/Design/Debug/TRANSLATE/Projects/Analyze/Help] and a native plug-in? > Alternative is a "Translations" project view option [Projects/Open Documents/Bookmarks/.../Translations] which shows the normal projects pane file list but also shows the strings in the files,with the editor window the Translation pane from Linugust. Don't get me wrong, Linguist needs to be kept, and work stand-alone so we don't have to give it to translators. But being able to preview the QML screens with translations would be very cool. > > But given benefit-reward ratio, I'd just like to launch it from Creator. > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- ekke (ekkehard gentz) independent software architect international development native mobile business apps BlackBerry 10 | Qt Mobile (Android, iOS) twitter: @ekkescorner LinkedIn: http://linkedin.com/in/ekkehard/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From tobias.hunger at gmail.com Sat Mar 12 08:41:23 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Sat, 12 Mar 2016 08:41:23 +0100 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: Am 11.03.2016 17:56 schrieb "Jason H" : > On OSX, Linguist is not in my spotlight :-( Anyone else in the same boat? Also it takes time to launch Spotlight, type 'lin' and get the results, it's jarring experience. If I could launch that in Creator, that'd be so much better. (I'm on a recent mid-range Mac mini, no SSD for me!) Just configure an external tool, just like Eike suggested earlier. If it works well for you, you can always contribute a patch enabling that for everybody. > Maybe Creator could get a "Translation" button in the left sidebar [Welcome/Edit/Design/Debug/TRANSLATE/Projects/Analyze/Help] and a native plug-in? So far we did not consider that: In our experience translations are not handled by developers most of the time, but by external translation services. So we think it makes little sense to use space for translation in a developer tool. The external services are better off with a tool much simpler than Qt Creator, too. Why would we want to bother them with a UI designer and code editors? Some way to preview UIs I different languages would indeed be cool though:-) Not just for QML, but also for widget UIs. Best regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From ekke at ekkes-corner.org Sat Mar 12 09:30:43 2016 From: ekke at ekkes-corner.org (ekke) Date: Sat, 12 Mar 2016 09:30:43 +0100 Subject: [Qt-creator] Linguist In-Reply-To: References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> Message-ID: <56E3D3B3.7080005@ekkes-corner.org> Am 12.03.16 um 08:41 schrieb Tobias Hunger: > > > Am 11.03.2016 17:56 schrieb "Jason H" >: > ..... > > > Maybe Creator could get a "Translation" button in the left sidebar > [Welcome/Edit/Design/Debug/TRANSLATE/Projects/Analyze/Help] and a > native plug-in? > > So far we did not consider that: In our experience translations are > not handled by developers most of the time, but by external > translation services. So we think it makes little sense to use space > for translation in a developer tool. > > .... this depends... most mobile app developers I know are also doing translations same for me: I'm always developing in english and immediately translate into german having some devices with english locale and some with german locale it's easy to see how it fits and also if I forgot a tr() or qsTr() so I'm always switching thanks to this thread I learned that I can configure this ekke -------------- next part -------------- An HTML attachment was scrubbed... URL: From papa at arbolone.ca Sat Mar 12 14:16:03 2016 From: papa at arbolone.ca (=?utf-8?Q?Pap=C3=A1?=) Date: Sat, 12 Mar 2016 08:16:03 -0500 Subject: [Qt-creator] (no subject) Message-ID: <20160312131604.E8F8AA06BF@relay.mailchannels.net> In my Laptop, the IDE's font is very, veeeeery small, but that is the case with most applications in my Laptop. However, for Notepad, cmd and other applications I was able to increase the font size directly into the application, the win10 apps don't let you do this, :( example1 example2 Perhaps the Qt Creator developers could provide a way to increase the font size of it. Thanks in advance and keep up the good work! Sent from Mail for Windows 10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From pascom at orange.fr Sat Mar 12 21:49:08 2016 From: pascom at orange.fr (Pascal COMBES) Date: Sat, 12 Mar 2016 21:49:08 +0100 Subject: [Qt-creator] (no subject) In-Reply-To: <20160312131604.E8F8AA06BF@relay.mailchannels.net> References: <20160312131604.E8F8AA06BF@relay.mailchannels.net> Message-ID: <56E480C4.1090804@orange.fr> Le 12/03/2016 14:16, Papá a écrit : > > In my Laptop, the IDE's font is very, veeeeery small, but that is the > case with most applications in my Laptop. However, for Notepad, cmd > and other applications I was able to increase the font size directly > into the application, the win10 apps don't let you do this, :( > example1 > > example2 > > Perhaps the Qt Creator developers could provide a way to increase the > font size of it. > Thanks in advance and keep up the good work! > > > Hello, You can increase or decrease the font size in the code editor with Ctrl+mouse scroll or Ctrl++ and Ctrl+-. However this affects only the font size in the editor. I don't know if it is what you want. Alternatively, if all your applications are too small you can consider decreasing the resolution of your screen in windows settings (I don't know where these settings are in Windows 10. Best regards, Pascal. From realnc at gmail.com Sun Mar 13 09:27:22 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 13 Mar 2016 10:27:22 +0200 Subject: [Qt-creator] (no subject) In-Reply-To: <20160312131604.E8F8AA06BF@relay.mailchannels.net> References: <20160312131604.E8F8AA06BF@relay.mailchannels.net> Message-ID: On 12/03/16 15:16, Papá wrote: > In my Laptop, the IDE's font is very, veeeeery small, but that is the > case with most applications in my Laptop. However, for Notepad, cmd and > other applications I was able to increase the font size directly into > the application, the win10 apps don't let you do this, :( Give this a try: http://www.windows10update.com/2015/05/windows-10-tutorials-79-change-dpi-scaling-level-for-displays/ From realnc at gmail.com Sun Mar 13 09:34:00 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 13 Mar 2016 10:34:00 +0200 Subject: [Qt-creator] Clang 3.8 and Creator code model Message-ID: I upgraded to LLVM and Clang 3.8.0. The Clang code model doesn't work anymore. There are no error messages anywhere. It just silently doesn't do anything (no auto-completion, no code checking, nothing.) Is this supposed to happen? From atelszewski at gmail.com Sun Mar 13 09:36:53 2016 From: atelszewski at gmail.com (Andrzej Telszewski) Date: Sun, 13 Mar 2016 09:36:53 +0100 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: Message-ID: <56E526A5.1050502@gmail.com> On 13/03/16 09:34, Nikos Chantziaras wrote: > I upgraded to LLVM and Clang 3.8.0. The Clang code model doesn't work > anymore. There are no error messages anywhere. It just silently doesn't > do anything (no auto-completion, no code checking, nothing.) > > Is this supposed to happen? I also updated to LLVM and Clang 3.8.0 and completion stopped working. Recompiling Qt Creator solved the problem. -- Best regards, Andrzej Telszewski From realnc at gmail.com Sun Mar 13 10:15:12 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 13 Mar 2016 11:15:12 +0200 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: <56E526A5.1050502@gmail.com> References: <56E526A5.1050502@gmail.com> Message-ID: On 13/03/16 10:36, Andrzej Telszewski wrote: > On 13/03/16 09:34, Nikos Chantziaras wrote: >> I upgraded to LLVM and Clang 3.8.0. The Clang code model doesn't work >> anymore. There are no error messages anywhere. It just silently doesn't >> do anything (no auto-completion, no code checking, nothing.) >> >> Is this supposed to happen? > > I also updated to LLVM and Clang 3.8.0 and completion stopped working. > Recompiling Qt Creator solved the problem. I just recompiled Creator (I'm 3.6.0). It didn't help at all :-/ From realnc at gmail.com Sun Mar 13 10:26:49 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 13 Mar 2016 11:26:49 +0200 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: <56E526A5.1050502@gmail.com> Message-ID: On 13/03/16 11:15, Nikos Chantziaras wrote: > On 13/03/16 10:36, Andrzej Telszewski wrote: >> On 13/03/16 09:34, Nikos Chantziaras wrote: >>> I upgraded to LLVM and Clang 3.8.0. The Clang code model doesn't work >>> anymore. [...] >> >> I also updated to LLVM and Clang 3.8.0 and completion stopped working. >> Recompiling Qt Creator solved the problem. > > I just recompiled Creator (I'm 3.6.0). It didn't help at all :-/ Let me correct that. It started working (it seemed it needed to reparse the sources first.) Code checking and completion works now, except for one important (to me) detail: Typing "." does not automatically convert to "->" anymore when using raw pointers. Any known fix for that? From tobias.hunger at gmail.com Sun Mar 13 12:09:44 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Sun, 13 Mar 2016 12:09:44 +0100 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: <56E526A5.1050502@gmail.com> Message-ID: Hi, as far as I know we will stick with clang 3.6 for our Qt Creator 4.0 release. I am sure Eike will correct me on Monday if I am wrong:-) Clang 3.7 has some bugs that breaks the code model completely. Clang 3.8 has some features we want, but also some changes we need to incorporate first. That is still missing AFAIK. Best regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From flexferrum at gmail.com Sun Mar 13 13:16:20 2016 From: flexferrum at gmail.com (Sergei Sadovnikov) Date: Sun, 13 Mar 2016 15:16:20 +0300 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: <56E526A5.1050502@gmail.com> Message-ID: Hi Tobias, Could you explain: what special features do you want from Clang 3.8? I looked into release notes and didn't fine any notices about libclang/codecompletion/codemodel. Best regards, Sergei. 2016-03-13 14:09 GMT+03:00 Tobias Hunger : > Hi, > > as far as I know we will stick with clang 3.6 for our Qt Creator 4.0 > release. I am sure Eike will correct me on Monday if I am wrong:-) > > Clang 3.7 has some bugs that breaks the code model completely. > > Clang 3.8 has some features we want, but also some changes we need to > incorporate first. That is still missing AFAIK. > > Best regards, > Tobias > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From papa at arbolone.ca Mon Mar 14 00:05:58 2016 From: papa at arbolone.ca (=?utf-8?Q?Pap=C3=A1?=) Date: Sun, 13 Mar 2016 19:05:58 -0400 Subject: [Qt-creator] (no subject) In-Reply-To: References: <20160312131604.E8F8AA06BF@relay.mailchannels.net> Message-ID: <20160313230559.C80CE169F3B@relay.mailchannels.net> Thanks Nikos, I tried it but to no avail. No matter how much I increase the or reduce the DPI, Qt Creator’s menus do not change. I guess this is more of a Engineering challenge than I setting/users settings. To the benefit of the Developers, this problem does not occur in my Desktop. Have a good week. Sent from Mail for Windows 10 From: Nikos Chantziaras Sent: Sunday, March 13, 2016 4:28 AM To: qt-creator at qt-project.org Subject: Re: [Qt-creator] (no subject) On 12/03/16 15:16, Papá wrote: > In my Laptop, the IDE's font is very, veeeeery small, but that is the > case with most applications in my Laptop. However, for Notepad, cmd and > other applications I was able to increase the font size directly into > the application, the win10 apps don't let you do this, :( Give this a try: http://www.windows10update.com/2015/05/windows-10-tutorials-79-change-dpi-scaling-level-for-displays/ _______________________________________________ Qt-creator mailing list Qt-creator at qt-project.org http://lists.qt-project.org/mailman/listinfo/qt-creator -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Mon Mar 14 07:02:33 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Mon, 14 Mar 2016 08:02:33 +0200 Subject: [Qt-creator] (no subject) In-Reply-To: <20160313230559.C80CE169F3B@relay.mailchannels.net> References: <20160312131604.E8F8AA06BF@relay.mailchannels.net> <20160313230559.C80CE169F3B@relay.mailchannels.net> Message-ID: What happens if you set: QT_DEVICE_PIXEL_RATIO=2 in the environment before running Creator? This should make everything double-sized. On 14/03/16 01:05, Papá wrote: > Thanks Nikos, I tried it but to no avail. No matter how much I increase > the or reduce the DPI, Qt Creator’s menus do not change. I guess this is > more of a Engineering challenge than I setting/users settings. To the > benefit of the Developers, this problem does not occur in my Desktop. > > Have a good week. > > Sent from Mail for > Windows 10 > > *From: *Nikos Chantziaras > *Sent: *Sunday, March 13, 2016 4:28 AM > *To: *qt-creator at qt-project.org > *Subject: *Re: [Qt-creator] (no subject) > > On 12/03/16 15:16, Papá wrote: > > > In my Laptop, the IDE's font is very, veeeeery small, but that is the > > > case with most applications in my Laptop. However, for Notepad, cmd and > > > other applications I was able to increase the font size directly into > > > the application, the win10 apps don't let you do this, :( > > Give this a try: > > http://www.windows10update.com/2015/05/windows-10-tutorials-79-change-dpi-scaling-level-for-displays/ > > _______________________________________________ > > Qt-creator mailing list > > Qt-creator at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/qt-creator > > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > From Eike.Ziller at theqtcompany.com Mon Mar 14 08:54:57 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Mon, 14 Mar 2016 07:54:57 +0000 Subject: [Qt-creator] Linguist In-Reply-To: <56E2FBCC.8030207@ekkes-corner.org> References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> <56E2FBCC.8030207@ekkes-corner.org> Message-ID: <58973561-AB7F-42A2-968F-F1873DDA79CD@theqtcompany.com> > On Mar 11, 2016, at 6:09 PM, ekke wrote: > > Am 11.03.16 um 17:56 schrieb Jason H: >>>> The question is why it's not there by default. >>>> >>> My question is more why the lrelease and lupdate items are in Tools > External, and not in the context menu for .pro files ;) >>> But since I added them there..... I know that it was mostly as an example for the external tools, and how they can get access to state in Qt Creator (the current project in this case). >>> Just opening Linguist is not there, because you can either open it with a .ts file directly, or use whatever OS means you have to open applications (e.g. Spotlight on OS X, Start menu search on Windows, and the search thingy in Ubuntu, all sound like a much quicker and more comfortable way than navigating through menus to Tools > External > Linguist > Open Linguist). >>> I’m not against having it there though, so feel free to upload a change. >>> >> My question is why we have to deal with lrelease at all? Seems like there should be a qmake auto build for .qm updates from .ts? > would be cool to get ts / qm files "by magic” > coming from BlackBerry Cascades where Eclipse Momentics IDE only needs to add the languages to the project and then does this stuff "under the hood". > ... was not easy to start with QtCreator workflow for translations for me I’d claim that that is more the “Qt” or “qmake” workflow. For Qt Creator itself we manually added the automatic “releasing” of translations to the .pro files. I agree that it’s ugly that this is necessary. (See http://code.qt.io/cgit/qt-creator/qt-creator.git/tree/share/qtcreator/translations/translations.pro#n122 most of that file is because of special stuff we do in Qt Creator though (translating .json files etc), so the most relevant is the updateqm extra compiler line 122ff.) > see also > http://forum.qt.io/topic/64123/localizing-qt-quick-apps > >> >> On OSX, Linguist is not in my spotlight :-( Anyone else in the same boat? Also it takes time to launch Spotlight, type 'lin' and get the results, it's jarring experience. If I could launch that in Creator, that'd be so much better. (I'm on a recent mid-range Mac mini, no SSD for me!) Well, out of the box you can open Linguist with a specific .ts file from Qt Creator ;) >> Maybe Creator could get a "Translation" button in the left sidebar [Welcome/Edit/Design/Debug/TRANSLATE/Projects/Analyze/Help] and a native plug-in? >> Alternative is a "Translations" project view option [Projects/Open Documents/Bookmarks/.../Translations] which shows the normal projects pane file list but also shows the strings in the files,with the editor window the Translation pane from Linugust. The suggestion to “integrate” Linguist into Qt Creator exists basically since the existence of Qt Creator, but the effort involved in that would far outweigh the actual usefulness... >> Don't get me wrong, Linguist needs to be kept, and work stand-alone so we don't have to give it to translators. But being able to preview the QML screens with translations would be very cool. >> >> But given benefit-reward ratio, I'd just like to launch it from Creator. >> >> >> _______________________________________________ >> Qt-creator mailing list >> >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator > > > -- > ekke (ekkehard gentz) > > independent software architect > international development native mobile business apps > BlackBerry 10 | Qt Mobile (Android, iOS) > twitter: @ekkescorner > LinkedIn: http://linkedin.com/in/ekkehard/ > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Eike.Ziller at theqtcompany.com Mon Mar 14 09:01:09 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Mon, 14 Mar 2016 08:01:09 +0000 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: <56E526A5.1050502@gmail.com> Message-ID: <53672EEF-50A9-4EBD-8564-8C1D22305953@theqtcompany.com> > On Mar 13, 2016, at 10:26 AM, Nikos Chantziaras wrote: > > On 13/03/16 11:15, Nikos Chantziaras wrote: >> On 13/03/16 10:36, Andrzej Telszewski wrote: >>> On 13/03/16 09:34, Nikos Chantziaras wrote: >>>> I upgraded to LLVM and Clang 3.8.0. The Clang code model doesn't work >>>> anymore. [...] >>> >>> I also updated to LLVM and Clang 3.8.0 and completion stopped working. >>> Recompiling Qt Creator solved the problem. >> >> I just recompiled Creator (I'm 3.6.0). It didn't help at all :-/ > > Let me correct that. It started working (it seemed it needed to reparse the sources first.) > > Code checking and completion works now, except for one important (to me) detail: Typing "." does not automatically convert to "->" anymore when using raw pointers. > > Any known fix for that? Afaik that feature has only been implemented for the Clang code model in 4.0. Br, Eike -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Eike.Ziller at theqtcompany.com Mon Mar 14 09:01:09 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Mon, 14 Mar 2016 08:01:09 +0000 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: <56E526A5.1050502@gmail.com> Message-ID: <9A97DE28-4E73-49D9-A4B2-8A333B273942@theqtcompany.com> > On Mar 13, 2016, at 10:26 AM, Nikos Chantziaras wrote: > > On 13/03/16 11:15, Nikos Chantziaras wrote: >> On 13/03/16 10:36, Andrzej Telszewski wrote: >>> On 13/03/16 09:34, Nikos Chantziaras wrote: >>>> I upgraded to LLVM and Clang 3.8.0. The Clang code model doesn't work >>>> anymore. [...] >>> >>> I also updated to LLVM and Clang 3.8.0 and completion stopped working. >>> Recompiling Qt Creator solved the problem. >> >> I just recompiled Creator (I'm 3.6.0). It didn't help at all :-/ > > Let me correct that. It started working (it seemed it needed to reparse the sources first.) > > Code checking and completion works now, except for one important (to me) detail: Typing "." does not automatically convert to "->" anymore when using raw pointers. > > Any known fix for that? Afaik that feature has only been implemented for the Clang code model in 4.0. Br, Eike -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From Eike.Ziller at theqtcompany.com Mon Mar 14 09:06:50 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Mon, 14 Mar 2016 08:06:50 +0000 Subject: [Qt-creator] (no subject) In-Reply-To: References: <20160312131604.E8F8AA06BF@relay.mailchannels.net> <20160313230559.C80CE169F3B@relay.mailchannels.net> Message-ID: > On Mar 14, 2016, at 7:02 AM, Nikos Chantziaras wrote: > > What happens if you set: > > QT_DEVICE_PIXEL_RATIO=2 > > in the environment before running Creator? This should make everything double-sized. And you might want to try using the Qt 5.6 based Qt Creator 3.6.1 (to be released soon, snapshots can be found here: http://download.qt.io/snapshots/qtcreator/3.6.1/latest/) and use the environment variable QT_SCALE_FACTOR=2 Br, Eike > On 14/03/16 01:05, Papá wrote: >> Thanks Nikos, I tried it but to no avail. No matter how much I increase >> the or reduce the DPI, Qt Creator’s menus do not change. I guess this is >> more of a Engineering challenge than I setting/users settings. To the >> benefit of the Developers, this problem does not occur in my Desktop. >> >> Have a good week. >> >> Sent from Mail for >> Windows 10 >> >> *From: *Nikos Chantziaras >> *Sent: *Sunday, March 13, 2016 4:28 AM >> *To: *qt-creator at qt-project.org >> *Subject: *Re: [Qt-creator] (no subject) >> >> On 12/03/16 15:16, Papá wrote: >> >> > In my Laptop, the IDE's font is very, veeeeery small, but that is the >> >> > case with most applications in my Laptop. However, for Notepad, cmd and >> >> > other applications I was able to increase the font size directly into >> >> > the application, the win10 apps don't let you do this, :( >> >> Give this a try: >> >> http://www.windows10update.com/2015/05/windows-10-tutorials-79-change-dpi-scaling-level-for-displays/ >> >> _______________________________________________ >> >> Qt-creator mailing list >> >> Qt-creator at qt-project.org >> >> http://lists.qt-project.org/mailman/listinfo/qt-creator >> >> >> >> _______________________________________________ >> Qt-creator mailing list >> Qt-creator at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/qt-creator >> > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From ekke at ekkes-corner.org Mon Mar 14 09:12:48 2016 From: ekke at ekkes-corner.org (ekke) Date: Mon, 14 Mar 2016 09:12:48 +0100 Subject: [Qt-creator] Linguist In-Reply-To: <58973561-AB7F-42A2-968F-F1873DDA79CD@theqtcompany.com> References: <1457476058.19124.1.camel@gmx.de> <9929C6D6-4CB6-4D8B-B3F9-AD58A5857F08@theqtcompany.com> <56E2FBCC.8030207@ekkes-corner.org> <58973561-AB7F-42A2-968F-F1873DDA79CD@theqtcompany.com> Message-ID: <56E67280.1080708@ekkes-corner.org> Am 14.03.16 um 08:54 schrieb Ziller Eike: >> .... >> would be cool to get ts / qm files "by magic” >> coming from BlackBerry Cascades where Eclipse Momentics IDE only needs to add the languages to the project and then does this stuff "under the hood". >> ... was not easy to start with QtCreator workflow for translations for me > I’d claim that that is more the “Qt” or “qmake” workflow. > > For Qt Creator itself we manually added the automatic “releasing” of translations to the .pro files. > I agree that it’s ugly that this is necessary. > (See http://code.qt.io/cgit/qt-creator/qt-creator.git/tree/share/qtcreator/translations/translations.pro#n122 > most of that file is because of special stuff we do in Qt Creator though (translating .json files etc), so the most > relevant is the updateqm extra compiler line 122ff.) > > .... thanks for this helpful info From tobias.hunger at gmail.com Mon Mar 14 11:40:53 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Mon, 14 Mar 2016 11:40:53 +0100 Subject: [Qt-creator] Clang 3.8 and Creator code model In-Reply-To: References: <56E526A5.1050502@gmail.com> Message-ID: Hello Sergei, On Sun, Mar 13, 2016 at 1:16 PM, Sergei Sadovnikov wrote: > Hi Tobias, > > Could you explain: what special features do you want from Clang 3.8? I > looked into release notes and didn't fine any notices about > libclang/codecompletion/codemodel. Sorry, but I can not. I was just repeating what I was told when I complained about issues with using clang-trunk with Qt Creator a while back:-) Best Regards, Tobias From ariel at edis.mx Mon Mar 14 18:16:37 2016 From: ariel at edis.mx (Ariel Molina) Date: Mon, 14 Mar 2016 11:16:37 -0600 Subject: [Qt-creator] (no subject) In-Reply-To: References: <20160312131604.E8F8AA06BF@relay.mailchannels.net> Message-ID: Yes, your problem seems to be you upscaled some of your System Fonts and you do it manually, not that QtC downscaled it's own fonts. On Sun, Mar 13, 2016 at 2:27 AM, Nikos Chantziaras wrote: > On 12/03/16 15:16, Papá wrote: > >> In my Laptop, the IDE's font is very, veeeeery small, but that is the >> case with most applications in my Laptop. However, for Notepad, cmd and >> other applications I was able to increase the font size directly into >> the application, the win10 apps don't let you do this, :( >> > > Give this a try: > > > http://www.windows10update.com/2015/05/windows-10-tutorials-79-change-dpi-scaling-level-for-displays/ > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- Ariel Molina R. Oficina: +52 (222) 3723196 Movil: +521 2226 758874 http://edis.mx -------------- next part -------------- An HTML attachment was scrubbed... URL: From Eike.Ziller at theqtcompany.com Thu Mar 17 14:05:04 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Thu, 17 Mar 2016 13:05:04 +0000 Subject: [Qt-creator] Dev: Bumping Qt requirement of Qt Creator 4.0 Message-ID: <6ADD1178-3058-4E87-BA4C-F9FB580A43C8@theqtcompany.com> In the light of the release of Qt 5.6, I suggest that we bump the minimum required Qt version for Qt Creator 4.0 (and master) to Qt 5.5. Br, Eike -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From freddy311082 at gmail.com Fri Mar 18 21:34:18 2016 From: freddy311082 at gmail.com (Freddy Martinez Garcia) Date: Fri, 18 Mar 2016 17:34:18 -0300 Subject: [Qt-creator] Valgrind and Qt Creator Message-ID: Guys... i'm using qt-creator with non qt project and when I run the valgrind everything is perfect, but If i run a qt project with valgrind, my project doesn't start.. Is this the right place to ask about it or I have to go to the valgrind mail list ? regards *============================================="El tamaño de tus logros depende del tamaño de tus metas." * *C++ and Qt Senior Developer* *Lic. Computer Science* *Buenos Aires, Argentina* -------------- next part -------------- An HTML attachment was scrubbed... URL: From apoenitz at t-online.de Fri Mar 18 23:13:28 2016 From: apoenitz at t-online.de (=?iso-8859-1?Q?Andr=E9_P=F6nitz?=) Date: Fri, 18 Mar 2016 23:13:28 +0100 Subject: [Qt-creator] Valgrind and Qt Creator In-Reply-To: References: Message-ID: <20160318221328.GA2589@klara.mpi.htwm.de> On Fri, Mar 18, 2016 at 05:34:18PM -0300, Freddy Martinez Garcia wrote: > Guys... i'm using qt-creator with non qt project and when I run the > valgrind everything is perfect, but If i run a qt project with valgrind, my > project doesn't start.. Is this the right place to ask about it or I have > to go to the valgrind mail list ? A valgrind list would be the right place to ask if your application doesn't work with valgrind on the command line. Assuming it does work there, this here (or even bugreports.qt-project.org) is the place to ask about issues with the Qt Creator integration. What version of Qt Creator, valgrind and your OS are we talking about? Is there any output in tha Application Output pane or anything else that looks interesting? Andre' From freddy311082 at gmail.com Sat Mar 19 01:50:10 2016 From: freddy311082 at gmail.com (Freddy Martinez) Date: Fri, 18 Mar 2016 21:50:10 -0300 Subject: [Qt-creator] Valgrind and Qt Creator In-Reply-To: <20160318221328.GA2589@klara.mpi.htwm.de> References: <20160318221328.GA2589@klara.mpi.htwm.de> Message-ID: <453599F0-1B5B-4360-B345-AF9C16EE8F22@gmail.com> Hi Andre… this is the valgrind output freddy at Freddys-MacBook-Pro:/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib$ valgrind ./nautam-server ==99053== Memcheck, a memory error detector ==99053== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al. ==99053== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info ==99053== Command: ./nautam-server ==99053== --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommon.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsystem_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libtask_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libforms_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libformula_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libreports_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libdb_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/librules_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsecurity_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libqueries_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommunication_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libprocedure_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperspectives_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperiods_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsequence_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcryptopp.1.0.0.dylib" ==99053== Conditional jump or move depends on uninitialised value(s) ==99053== at 0x7FFF5FC24A87: bcmp (in /usr/lib/dyld) ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== ==99053== Use of uninitialised value of size 8 ==99053== at 0x7FFF5FC24A9F: bcmp (in /usr/lib/dyld) ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== ==99053== Use of uninitialised value of size 8 ==99053== at 0x7FFF5FC24AA4: bcmp (in /usr/lib/dyld) ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== ==99053== Conditional jump or move depends on uninitialised value(s) ==99053== at 0x7FFF5FC11977: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0E8CD: ImageLoader::recursiveLoadLibraries(ImageLoader::LinkContext const&, bool, ImageLoader::RPathChain const&) (in /usr/lib/dyld) ==99053== --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 2 times) --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 4 times) --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 8 times) --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 16 times) valgrind: m_syswrap/syswrap-amd64-darwin.c:507 (void wqthread_hijack(Addr, Addr, Addr, Addr, Int, Addr)): Assertion 'tst->os_state.pthread - magic_delta == self' failed. host stacktrace: ==99053== at 0x238040B6C: ??? ==99053== by 0x238040F77: ??? ==99053== by 0x238040F5A: ??? ==99053== by 0x2380DFCD9: ??? sched status: running_tid=0 Thread 1: status = VgTs_WaitSys (lwpid 2567) ==99053== at 0x10449E3C2: semaphore_wait_trap (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x1042A85D5: _dispatch_group_wait_slow (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x102AE777D: CFPREFERENCES_IS_WAITING_FOR_CFPREFSD (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102AE7897: -[CFPrefsSearchListSource generationCountFromListOfSources:count:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102996CB3: -[CFPrefsSearchListSource alreadylocked_copyDictionary] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x10299198B: -[CFPrefsSearchListSource alreadylocked_copyValueForKey:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102B1964B: ___CFPreferencesCopyAppValueWithContainer_block_invoke (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102AE674F: +[CFPrefsSearchListSource withSearchListForIdentifier:container:perform:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102B195C6: _CFPreferencesCopyAppValueWithContainer (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A3E92: ___CFBundleCopyUserLanguages_block_invoke (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1042A033E: _dispatch_client_callout (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A0236: dispatch_once_f (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x102A87FAA: _CFBundleCopyLanguageSearchListInBundle (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102A62892: _copyQueryTable (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102A62071: _copyResourceURLsFromBundle (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A22E5: _CFBundleCopyFindResources (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A1E1F: CFBundleCopyResourceURL (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A1CC2: CFBundleGetLocalInfoDictionary (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A1C30: CFBundleGetValueForInfoDictionaryKey (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1021AE88B: QCoreApplicationPrivate::macMenuBarName() (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x1021AE957: QCoreApplicationPrivate::appName() const (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x1021B047D: QCoreApplication::init() (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x1021B06BA: QCoreApplication::QCoreApplication(int&, char**, int) (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x10003C5B8: main (in ./nautam-server) Thread 2: status = VgTs_WaitSys (lwpid 6147) ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x2380DFCD9: ??? ==99053== by 0x2382081AF: ??? ==99053== by 0x16F: ??? ==99053== by 0x700000397C3F: ??? Thread 3: status = VgTs_WaitSys (lwpid 5635) ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x1045F4C52: _pthread_workqueue_addthreads (in /usr/lib/system/libsystem_pthread.dylib) ==99053== by 0x1042A3672: _dispatch_queue_wakeup_global_slow (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A5133: _dispatch_queue_drain (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A6112: _dispatch_mgr_queue_drain (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A5FED: _dispatch_mgr_invoke (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A5D00: _dispatch_mgr_thread (in /usr/lib/system/libdispatch.dylib) Thread 4: status = VgTs_WaitSys (lwpid 7171) ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x1045F5728: _pthread_wqthread (in /usr/lib/system/libsystem_pthread.dylib) ==99053== by 0x1045F3364: start_wqthread (in /usr/lib/system/libsystem_pthread.dylib) Note: see also the FAQ in the source distribution. It contains workarounds to several common problems. In particular, if Valgrind aborted or crashed after identifying problems in your program, there's a good chance that fixing those problems will prevent Valgrind aborting or crashing, especially if it happened in m_mallocfree.c. If that doesn't help, please report this bug to: www.valgrind.org In the bug report, send all the above text, the valgrind version, and what OS and version you are using. Thanks. i’m running my program without any valgrind option because I want to see if I can fix it… if anyone can help me I appreciate that… my application is a server that must be running all the time. If I run the application with valgrind, the program exit, if I run it from the command line as any other application, it works fine best regard ============================================= "El tamaño de tus logros depende del tamaño de tus metas." C++ and Qt Senior Developer B.S. Computer Science Buenos Aires, Argentina > On Mar 18, 2016, at 7:13 PM, André Pönitz wrote: > > On Fri, Mar 18, 2016 at 05:34:18PM -0300, Freddy Martinez Garcia wrote: >> Guys... i'm using qt-creator with non qt project and when I run the >> valgrind everything is perfect, but If i run a qt project with valgrind, my >> project doesn't start.. Is this the right place to ask about it or I have >> to go to the valgrind mail list ? > > A valgrind list would be the right place to ask if your application > doesn't work with valgrind on the command line. > > Assuming it does work there, this here (or even bugreports.qt-project.org) > is the place to ask about issues with the Qt Creator integration. > > What version of Qt Creator, valgrind and your OS are we talking about? > > Is there any output in tha Application Output pane or anything else > that looks interesting? > > Andre' -------------- next part -------------- An HTML attachment was scrubbed... URL: From freddy311082 at gmail.com Sat Mar 19 01:50:10 2016 From: freddy311082 at gmail.com (Freddy Martinez) Date: Fri, 18 Mar 2016 21:50:10 -0300 Subject: [Qt-creator] Valgrind and Qt Creator In-Reply-To: <20160318221328.GA2589@klara.mpi.htwm.de> References: <20160318221328.GA2589@klara.mpi.htwm.de> Message-ID: <453599F0-1B5B-4360-B345-AF9C16EE8F22@gmail.com> Hi Andre… this is the valgrind output freddy at Freddys-MacBook-Pro:/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib$ valgrind ./nautam-server ==99053== Memcheck, a memory error detector ==99053== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al. ==99053== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info ==99053== Command: ./nautam-server ==99053== --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommon.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsystem_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libtask_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libforms_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libformula_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libreports_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libdb_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/librules_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsecurity_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libqueries_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommunication_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libprocedure_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperspectives_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperiods_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsequence_manager.1.0.0.dylib" --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcryptopp.1.0.0.dylib" ==99053== Conditional jump or move depends on uninitialised value(s) ==99053== at 0x7FFF5FC24A87: bcmp (in /usr/lib/dyld) ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== ==99053== Use of uninitialised value of size 8 ==99053== at 0x7FFF5FC24A9F: bcmp (in /usr/lib/dyld) ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== ==99053== Use of uninitialised value of size 8 ==99053== at 0x7FFF5FC24AA4: bcmp (in /usr/lib/dyld) ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== ==99053== Conditional jump or move depends on uninitialised value(s) ==99053== at 0x7FFF5FC11977: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) ==99053== by 0x7FFF5FC0E8CD: ImageLoader::recursiveLoadLibraries(ImageLoader::LinkContext const&, bool, ImageLoader::RPathChain const&) (in /usr/lib/dyld) ==99053== --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 2 times) --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 4 times) --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 8 times) --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 16 times) valgrind: m_syswrap/syswrap-amd64-darwin.c:507 (void wqthread_hijack(Addr, Addr, Addr, Addr, Int, Addr)): Assertion 'tst->os_state.pthread - magic_delta == self' failed. host stacktrace: ==99053== at 0x238040B6C: ??? ==99053== by 0x238040F77: ??? ==99053== by 0x238040F5A: ??? ==99053== by 0x2380DFCD9: ??? sched status: running_tid=0 Thread 1: status = VgTs_WaitSys (lwpid 2567) ==99053== at 0x10449E3C2: semaphore_wait_trap (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x1042A85D5: _dispatch_group_wait_slow (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x102AE777D: CFPREFERENCES_IS_WAITING_FOR_CFPREFSD (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102AE7897: -[CFPrefsSearchListSource generationCountFromListOfSources:count:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102996CB3: -[CFPrefsSearchListSource alreadylocked_copyDictionary] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x10299198B: -[CFPrefsSearchListSource alreadylocked_copyValueForKey:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102B1964B: ___CFPreferencesCopyAppValueWithContainer_block_invoke (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102AE674F: +[CFPrefsSearchListSource withSearchListForIdentifier:container:perform:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102B195C6: _CFPreferencesCopyAppValueWithContainer (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A3E92: ___CFBundleCopyUserLanguages_block_invoke (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1042A033E: _dispatch_client_callout (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A0236: dispatch_once_f (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x102A87FAA: _CFBundleCopyLanguageSearchListInBundle (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102A62892: _copyQueryTable (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x102A62071: _copyResourceURLsFromBundle (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A22E5: _CFBundleCopyFindResources (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A1E1F: CFBundleCopyResourceURL (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A1CC2: CFBundleGetLocalInfoDictionary (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1029A1C30: CFBundleGetValueForInfoDictionaryKey (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) ==99053== by 0x1021AE88B: QCoreApplicationPrivate::macMenuBarName() (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x1021AE957: QCoreApplicationPrivate::appName() const (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x1021B047D: QCoreApplication::init() (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x1021B06BA: QCoreApplication::QCoreApplication(int&, char**, int) (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) ==99053== by 0x10003C5B8: main (in ./nautam-server) Thread 2: status = VgTs_WaitSys (lwpid 6147) ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x2380DFCD9: ??? ==99053== by 0x2382081AF: ??? ==99053== by 0x16F: ??? ==99053== by 0x700000397C3F: ??? Thread 3: status = VgTs_WaitSys (lwpid 5635) ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x1045F4C52: _pthread_workqueue_addthreads (in /usr/lib/system/libsystem_pthread.dylib) ==99053== by 0x1042A3672: _dispatch_queue_wakeup_global_slow (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A5133: _dispatch_queue_drain (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A6112: _dispatch_mgr_queue_drain (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A5FED: _dispatch_mgr_invoke (in /usr/lib/system/libdispatch.dylib) ==99053== by 0x1042A5D00: _dispatch_mgr_thread (in /usr/lib/system/libdispatch.dylib) Thread 4: status = VgTs_WaitSys (lwpid 7171) ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) ==99053== by 0x1045F5728: _pthread_wqthread (in /usr/lib/system/libsystem_pthread.dylib) ==99053== by 0x1045F3364: start_wqthread (in /usr/lib/system/libsystem_pthread.dylib) Note: see also the FAQ in the source distribution. It contains workarounds to several common problems. In particular, if Valgrind aborted or crashed after identifying problems in your program, there's a good chance that fixing those problems will prevent Valgrind aborting or crashing, especially if it happened in m_mallocfree.c. If that doesn't help, please report this bug to: www.valgrind.org In the bug report, send all the above text, the valgrind version, and what OS and version you are using. Thanks. i’m running my program without any valgrind option because I want to see if I can fix it… if anyone can help me I appreciate that… my application is a server that must be running all the time. If I run the application with valgrind, the program exit, if I run it from the command line as any other application, it works fine best regard ============================================= "El tamaño de tus logros depende del tamaño de tus metas." C++ and Qt Senior Developer B.S. Computer Science Buenos Aires, Argentina > On Mar 18, 2016, at 7:13 PM, André Pönitz wrote: > > On Fri, Mar 18, 2016 at 05:34:18PM -0300, Freddy Martinez Garcia wrote: >> Guys... i'm using qt-creator with non qt project and when I run the >> valgrind everything is perfect, but If i run a qt project with valgrind, my >> project doesn't start.. Is this the right place to ask about it or I have >> to go to the valgrind mail list ? > > A valgrind list would be the right place to ask if your application > doesn't work with valgrind on the command line. > > Assuming it does work there, this here (or even bugreports.qt-project.org) > is the place to ask about issues with the Qt Creator integration. > > What version of Qt Creator, valgrind and your OS are we talking about? > > Is there any output in tha Application Output pane or anything else > that looks interesting? > > Andre' -------------- next part -------------- An HTML attachment was scrubbed... URL: From adrozdoff at gmail.com Mon Mar 21 09:46:47 2016 From: adrozdoff at gmail.com (Alexander Drozdov) Date: Mon, 21 Mar 2016 18:46:47 +1000 Subject: [Qt-creator] Dev: Bumping Qt requirement of Qt Creator 4.0 In-Reply-To: <6ADD1178-3058-4E87-BA4C-F9FB580A43C8@theqtcompany.com> References: <6ADD1178-3058-4E87-BA4C-F9FB580A43C8@theqtcompany.com> Message-ID: Only Qt? What about clang/llvm, qbs (if bundled omited) and other? 2016-03-17 23:05 GMT+10:00 Ziller Eike : > In the light of the release of Qt 5.6, I suggest that we bump the minimum > required Qt version for Qt Creator 4.0 (and master) to Qt 5.5. > > Br, Eike > > -- > Eike Ziller, Principle Software Engineer - The Qt Company GmbH > > The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin > Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja > Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht > Charlottenburg, HRB 144331 B > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -- WBR, Alexander Drozdov http://htrd.su -------------- next part -------------- An HTML attachment was scrubbed... URL: From Eike.Ziller at theqtcompany.com Mon Mar 21 10:11:53 2016 From: Eike.Ziller at theqtcompany.com (Ziller Eike) Date: Mon, 21 Mar 2016 09:11:53 +0000 Subject: [Qt-creator] Valgrind and Qt Creator In-Reply-To: <453599F0-1B5B-4360-B345-AF9C16EE8F22@gmail.com> References: <20160318221328.GA2589@klara.mpi.htwm.de> <453599F0-1B5B-4360-B345-AF9C16EE8F22@gmail.com> Message-ID: Hi, Valgrind on OS X mostly does not work. Valgrind 3.11.0 claims to support Mac OS X 10.10, with initial support for 10.11, but we’ve already previously seen that it tends to have even more problems with Qt. Please follow the advice of the debug output below and report bugs to Valgrind. Br, Eike > On Mar 19, 2016, at 1:50 AM, Freddy Martinez wrote: > > Hi Andre… this is the valgrind output > > freddy at Freddys-MacBook-Pro:/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib$ valgrind ./nautam-server > ==99053== Memcheck, a memory error detector > ==99053== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al. > ==99053== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info > ==99053== Command: ./nautam-server > ==99053== > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommon.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsystem_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libtask_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libforms_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libformula_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libreports_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libdb_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/librules_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsecurity_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libqueries_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommunication_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libprocedure_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperspectives_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperiods_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsequence_manager.1.0.0.dylib" > --99053-- run: /usr/bin/dsymutil "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcryptopp.1.0.0.dylib" > ==99053== Conditional jump or move depends on uninitialised value(s) > ==99053== at 0x7FFF5FC24A87: bcmp (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > ==99053== > ==99053== Use of uninitialised value of size 8 > ==99053== at 0x7FFF5FC24A9F: bcmp (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > ==99053== > ==99053== Use of uninitialised value of size 8 > ==99053== at 0x7FFF5FC24AA4: bcmp (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC11974: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > ==99053== > ==99053== Conditional jump or move depends on uninitialised value(s) > ==99053== at 0x7FFF5FC11977: ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, unsigned char const*, unsigned long, long long, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC16BEA: ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned char const*, unsigned long, unsigned long long, unsigned long long, stat const&, unsigned int, unsigned int, linkedit_data_command const*, encryption_info_command const*, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC10ADE: ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char const*, unsigned long long, unsigned long long, stat const&, ImageLoader::LinkContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char const*, dyld::LoadContext const&, std::__1::vector >*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, dyld::LoadContext const&) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > ==99053== by 0x7FFF5FC0E8CD: ImageLoader::recursiveLoadLibraries(ImageLoader::LinkContext const&, bool, ImageLoader::RPathChain const&) (in /usr/lib/dyld) > ==99053== > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 2 times) > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 4 times) > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 8 times) > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 16 times) > > valgrind: m_syswrap/syswrap-amd64-darwin.c:507 (void wqthread_hijack(Addr, Addr, Addr, Addr, Int, Addr)): Assertion 'tst->os_state.pthread - magic_delta == self' failed. > > host stacktrace: > ==99053== at 0x238040B6C: ??? > ==99053== by 0x238040F77: ??? > ==99053== by 0x238040F5A: ??? > ==99053== by 0x2380DFCD9: ??? > > sched status: > running_tid=0 > > Thread 1: status = VgTs_WaitSys (lwpid 2567) > ==99053== at 0x10449E3C2: semaphore_wait_trap (in /usr/lib/system/libsystem_kernel.dylib) > ==99053== by 0x1042A85D5: _dispatch_group_wait_slow (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x102AE777D: CFPREFERENCES_IS_WAITING_FOR_CFPREFSD (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102AE7897: -[CFPrefsSearchListSource generationCountFromListOfSources:count:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102996CB3: -[CFPrefsSearchListSource alreadylocked_copyDictionary] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x10299198B: -[CFPrefsSearchListSource alreadylocked_copyValueForKey:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102B1964B: ___CFPreferencesCopyAppValueWithContainer_block_invoke (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102AE674F: +[CFPrefsSearchListSource withSearchListForIdentifier:container:perform:] (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102B195C6: _CFPreferencesCopyAppValueWithContainer (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1029A3E92: ___CFBundleCopyUserLanguages_block_invoke (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1042A033E: _dispatch_client_callout (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x1042A0236: dispatch_once_f (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x102A87FAA: _CFBundleCopyLanguageSearchListInBundle (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102A62892: _copyQueryTable (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x102A62071: _copyResourceURLsFromBundle (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1029A22E5: _CFBundleCopyFindResources (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1029A1E1F: CFBundleCopyResourceURL (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1029A1CC2: CFBundleGetLocalInfoDictionary (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1029A1C30: CFBundleGetValueForInfoDictionaryKey (in /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > ==99053== by 0x1021AE88B: QCoreApplicationPrivate::macMenuBarName() (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > ==99053== by 0x1021AE957: QCoreApplicationPrivate::appName() const (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > ==99053== by 0x1021B047D: QCoreApplication::init() (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > ==99053== by 0x1021B06BA: QCoreApplication::QCoreApplication(int&, char**, int) (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > ==99053== by 0x10003C5B8: main (in ./nautam-server) > > Thread 2: status = VgTs_WaitSys (lwpid 6147) > ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) > ==99053== by 0x2380DFCD9: ??? > ==99053== by 0x2382081AF: ??? > ==99053== by 0x16F: ??? > ==99053== by 0x700000397C3F: ??? > > Thread 3: status = VgTs_WaitSys (lwpid 5635) > ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) > ==99053== by 0x1045F4C52: _pthread_workqueue_addthreads (in /usr/lib/system/libsystem_pthread.dylib) > ==99053== by 0x1042A3672: _dispatch_queue_wakeup_global_slow (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x1042A5133: _dispatch_queue_drain (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x1042A6112: _dispatch_mgr_queue_drain (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x1042A5FED: _dispatch_mgr_invoke (in /usr/lib/system/libdispatch.dylib) > ==99053== by 0x1042A5D00: _dispatch_mgr_thread (in /usr/lib/system/libdispatch.dylib) > > Thread 4: status = VgTs_WaitSys (lwpid 7171) > ==99053== at 0x1044A46DE: __workq_kernreturn (in /usr/lib/system/libsystem_kernel.dylib) > ==99053== by 0x1045F5728: _pthread_wqthread (in /usr/lib/system/libsystem_pthread.dylib) > ==99053== by 0x1045F3364: start_wqthread (in /usr/lib/system/libsystem_pthread.dylib) > > > Note: see also the FAQ in the source distribution. > It contains workarounds to several common problems. > In particular, if Valgrind aborted or crashed after > identifying problems in your program, there's a good chance > that fixing those problems will prevent Valgrind aborting or > crashing, especially if it happened in m_mallocfree.c. > > If that doesn't help, please report this bug to: www.valgrind.org > > In the bug report, send all the above text, the valgrind > version, and what OS and version you are using. Thanks. > > i’m running my program without any valgrind option because I want to see if I can fix it… > > if anyone can help me I appreciate that… my application is a server that must be running all the time. If I run the application with valgrind, the program exit, if I run it from the command line as any other application, it works fine > > best regard > > ============================================= > "El tamaño de tus logros depende del tamaño de tus metas." > C++ and Qt Senior Developer > B.S. Computer Science > Buenos Aires, Argentina > >> On Mar 18, 2016, at 7:13 PM, André Pönitz wrote: >> >> On Fri, Mar 18, 2016 at 05:34:18PM -0300, Freddy Martinez Garcia wrote: >>> Guys... i'm using qt-creator with non qt project and when I run the >>> valgrind everything is perfect, but If i run a qt project with valgrind, my >>> project doesn't start.. Is this the right place to ask about it or I have >>> to go to the valgrind mail list ? >> >> A valgrind list would be the right place to ask if your application >> doesn't work with valgrind on the command line. >> >> Assuming it does work there, this here (or even bugreports.qt-project.org) >> is the place to ask about issues with the Qt Creator integration. >> >> What version of Qt Creator, valgrind and your OS are we talking about? >> >> Is there any output in tha Application Output pane or anything else >> that looks interesting? >> >> Andre' > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From freddy311082 at gmail.com Mon Mar 21 13:31:38 2016 From: freddy311082 at gmail.com (Freddy Martinez Garcia) Date: Mon, 21 Mar 2016 09:31:38 -0300 Subject: [Qt-creator] Valgrind and Qt Creator In-Reply-To: References: <20160318221328.GA2589@klara.mpi.htwm.de> <453599F0-1B5B-4360-B345-AF9C16EE8F22@gmail.com> Message-ID: Thanls Ziller. I will Regards El mar. 21, 2016 6:11, "Ziller Eike" escribió: > Hi, > > Valgrind on OS X mostly does not work. > Valgrind 3.11.0 claims to support Mac OS X 10.10, with initial support for > 10.11, > but we’ve already previously seen that it tends to have even more problems > with Qt. > Please follow the advice of the debug output below and report bugs to > Valgrind. > > Br, Eike > > > On Mar 19, 2016, at 1:50 AM, Freddy Martinez > wrote: > > > > Hi Andre… this is the valgrind output > > > > freddy at Freddys-MacBook-Pro:/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib$ > valgrind ./nautam-server > > ==99053== Memcheck, a memory error detector > > ==99053== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al. > > ==99053== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright > info > > ==99053== Command: ./nautam-server > > ==99053== > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommon.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsystem_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libtask_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libforms_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libformula_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libreports_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libdb_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/librules_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsecurity_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libqueries_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommunication_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libprocedure_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperspectives_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperiods_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsequence_manager.1.0.0.dylib" > > --99053-- run: /usr/bin/dsymutil > "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcryptopp.1.0.0.dylib" > > ==99053== Conditional jump or move depends on uninitialised value(s) > > ==99053== at 0x7FFF5FC24A87: bcmp (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC11974: > ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, > unsigned char const*, unsigned long, long long, ImageLoader::LinkContext > const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC16BEA: > ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned > char const*, unsigned long, unsigned long long, unsigned long long, stat > const&, unsigned int, unsigned int, linkedit_data_command const*, > encryption_info_command const*, ImageLoader::LinkContext const&) (in > /usr/lib/dyld) > > ==99053== by 0x7FFF5FC10ADE: > ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char > const*, unsigned long long, unsigned long long, stat const&, > ImageLoader::LinkContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char > const*, dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, > dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, > char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > > ==99053== > > ==99053== Use of uninitialised value of size 8 > > ==99053== at 0x7FFF5FC24A9F: bcmp (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC11974: > ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, > unsigned char const*, unsigned long, long long, ImageLoader::LinkContext > const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC16BEA: > ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned > char const*, unsigned long, unsigned long long, unsigned long long, stat > const&, unsigned int, unsigned int, linkedit_data_command const*, > encryption_info_command const*, ImageLoader::LinkContext const&) (in > /usr/lib/dyld) > > ==99053== by 0x7FFF5FC10ADE: > ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char > const*, unsigned long long, unsigned long long, stat const&, > ImageLoader::LinkContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char > const*, dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, > dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, > char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > > ==99053== > > ==99053== Use of uninitialised value of size 8 > > ==99053== at 0x7FFF5FC24AA4: bcmp (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC11974: > ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, > unsigned char const*, unsigned long, long long, ImageLoader::LinkContext > const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC16BEA: > ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned > char const*, unsigned long, unsigned long long, unsigned long long, stat > const&, unsigned int, unsigned int, linkedit_data_command const*, > encryption_info_command const*, ImageLoader::LinkContext const&) (in > /usr/lib/dyld) > > ==99053== by 0x7FFF5FC10ADE: > ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char > const*, unsigned long long, unsigned long long, stat const&, > ImageLoader::LinkContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char > const*, dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, > dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, > char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > > ==99053== > > ==99053== Conditional jump or move depends on uninitialised value(s) > > ==99053== at 0x7FFF5FC11977: > ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, > unsigned char const*, unsigned long, long long, ImageLoader::LinkContext > const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC16BEA: > ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned > char const*, unsigned long, unsigned long long, unsigned long long, stat > const&, unsigned int, unsigned int, linkedit_data_command const*, > encryption_info_command const*, ImageLoader::LinkContext const&) (in > /usr/lib/dyld) > > ==99053== by 0x7FFF5FC10ADE: > ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char > const*, unsigned long long, unsigned long long, stat const&, > ImageLoader::LinkContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char > const*, dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char > const*, dyld::LoadContext const&, std::__1::vector std::__1::allocator >*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, > dyld::LoadContext const&) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, > char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) > > ==99053== by 0x7FFF5FC0E8CD: > ImageLoader::recursiveLoadLibraries(ImageLoader::LinkContext const&, bool, > ImageLoader::RPathChain const&) (in /usr/lib/dyld) > > ==99053== > > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option > > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated > 2 times) > > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated > 4 times) > > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated > 8 times) > > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated > 16 times) > > > > valgrind: m_syswrap/syswrap-amd64-darwin.c:507 (void > wqthread_hijack(Addr, Addr, Addr, Addr, Int, Addr)): Assertion > 'tst->os_state.pthread - magic_delta == self' failed. > > > > host stacktrace: > > ==99053== at 0x238040B6C: ??? > > ==99053== by 0x238040F77: ??? > > ==99053== by 0x238040F5A: ??? > > ==99053== by 0x2380DFCD9: ??? > > > > sched status: > > running_tid=0 > > > > Thread 1: status = VgTs_WaitSys (lwpid 2567) > > ==99053== at 0x10449E3C2: semaphore_wait_trap (in > /usr/lib/system/libsystem_kernel.dylib) > > ==99053== by 0x1042A85D5: _dispatch_group_wait_slow (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x102AE777D: CFPREFERENCES_IS_WAITING_FOR_CFPREFSD (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102AE7897: -[CFPrefsSearchListSource > generationCountFromListOfSources:count:] (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102996CB3: -[CFPrefsSearchListSource > alreadylocked_copyDictionary] (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x10299198B: -[CFPrefsSearchListSource > alreadylocked_copyValueForKey:] (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102B1964B: > ___CFPreferencesCopyAppValueWithContainer_block_invoke (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102AE674F: +[CFPrefsSearchListSource > withSearchListForIdentifier:container:perform:] (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102B195C6: _CFPreferencesCopyAppValueWithContainer (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1029A3E92: ___CFBundleCopyUserLanguages_block_invoke > (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1042A033E: _dispatch_client_callout (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x1042A0236: dispatch_once_f (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x102A87FAA: _CFBundleCopyLanguageSearchListInBundle (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102A62892: _copyQueryTable (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x102A62071: _copyResourceURLsFromBundle (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1029A22E5: _CFBundleCopyFindResources (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1029A1E1F: CFBundleCopyResourceURL (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1029A1CC2: CFBundleGetLocalInfoDictionary (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1029A1C30: CFBundleGetValueForInfoDictionaryKey (in > /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) > > ==99053== by 0x1021AE88B: QCoreApplicationPrivate::macMenuBarName() > (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > > ==99053== by 0x1021AE957: QCoreApplicationPrivate::appName() const > (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > > ==99053== by 0x1021B047D: QCoreApplication::init() (in > /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > > ==99053== by 0x1021B06BA: QCoreApplication::QCoreApplication(int&, > char**, int) (in > /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) > > ==99053== by 0x10003C5B8: main (in ./nautam-server) > > > > Thread 2: status = VgTs_WaitSys (lwpid 6147) > > ==99053== at 0x1044A46DE: __workq_kernreturn (in > /usr/lib/system/libsystem_kernel.dylib) > > ==99053== by 0x2380DFCD9: ??? > > ==99053== by 0x2382081AF: ??? > > ==99053== by 0x16F: ??? > > ==99053== by 0x700000397C3F: ??? > > > > Thread 3: status = VgTs_WaitSys (lwpid 5635) > > ==99053== at 0x1044A46DE: __workq_kernreturn (in > /usr/lib/system/libsystem_kernel.dylib) > > ==99053== by 0x1045F4C52: _pthread_workqueue_addthreads (in > /usr/lib/system/libsystem_pthread.dylib) > > ==99053== by 0x1042A3672: _dispatch_queue_wakeup_global_slow (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x1042A5133: _dispatch_queue_drain (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x1042A6112: _dispatch_mgr_queue_drain (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x1042A5FED: _dispatch_mgr_invoke (in > /usr/lib/system/libdispatch.dylib) > > ==99053== by 0x1042A5D00: _dispatch_mgr_thread (in > /usr/lib/system/libdispatch.dylib) > > > > Thread 4: status = VgTs_WaitSys (lwpid 7171) > > ==99053== at 0x1044A46DE: __workq_kernreturn (in > /usr/lib/system/libsystem_kernel.dylib) > > ==99053== by 0x1045F5728: _pthread_wqthread (in > /usr/lib/system/libsystem_pthread.dylib) > > ==99053== by 0x1045F3364: start_wqthread (in > /usr/lib/system/libsystem_pthread.dylib) > > > > > > Note: see also the FAQ in the source distribution. > > It contains workarounds to several common problems. > > In particular, if Valgrind aborted or crashed after > > identifying problems in your program, there's a good chance > > that fixing those problems will prevent Valgrind aborting or > > crashing, especially if it happened in m_mallocfree.c. > > > > If that doesn't help, please report this bug to: www.valgrind.org > > > > In the bug report, send all the above text, the valgrind > > version, and what OS and version you are using. Thanks. > > > > i’m running my program without any valgrind option because I want to see > if I can fix it… > > > > if anyone can help me I appreciate that… my application is a server that > must be running all the time. If I run the application with valgrind, the > program exit, if I run it from the command line as any other application, > it works fine > > > > best regard > > > > ============================================= > > "El tamaño de tus logros depende del tamaño de tus metas." > > C++ and Qt Senior Developer > > B.S. Computer Science > > Buenos Aires, Argentina > > > >> On Mar 18, 2016, at 7:13 PM, André Pönitz wrote: > >> > >> On Fri, Mar 18, 2016 at 05:34:18PM -0300, Freddy Martinez Garcia wrote: > >>> Guys... i'm using qt-creator with non qt project and when I run the > >>> valgrind everything is perfect, but If i run a qt project with > valgrind, my > >>> project doesn't start.. Is this the right place to ask about it or I > have > >>> to go to the valgrind mail list ? > >> > >> A valgrind list would be the right place to ask if your application > >> doesn't work with valgrind on the command line. > >> > >> Assuming it does work there, this here (or even > bugreports.qt-project.org) > >> is the place to ask about issues with the Qt Creator integration. > >> > >> What version of Qt Creator, valgrind and your OS are we talking about? > >> > >> Is there any output in tha Application Output pane or anything else > >> that looks interesting? > >> > >> Andre' > > > > _______________________________________________ > > Qt-creator mailing list > > Qt-creator at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/qt-creator > > -- > Eike Ziller, Principle Software Engineer - The Qt Company GmbH > > The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin > Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja > Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht > Charlottenburg, HRB 144331 B > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From freddy311082 at gmail.com Mon Mar 21 14:53:32 2016 From: freddy311082 at gmail.com (Freddy Martinez Garcia) Date: Mon, 21 Mar 2016 10:53:32 -0300 Subject: [Qt-creator] Valgrind and Qt Creator In-Reply-To: References: <20160318221328.GA2589@klara.mpi.htwm.de> <453599F0-1B5B-4360-B345-AF9C16EE8F22@gmail.com> Message-ID: Hi Ziller... something that I forgot... valgrind doesn't work in linux nither with my qt application... Running my server application with valgrind in Ubuntu 15.10 it didn't work regards *============================================="El tamaño de tus logros depende del tamaño de tus metas." * *C++ and Qt Senior Developer* *Lic. Computer Science* *Buenos Aires, Argentina* 2016-03-21 9:31 GMT-03:00 Freddy Martinez Garcia : > Thanls Ziller. I will > > Regards > El mar. 21, 2016 6:11, "Ziller Eike" > escribió: > >> Hi, >> >> Valgrind on OS X mostly does not work. >> Valgrind 3.11.0 claims to support Mac OS X 10.10, with initial support >> for 10.11, >> but we’ve already previously seen that it tends to have even more >> problems with Qt. >> Please follow the advice of the debug output below and report bugs to >> Valgrind. >> >> Br, Eike >> >> > On Mar 19, 2016, at 1:50 AM, Freddy Martinez >> wrote: >> > >> > Hi Andre… this is the valgrind output >> > >> > freddy at Freddys-MacBook-Pro:/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib$ >> valgrind ./nautam-server >> > ==99053== Memcheck, a memory error detector >> > ==99053== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et >> al. >> > ==99053== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright >> info >> > ==99053== Command: ./nautam-server >> > ==99053== >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommon.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsystem_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libtask_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libforms_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libformula_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libreports_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libdb_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/librules_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsecurity_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libqueries_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcommunication_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libprocedure_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperspectives_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libperiods_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libsequence_manager.1.0.0.dylib" >> > --99053-- run: /usr/bin/dsymutil >> "/Volumes/DATA/nautam/nautam1.1/demo/Nautam/lib/libcryptopp.1.0.0.dylib" >> > ==99053== Conditional jump or move depends on uninitialised value(s) >> > ==99053== at 0x7FFF5FC24A87: bcmp (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC11974: >> ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, >> unsigned char const*, unsigned long, long long, ImageLoader::LinkContext >> const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC16BEA: >> ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned >> char const*, unsigned long, unsigned long long, unsigned long long, stat >> const&, unsigned int, unsigned int, linkedit_data_command const*, >> encryption_info_command const*, ImageLoader::LinkContext const&) (in >> /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC10ADE: >> ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char >> const*, unsigned long long, unsigned long long, stat const&, >> ImageLoader::LinkContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char >> const*, dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, >> dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, >> char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) >> > ==99053== >> > ==99053== Use of uninitialised value of size 8 >> > ==99053== at 0x7FFF5FC24A9F: bcmp (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC11974: >> ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, >> unsigned char const*, unsigned long, long long, ImageLoader::LinkContext >> const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC16BEA: >> ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned >> char const*, unsigned long, unsigned long long, unsigned long long, stat >> const&, unsigned int, unsigned int, linkedit_data_command const*, >> encryption_info_command const*, ImageLoader::LinkContext const&) (in >> /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC10ADE: >> ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char >> const*, unsigned long long, unsigned long long, stat const&, >> ImageLoader::LinkContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char >> const*, dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, >> dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, >> char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) >> > ==99053== >> > ==99053== Use of uninitialised value of size 8 >> > ==99053== at 0x7FFF5FC24AA4: bcmp (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC11974: >> ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, >> unsigned char const*, unsigned long, long long, ImageLoader::LinkContext >> const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC16BEA: >> ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned >> char const*, unsigned long, unsigned long long, unsigned long long, stat >> const&, unsigned int, unsigned int, linkedit_data_command const*, >> encryption_info_command const*, ImageLoader::LinkContext const&) (in >> /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC10ADE: >> ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char >> const*, unsigned long long, unsigned long long, stat const&, >> ImageLoader::LinkContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char >> const*, dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, >> dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, >> char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) >> > ==99053== >> > ==99053== Conditional jump or move depends on uninitialised value(s) >> > ==99053== at 0x7FFF5FC11977: >> ImageLoaderMachO::validateFirstPages(linkedit_data_command const*, int, >> unsigned char const*, unsigned long, long long, ImageLoader::LinkContext >> const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC16BEA: >> ImageLoaderMachOCompressed::instantiateFromFile(char const*, int, unsigned >> char const*, unsigned long, unsigned long long, unsigned long long, stat >> const&, unsigned int, unsigned int, linkedit_data_command const*, >> encryption_info_command const*, ImageLoader::LinkContext const&) (in >> /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC10ADE: >> ImageLoaderMachO::instantiateFromFile(char const*, int, unsigned char >> const*, unsigned long long, unsigned long long, stat const&, >> ImageLoader::LinkContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC038C2: dyld::loadPhase6(int, stat const&, char >> const*, dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC084CF: dyld::loadPhase5(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC081EF: dyld::loadPhase4(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC07F54: dyld::loadPhase3(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC076A9: dyld::loadPhase1(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0347A: dyld::loadPhase0(char const*, char >> const*, dyld::LoadContext const&, std::__1::vector> std::__1::allocator >*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0315E: dyld::load(char const*, >> dyld::LoadContext const&) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0876C: dyld::libraryLocator(char const*, bool, >> char const*, ImageLoader::RPathChain const*) (in /usr/lib/dyld) >> > ==99053== by 0x7FFF5FC0E8CD: >> ImageLoader::recursiveLoadLibraries(ImageLoader::LinkContext const&, bool, >> ImageLoader::RPathChain const&) (in /usr/lib/dyld) >> > ==99053== >> > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option >> > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated >> 2 times) >> > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated >> 4 times) >> > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated >> 8 times) >> > --99053-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated >> 16 times) >> > >> > valgrind: m_syswrap/syswrap-amd64-darwin.c:507 (void >> wqthread_hijack(Addr, Addr, Addr, Addr, Int, Addr)): Assertion >> 'tst->os_state.pthread - magic_delta == self' failed. >> > >> > host stacktrace: >> > ==99053== at 0x238040B6C: ??? >> > ==99053== by 0x238040F77: ??? >> > ==99053== by 0x238040F5A: ??? >> > ==99053== by 0x2380DFCD9: ??? >> > >> > sched status: >> > running_tid=0 >> > >> > Thread 1: status = VgTs_WaitSys (lwpid 2567) >> > ==99053== at 0x10449E3C2: semaphore_wait_trap (in >> /usr/lib/system/libsystem_kernel.dylib) >> > ==99053== by 0x1042A85D5: _dispatch_group_wait_slow (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x102AE777D: CFPREFERENCES_IS_WAITING_FOR_CFPREFSD (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102AE7897: -[CFPrefsSearchListSource >> generationCountFromListOfSources:count:] (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102996CB3: -[CFPrefsSearchListSource >> alreadylocked_copyDictionary] (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x10299198B: -[CFPrefsSearchListSource >> alreadylocked_copyValueForKey:] (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102B1964B: >> ___CFPreferencesCopyAppValueWithContainer_block_invoke (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102AE674F: +[CFPrefsSearchListSource >> withSearchListForIdentifier:container:perform:] (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102B195C6: _CFPreferencesCopyAppValueWithContainer >> (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1029A3E92: ___CFBundleCopyUserLanguages_block_invoke >> (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1042A033E: _dispatch_client_callout (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x1042A0236: dispatch_once_f (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x102A87FAA: _CFBundleCopyLanguageSearchListInBundle >> (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102A62892: _copyQueryTable (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x102A62071: _copyResourceURLsFromBundle (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1029A22E5: _CFBundleCopyFindResources (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1029A1E1F: CFBundleCopyResourceURL (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1029A1CC2: CFBundleGetLocalInfoDictionary (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1029A1C30: CFBundleGetValueForInfoDictionaryKey (in >> /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation) >> > ==99053== by 0x1021AE88B: QCoreApplicationPrivate::macMenuBarName() >> (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) >> > ==99053== by 0x1021AE957: QCoreApplicationPrivate::appName() const >> (in /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) >> > ==99053== by 0x1021B047D: QCoreApplication::init() (in >> /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) >> > ==99053== by 0x1021B06BA: QCoreApplication::QCoreApplication(int&, >> char**, int) (in >> /Applications/Qt/5.5/clang_64/lib/QtCore.framework/Versions/5/QtCore) >> > ==99053== by 0x10003C5B8: main (in ./nautam-server) >> > >> > Thread 2: status = VgTs_WaitSys (lwpid 6147) >> > ==99053== at 0x1044A46DE: __workq_kernreturn (in >> /usr/lib/system/libsystem_kernel.dylib) >> > ==99053== by 0x2380DFCD9: ??? >> > ==99053== by 0x2382081AF: ??? >> > ==99053== by 0x16F: ??? >> > ==99053== by 0x700000397C3F: ??? >> > >> > Thread 3: status = VgTs_WaitSys (lwpid 5635) >> > ==99053== at 0x1044A46DE: __workq_kernreturn (in >> /usr/lib/system/libsystem_kernel.dylib) >> > ==99053== by 0x1045F4C52: _pthread_workqueue_addthreads (in >> /usr/lib/system/libsystem_pthread.dylib) >> > ==99053== by 0x1042A3672: _dispatch_queue_wakeup_global_slow (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x1042A5133: _dispatch_queue_drain (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x1042A6112: _dispatch_mgr_queue_drain (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x1042A5FED: _dispatch_mgr_invoke (in >> /usr/lib/system/libdispatch.dylib) >> > ==99053== by 0x1042A5D00: _dispatch_mgr_thread (in >> /usr/lib/system/libdispatch.dylib) >> > >> > Thread 4: status = VgTs_WaitSys (lwpid 7171) >> > ==99053== at 0x1044A46DE: __workq_kernreturn (in >> /usr/lib/system/libsystem_kernel.dylib) >> > ==99053== by 0x1045F5728: _pthread_wqthread (in >> /usr/lib/system/libsystem_pthread.dylib) >> > ==99053== by 0x1045F3364: start_wqthread (in >> /usr/lib/system/libsystem_pthread.dylib) >> > >> > >> > Note: see also the FAQ in the source distribution. >> > It contains workarounds to several common problems. >> > In particular, if Valgrind aborted or crashed after >> > identifying problems in your program, there's a good chance >> > that fixing those problems will prevent Valgrind aborting or >> > crashing, especially if it happened in m_mallocfree.c. >> > >> > If that doesn't help, please report this bug to: www.valgrind.org >> > >> > In the bug report, send all the above text, the valgrind >> > version, and what OS and version you are using. Thanks. >> > >> > i’m running my program without any valgrind option because I want to >> see if I can fix it… >> > >> > if anyone can help me I appreciate that… my application is a server >> that must be running all the time. If I run the application with valgrind, >> the program exit, if I run it from the command line as any other >> application, it works fine >> > >> > best regard >> > >> > ============================================= >> > "El tamaño de tus logros depende del tamaño de tus metas." >> > C++ and Qt Senior Developer >> > B.S. Computer Science >> > Buenos Aires, Argentina >> > >> >> On Mar 18, 2016, at 7:13 PM, André Pönitz >> wrote: >> >> >> >> On Fri, Mar 18, 2016 at 05:34:18PM -0300, Freddy Martinez Garcia wrote: >> >>> Guys... i'm using qt-creator with non qt project and when I run the >> >>> valgrind everything is perfect, but If i run a qt project with >> valgrind, my >> >>> project doesn't start.. Is this the right place to ask about it or I >> have >> >>> to go to the valgrind mail list ? >> >> >> >> A valgrind list would be the right place to ask if your application >> >> doesn't work with valgrind on the command line. >> >> >> >> Assuming it does work there, this here (or even >> bugreports.qt-project.org) >> >> is the place to ask about issues with the Qt Creator integration. >> >> >> >> What version of Qt Creator, valgrind and your OS are we talking about? >> >> >> >> Is there any output in tha Application Output pane or anything else >> >> that looks interesting? >> >> >> >> Andre' >> > >> > _______________________________________________ >> > Qt-creator mailing list >> > Qt-creator at qt-project.org >> > http://lists.qt-project.org/mailman/listinfo/qt-creator >> >> -- >> Eike Ziller, Principle Software Engineer - The Qt Company GmbH >> >> The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin >> Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja >> Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht >> Charlottenburg, HRB 144331 B >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From hkarel at yandex.ru Mon Mar 21 15:33:24 2016 From: hkarel at yandex.ru (=?UTF-8?B?0JrQsNGA0LXQu9C40L0g0J/QsNCy0LXQuw==?=) Date: Mon, 21 Mar 2016 17:33:24 +0300 Subject: [Qt-creator] How run client ans server parts of the one project in debug mode In-Reply-To: <56DCAA21.40806@yandex.ru> References: <56DC65AA.9060903@yandex.ru> <3929879.VM2KY5d8d8@zmeu> <56DC75AA.2010905@yandex.ru> <56DC808E.2020603@gmail.com> <56DC82CF.1090701@gmail.com> <56DCA578.9000101@yandex.ru> <56DCAA21.40806@yandex.ru> Message-ID: <56F00634.4000003@yandex.ru> I decided to describe a solution that I've found for this situation: Create a symlink to the project. Start a new instance of QtC and open this "new" project. For the "new" project define new build directory. It's all. Now you can build and debug. The disadvantage of the solution is that the project should be built twice. For small projects - this is not a problem, but for large projects it can be uncomfortable. All the same, I would like to QtC developers have paid attention to the problem of debugging the client and server applications when they are parts in the same project. 07.03.2016 01:07, Карелин Павел пишет: > > > 07.03.2016 00:52, Orgad Shaneh пишет: >> >> >> בתאריך 6 במרץ 2016 23:47,‏ "Карелин Павел" כתב: >> > >> > >> > >> > 06.03.2016 22:19, Andrzej Telszewski пишет: >> > >> >> On 06/03/16 20:10, Andrzej Telszewski wrote: >> >>> >> >>> On 06/03/16 19:23, Карелин Павел wrote: >> >>>> >> >>>> >> >>>> >> >>>> 06.03.2016 21:20, Bogdan Vatra пишет: >> >>>>> >> >>>>> Hi, >> >>>>> >> >>>>> I think you'll need to run two QtCreator instances. >> >>>> >> >>>> I tried it, got this error: "Cannot lock build graph file >> >>>> >> '/home/hkarel/CProjects/Gamma7/build/qtc_Qt_551_a195e3b4-debug/qtc_Qt_551_a195e3b4-debug.bg': >> >> >>>> >> >>>> Already locked by 'QtCreator' (PID 4475)." >> >>> >> >>> >> >>> Is the server and client the same binary? >> >>> If so, I think you'd have to separate them and then you could run two >> >>> instances. >> > >> > Client and server are separated >> > >> >>> >> >> >> >> Or maybe another solution. Under "Debug" -> "Start Debugging" you >> have various possibilities. Maybe you could "Attach to Running >> Application" provided that you have started your application outside >> of Qt Creator? >> > >> > I tried to attach to running application, but debbuger immediately >> quits >> > >> >> >> >> I don't know how the locking is going to behave, though. >> > >> > In principle, it would suit me a running the client and server >> parts of my application in two different instances of QtCreator. But >> I can not open the same project in two different instances of QtCreator. >> >> Why do they have to be in the same project? Use 2 projects and 2 >> instances. >> >> - Orgad >> > Because in project I have a client, a server and shared libraries. I > would like to see it as a single solution in the development and > debugging. But judging from your question, I have to invent a > surrogate solution for debugging. > > -- > Pavel > > > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Wed Mar 23 16:15:58 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Wed, 23 Mar 2016 17:15:58 +0200 Subject: [Qt-creator] Do smart pointers work in Creator 4.0? Message-ID: I see that Creator 4.0 beta was released. Does someone know if either the built-in or the clang code model now supports auto-completion when dereferencing smart pointers? (std::shared_ptr and std::unique_ptr). From guenterlists at schwann.at Wed Mar 23 16:43:10 2016 From: guenterlists at schwann.at (Guenter Schwann) Date: Wed, 23 Mar 2016 16:43:10 +0100 Subject: [Qt-creator] Do smart pointers work in Creator 4.0? In-Reply-To: References: Message-ID: <5569917.hMzbs7tZOz@w541> On Wednesday, March 23, 2016 05:15:58 PM Nikos Chantziaras wrote: > I see that Creator 4.0 beta was released. Does someone know if either > the built-in or the clang code model now supports auto-completion when > dereferencing smart pointers? (std::shared_ptr and std::unique_ptr). std::shared_ptr and std::unique_ptr do work fine for me with the clang model. The model is getting quite slow with lost of templates, includes... But works pretty well in general and adds some nice features. Cheers Guenter From jan.krause.no19 at gmail.com Wed Mar 23 17:42:52 2016 From: jan.krause.no19 at gmail.com (Jan Krause) Date: Wed, 23 Mar 2016 17:42:52 +0100 Subject: [Qt-creator] QTC 3.6: parsing of qmake function shadowed(path) wrong? Message-ID: <56F2C78C.2070305@gmail.com> Hi all, since QTC 3.6 the internal parsing of qmake function shadowed(path) (@see http://doc.qt.io/qt-5/qmake-function-reference.html) seems to be wrong... especially if DESTDIR depends on the result -> the execution path of the current run-configuration of QTC is wrong... but qmake makes it right (@see messages in the example...) here a snippet of an example pro file: ------------------------------------------------ APP_PATH = $$shadowed($$PWD)/build BIN_PATH = $$APP_PATH/bin DESTDIR = $$BIN_PATH message (BIN_PATH: $$BIN_PATH) ------------------------------------------------------------ I created a bugreport (https://bugreports.qt.io/browse/QTCREATORBUG-15897) ... but maybe there are other reasons... and I'm wrong.... Any ideas? Cheers Jan -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Wed Mar 23 19:36:00 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Wed, 23 Mar 2016 20:36:00 +0200 Subject: [Qt-creator] Do smart pointers work in Creator 4.0? In-Reply-To: <5569917.hMzbs7tZOz@w541> References: <5569917.hMzbs7tZOz@w541> Message-ID: On 23/03/16 17:43, Guenter Schwann wrote: > On Wednesday, March 23, 2016 05:15:58 PM Nikos Chantziaras wrote: >> I see that Creator 4.0 beta was released. Does someone know if either >> the built-in or the clang code model now supports auto-completion when >> dereferencing smart pointers? (std::shared_ptr and std::unique_ptr). > > std::shared_ptr and std::unique_ptr do work fine for me with the clang model. > The model is getting quite slow with lost of templates, includes... But works > pretty well in general and adds some nice features. Great to hear! As I've been (slowly) migrating everything to C++11, the smart pointer issue has been increasingly becoming a major PITA. From aleixpol at kde.org Thu Mar 24 16:13:32 2016 From: aleixpol at kde.org (Aleix Pol) Date: Thu, 24 Mar 2016 16:13:32 +0100 Subject: [Qt-creator] [cmake-developers] CMake daemon for user tools In-Reply-To: References: <56923C33.6030506@gmail.com> <56C4ED48.2000604@gmail.com> Message-ID: Hey, I'll be going to Berlin the weekend of the 14th May (Board meeting, I'll be busy during the weekend). Would it make sense for me to book one day before/after to meet and hack on this? Aleix On Thu, Feb 18, 2016 at 2:20 PM, Aleix Pol wrote: > On Wed, Feb 17, 2016 at 10:59 PM, Stephen Kelly wrote: >> On 02/15/2016 07:24 PM, Tobias Hunger wrote: >> >> Hi Dominik, >> >> Am 15.02.2016 19:01 schrieb "Dominik Haumann" : >>> 1. Wouldn't it make sense you have a developer sprint ASAP for this? >> >> I'd be in, but I do not have the time to organize one. I could probably get >> a room in our office though (in Berlin). > > Eh... sorry for my silence. > > I'll be happy to join a small sprint. > > Aleix From jan.krause.no19 at gmail.com Fri Mar 25 01:58:20 2016 From: jan.krause.no19 at gmail.com (Jan Krause) Date: Fri, 25 Mar 2016 01:58:20 +0100 Subject: [Qt-creator] QTC 3.6: parsing of qmake function shadowed(path) wrong? In-Reply-To: <56F2C78C.2070305@gmail.com> References: <56F2C78C.2070305@gmail.com> Message-ID: <56F48D2C.8060609@gmail.com> this problem seems to occur if symbolic links for source directories are used... changing of QString QmakeProFileNode::sourceDir() const { return m_projectDir; } to QString QmakeProFileNode::sourceDir() const { return QDir(m_projectDir).canonicalPath(); } should resolve this... hopefully....;) cheers Jan Am 23.03.2016 um 17:42 schrieb Jan Krause: > Hi all, > > since QTC 3.6 the internal parsing of qmake function shadowed(path) > (@see http://doc.qt.io/qt-5/qmake-function-reference.html) seems to be > wrong... especially if DESTDIR depends on the result -> the execution > path of the current run-configuration of QTC is wrong... but qmake > makes it right (@see messages in the example...) > > here a snippet of an example pro file: > ------------------------------------------------ > APP_PATH = $$shadowed($$PWD)/build > > > BIN_PATH = $$APP_PATH/bin > > DESTDIR = $$BIN_PATH > message (BIN_PATH: $$BIN_PATH) > ------------------------------------------------------------ > > I created a bugreport > (https://bugreports.qt.io/browse/QTCREATORBUG-15897) ... but maybe > there are other reasons... and I'm wrong.... > > Any ideas? > > Cheers > Jan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tobias.hunger at gmail.com Fri Mar 25 09:25:24 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Fri, 25 Mar 2016 09:25:24 +0100 Subject: [Qt-creator] [cmake-developers] CMake daemon for user tools In-Reply-To: References: <56923C33.6030506@gmail.com> <56C4ED48.2000604@gmail.com> Message-ID: Am 24.03.2016 16:13 schrieb "Aleix Pol" : > Hey, > I'll be going to Berlin the weekend of the 14th May (Board meeting, > I'll be busy during the weekend). > > Would it make sense for me to book one day before/after to meet and > hack on this? Hey Aleix, I did already do some work on that front, trying to make the JSON API more complete, less prone to sudden aborts (there used to be asserts that could be triggered by user input), more consistent, and with better test coverage. See my fork on github for details. I am personally rather unsure about how to proceed. I can help make this branch production/merge ready, but I do not want to maintain it indefinitely after it is merged. It touches to many CMake internals for me to be comfortable hacking on that code. Currently you (Aleix), Domminik and me seem to be the only people that want this daemon, so I guess it would make sense for us to sit together and decide what we want, whether we can maintain it together (or where we can get help) and then hack into that direction. Best Regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Sun Mar 27 18:10:12 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 27 Mar 2016 19:10:12 +0300 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects Message-ID: With cmake projects, the "clean" or "rebuild" functions in the build menu don't do anything. I assume this is not normal. Is there a proper way to fix that? Currently, I just replaced the default command "cmake --build" with "make clean". For some reason, cmake doesn't have a "--clean" option? From cristian.adam at gmail.com Sun Mar 27 19:45:37 2016 From: cristian.adam at gmail.com (Cristian Adam) Date: Sun, 27 Mar 2016 19:45:37 +0200 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects In-Reply-To: References: Message-ID: Delete your project's .user file and reload the project. Qt Creator should then work fine. Qt Creator creates a "clean" target for all projects. "cmake --target clean" should be the manual command in case you really need it. Cheers, Cristian. On Mar 27, 2016 18:10, "Nikos Chantziaras" wrote: > With cmake projects, the "clean" or "rebuild" functions in the build menu > don't do anything. > > I assume this is not normal. Is there a proper way to fix that? Currently, > I just replaced the default command "cmake --build" with "make clean". For > some reason, cmake doesn't have a "--clean" option? > > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Sun Mar 27 19:54:58 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 27 Mar 2016 20:54:58 +0300 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects In-Reply-To: References: Message-ID: Yep, that worked. Thanks! Btw, do you know of a way to enable parallel builds globally? (For both cmake as well as qmake.) Right now, for every project, I need to edit the build step manually and add "-j4". Is there some global setting somewhere that I'm just not seeing? On 27/03/16 20:45, Cristian Adam wrote: > Delete your project's .user file and reload the project. Qt Creator > should then work fine. > > Qt Creator creates a "clean" target for all projects. "cmake --target > clean" should be the manual command in case you really need it. > > Cheers, > Cristian. > > On Mar 27, 2016 18:10, "Nikos Chantziaras" > wrote: > > With cmake projects, the "clean" or "rebuild" functions in the build > menu don't do anything. > > I assume this is not normal. Is there a proper way to fix that? > Currently, I just replaced the default command "cmake --build" with > "make clean". For some reason, cmake doesn't have a "--clean" option? From tobias.hunger at gmail.com Sun Mar 27 20:22:35 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Sun, 27 Mar 2016 20:22:35 +0200 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects In-Reply-To: References: Message-ID: Am 27.03.2016 19:55 schrieb "Nikos Chantziaras" : > > Yep, that worked. Thanks! > > Btw, do you know of a way to enable parallel builds globally? (For both cmake as well as qmake.) Right now, for every project, I need to edit the build step manually and add "-j4". Is there some global setting somewhere that I'm just not seeing? Qt Creator tries to be nice to building on the command line, so just do the same you donthere: Export the MAKEFLAGS environment variable with the flags you want each make call to have (-j4 in your case). Do that before starting Qt Creator and it effects all builds using make triggered by Qt Creator. Or have the kit set that variable if you only want it for builds with certain kits. For CMake I would recommend just switching to ninja as the build backend though. That does parallel builds out of the box and seems much faster at handling builds of projects where only a few files changed. Best Regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From realnc at gmail.com Sun Mar 27 20:31:59 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 27 Mar 2016 21:31:59 +0300 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects In-Reply-To: References: Message-ID: On 27/03/16 21:22, Tobias Hunger wrote: > For CMake I would recommend just switching to ninja as the build backend > though. That does parallel builds out of the box and seems much faster > at handling builds of projects where only a few files changed. I just assumed the default is the best one. There's so many to choose from, and I don't know what they even mean. The default is "CodeBlocks - Unix Makefiles". For "Ninja", there's several options: Ninja, CodeBlocks Ninja, CodeLite Ninja, Eclipse Ninja, etc. I have no idea whatsoever what these are. From annulen at yandex.ru Sun Mar 27 20:34:05 2016 From: annulen at yandex.ru (Konstantin Tokarev) Date: Sun, 27 Mar 2016 21:34:05 +0300 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects In-Reply-To: References: Message-ID: <295241459103645@web1g.yandex.ru> 27.03.2016, 21:32, "Nikos Chantziaras" : > On 27/03/16 21:22, Tobias Hunger wrote: >>  For CMake I would recommend just switching to ninja as the build backend >>  though. That does parallel builds out of the box and seems much faster >>  at handling builds of projects where only a few files changed. > > I just assumed the default is the best one. There's so many to choose > from, and I don't know what they even mean. > > The default is "CodeBlocks - Unix Makefiles". For "Ninja", there's > several options: Ninja, CodeBlocks Ninja, CodeLite Ninja, Eclipse Ninja, > etc. I have no idea whatsoever what these are. You need CodeBlocks Ninja. -- Regards, Konstantin From realnc at gmail.com Sun Mar 27 20:48:35 2016 From: realnc at gmail.com (Nikos Chantziaras) Date: Sun, 27 Mar 2016 21:48:35 +0300 Subject: [Qt-creator] 4.0 doesn't clean nor rebuild in cmake projects In-Reply-To: <295241459103645@web1g.yandex.ru> References: <295241459103645@web1g.yandex.ru> Message-ID: On 27/03/16 21:34, Konstantin Tokarev wrote: > 27.03.2016, 21:32, "Nikos Chantziaras" : >> On 27/03/16 21:22, Tobias Hunger wrote: >>> For CMake I would recommend just switching to ninja as the build backend >>> though. That does parallel builds out of the box and seems much faster >>> at handling builds of projects where only a few files changed. >> >> I just assumed the default is the best one. There's so many to choose >> from, and I don't know what they even mean. >> >> The default is "CodeBlocks - Unix Makefiles". For "Ninja", there's >> several options: Ninja, CodeBlocks Ninja, CodeLite Ninja, Eclipse Ninja, >> etc. I have no idea whatsoever what these are. > > You need CodeBlocks Ninja. Thanks for your help. Changing the backend to CodeBlocks Ninja seems to break all active project configurations though. Creator seems unable to update them. After I deleted *.user files (again) and reopened the projects, the configuration was working again. From nikolaus at nikolaus-demmel.de Mon Mar 28 01:49:43 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Mon, 28 Mar 2016 01:49:43 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line Message-ID: <3734C73D-F5F2-4ACD-8574-D458C3769CE3@nikolaus-demmel.de> Hi, I am cross-posting this from https://forum.qt.io/topic/65483/generating-qtcreator-project-for-cmakelists-txt-from-command-line, since it was suggested there that the topic were more appropriate for the mailing list. The following is a slightly adapted version of the original post over at forum.qt.io, including now a paragraph talking about how this might interact with QtCreator Kits. I was woundering if it is at all possible to generate a QtCreator project for a given `CMakeLists.txt` from command line, i.e. generating the `CMakeLists.txt.user` from command line while giving some parameters, such as the desired configurations and according build directories. My use case is the following. BACKGROUND: I am developing software within the ROS [1] framework. One of the many aspects of ROS is the `catkin` build system, which makes it easy for you to manage many different ROS packages and their depedencies. For example, the latest ROS distribution "Jade" has almost 1000 packages released [2]. `catkin` is CMake-based, and as such in a typical ROS workspace you might have 10s or even >100 CMake projects. Now while you can use `catkin` with the standard CMake workflow of `mkidr build && cd build && cmake ../src && make`, there also exists a command line tool suite `catkin_tools` [3], which aids you in building all catkin/CMake projects in your workspace independently, while still respecting the interdependencies. As such, it manages the `cmake` invocations and build-locations for each package. It also has support for profiles, such that you could handle different configurations at the same time (debug, relase, ...). My current workflow using this together with QtCreator is the following: I use `catkin build` to compile all my packages, which also entails calling cmake in a specific build directory for each package. Then, whenever I want to work on a specific package, I open the `CMakeLists.txt` as a project in QtCreator, but need to manually select the correct build directory. If I have different profiles, I need to select each of the build directories for the different configurations. DESIRED WORKFLOW: It would be nice if for a catkin workspace such as described above, I could generate QtCreator projects for *all* packages with the appropriate build directories set up. For example, to do the same for Eclipse, I can call something like `catkin build --force-cmake -G"Eclipse CDT4 - Unix Makefiles"`, and then import the generated projects into Eclipse, which is extremely convenient. It would be great to be able to generate the QtCreator project files in a similar fashion without user intervention. I think we really only would need to generate the `CMakeLists.txt.user` file. The `.cbp` files generated by cmake itself could be created when the user first opens the project in QtCreator and invokes cmake through QtCreator. The only parameters I currently would want to pass are the configurations and according build-directories. Note that when the user opens the generated QtCreator project, the build directory with cmake-cache etc already exists from the original `catkin build` call. As such I would not want QtCreator to inject any configuration into the cmake call such as which compiler to use, which Qt version, which version of cmake. This seems to conflict somewhat with the concept of Kits in QtCreator. What I am looking for here is using QtCreator with some sort of “current environment kit”, which should use the compiler, Qt, cmake as defined by the current environment, i.e. the same way that `cmake` and `make` (or in fact `catkin build`) calls from the command line would do. If this cannot easily be done, I think the next best thing would be allowing to select a specific Kit (by name) from command line. It would then be the responsibility of the user to setup a Kit in QtCreator first that is compatible with the environment on the command line. Does anyone have any input on what would the best way to achieve this? Should this be a standalone script/application, or somehow built into qtcreator with a specific command line invocation? Would it be easier to utilise the cmakeproject-related classes implemented in QtCreator, or generate the xml independently? Is there some kind of spec describing the contents of `CMakeLists.txt.user`? Is there some way to query information like the available Kits from QtCreator from command line in a machine readable form? Any comments or insights are welcome. Cheers & Happy Easter everyone, Niko [1] http://www.ros.org/ [2] http://repositories.ros.org/status_page/ros_jade_default.html [3] http://catkin-tools.readthedocs.org/en/latest/index.html From tobias.hunger at gmail.com Mon Mar 28 11:05:10 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Mon, 28 Mar 2016 11:05:10 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: <3734C73D-F5F2-4ACD-8574-D458C3769CE3@nikolaus-demmel.de> References: <3734C73D-F5F2-4ACD-8574-D458C3769CE3@nikolaus-demmel.de> Message-ID: Hi Nikolaus, do you have an easy to follow guide to set up some demo project for ROS? I have been trying to test ROS projects for a while now, but so far failed most of the time with setting up everything so that I could build some not-too-simple project that I can test Qt Creator and ROS together. I did fix some bugs with ROS and Qt Creator already, but so far have not had the opportunity to test the whole workflow myself. Something that works with arch Linux packages of ROS would be the most convenient for me:-) Best Regards, Tobias -------------- next part -------------- An HTML attachment was scrubbed... URL: From steveire at gmail.com Mon Mar 28 14:30:58 2016 From: steveire at gmail.com (Stephen Kelly) Date: Mon, 28 Mar 2016 14:30:58 +0200 Subject: [Qt-creator] [cmake-developers] CMake daemon for user tools In-Reply-To: References: <56923C33.6030506@gmail.com> <56C4ED48.2000604@gmail.com> Message-ID: <56F92402.3010205@gmail.com> On 03/25/2016 09:25 AM, Tobias Hunger wrote: > I am personally rather unsure about how to proceed. I can help make > this branch production/merge ready, but I do not want to maintain it > indefinitely after it is merged. It touches to many CMake internals > for me to be comfortable hacking on that code. Yes, this lack of interest in general in maintaining it upstream is currently a demotivator. I think we need a group of people interested in doing that. Brad/Ben, no response since I posted the blog or the code. Any word of interest in any of it? Thanks, Steve. From nikolaus at nikolaus-demmel.de Tue Mar 29 04:06:52 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Tue, 29 Mar 2016 04:06:52 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: Message-ID: Hi Tobias, thanks for your reply. > do you have an easy to follow guide to set up some demo project for ROS? I > have been trying to test ROS projects for a while now, but so far failed > most of the time with setting up everything so that I could build some > not-too-simple project that I can test Qt Creator and ROS together. I never used Arch Linux before, but I came up with this tutorial that should get you started with a catkin workspace of 77 ros-core packages from source: https://gist.github.com/NikolausDemmel/aeea157e00b2520aedee Note that if you just want to test catkin itself, there is actually no need to install all the ROS packages. You just would need a few of the python dependencies, and then create a workspace with catkin and some test packages. But you wanted a not-too-simple setup, so here you go :-). It is maybe worth reiterating that I am talking about the catkin_tools workflow. There is also catkin_make, which has a different approach in that it adds a top-level CMakeLists.txt to your workspace and builds all packages in one giant CMake project. This makes QtCreator integration easier, since there is just one project to import with one build directory. But the catkin-tools workflow has many advantages and is meant to eventually replace catkin_make as the default workflow: https://github.com/catkin/catkin_tools/issues/90 > I did fix some bugs with ROS and Qt Creator already, but so far have not > had the opportunity to test the whole workflow myself. Yeah I noticed things getting smoother over time. Thanks! With 3.6.1 I actually sill have an issue where `include_directories` is not properly propagated to the code-model such that QtCreator does not find include files. This is in conjunction with catkin. But I haven’t had that issue in 4.0 yet, so I stopped to investigate. Cheers, Niko P.S.: For your reference, there seem to also be people wanting a feature like this outside of ROS: http://stackoverflow.com/q/3898763/1813258 From nikolaus at nikolaus-demmel.de Tue Mar 29 04:15:53 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Tue, 29 Mar 2016 04:15:53 +0200 Subject: [Qt-creator] [cmake-developers] CMake daemon for user tools In-Reply-To: References: Message-ID: <92301709-7302-4ACC-BD2A-258DB6A92B65@nikolaus-demmel.de> Hi Stephen, > On 03/25/2016 09:25 AM, Tobias Hunger wrote: >> I am personally rather unsure about how to proceed. I can help make >> this branch production/merge ready, but I do not want to maintain it >> indefinitely after it is merged. It touches to many CMake internals >> for me to be comfortable hacking on that code. > > Yes, this lack of interest in general in maintaining it upstream is > currently a demotivator. I think we need a group of people interested in > doing that. > > Brad/Ben, no response since I posted the blog or the code. Any word of > interest in any of it? I just wanted to let you know that I think these additions have huge potential. Too much time is spent debugging cmake with `message(…)`. Unfortunately I currently do not have any capacity to join the effort, sorry. So all I can say is, keep it up! If you can get this to a useable state (hopefully with the help of others), it will be very useful indeed. Cheers, Nikolaus From tobias.hunger at gmail.com Wed Mar 30 10:52:34 2016 From: tobias.hunger at gmail.com (Tobias Hunger) Date: Wed, 30 Mar 2016 10:52:34 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: Message-ID: Hi Nikolaus, Thanks for the instructions! I just managed to do a successful catkin build (outside of Qt Creator for now:-) My first impression is that it will be tricky to fully integrate the catkin_tools workflow, even simple things like figuring out how to map source and build directories or even find the full path to files that compiler warnings are about. I will need to poke around a bit before I dare to make suggestions on how to support this though. Best Regards, Tobias On Tue, Mar 29, 2016 at 4:06 AM, Nikolaus Demmel wrote: > Hi Tobias, > > thanks for your reply. > >> do you have an easy to follow guide to set up some demo project for ROS? I >> have been trying to test ROS projects for a while now, but so far failed >> most of the time with setting up everything so that I could build some >> not-too-simple project that I can test Qt Creator and ROS together. > > I never used Arch Linux before, but I came up with this tutorial that should get you started with a catkin workspace of 77 ros-core packages from source: https://gist.github.com/NikolausDemmel/aeea157e00b2520aedee > > Note that if you just want to test catkin itself, there is actually no need to install all the ROS packages. You just would need a few of the python dependencies, and then create a workspace with catkin and some test packages. But you wanted a not-too-simple setup, so here you go :-). > > It is maybe worth reiterating that I am talking about the catkin_tools workflow. There is also catkin_make, which has a different approach in that it adds a top-level CMakeLists.txt to your workspace and builds all packages in one giant CMake project. This makes QtCreator integration easier, since there is just one project to import with one build directory. But the catkin-tools workflow has many advantages and is meant to eventually replace catkin_make as the default workflow: https://github.com/catkin/catkin_tools/issues/90 > >> I did fix some bugs with ROS and Qt Creator already, but so far have not >> had the opportunity to test the whole workflow myself. > > Yeah I noticed things getting smoother over time. Thanks! With 3.6.1 I actually sill have an issue where `include_directories` is not properly propagated to the code-model such that QtCreator does not find include files. This is in conjunction with catkin. But I haven’t had that issue in 4.0 yet, so I stopped to investigate. > > Cheers, > Niko > > P.S.: For your reference, there seem to also be people wanting a feature like this outside of ROS: http://stackoverflow.com/q/3898763/1813258 > From nikolaus at nikolaus-demmel.de Thu Mar 31 09:59:59 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Thu, 31 Mar 2016 09:59:59 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: Message-ID: Hi Tobias, thanks for trying it out. I think I made the wrong impression of what feature I am actually asking. It should be much simpler. :-) The catkin_tools workflow is currently supported quite well with qtcreator, given that catkin_tools creates more-or-less standard cmake build directories, one for each ROS package. So go ahead in your new ROS workspace after calling `catkin build`, open the CMakeLists.txt of one of the ROS packages in QtCreator as a project, then select one configuration and point the build directory to /build/. Now you can call make from within QtCreator, rerun cmake, have code completion, click on warnings to get to the source location, use the debugger, etc. That is all great. The step that is inconvenient, which I was asking about in this thread, is the fact that for each ROS package, you have to manually select the correct build directory and Kit when opening the project. Since there can be many ROS packages in a workspace (>100), it gets cumbersome. That’s why I am looking for a way to generate the project file (CMakeLists.txt.user) from command line without user interaction. I want to be able to specify the build directory(s), and also the configuration(s)/Kit(s). For the latter it would be great if there was a “environment Kit”, which selects compiler/qmake/cmake in the same way that cmake would on the command line (in fact cmake has already been run from the initial `catkin build` call, so QtCreator could just ask cmake what compiler it configured, which would maybe be a sort of “cmake Kit”, where you let cmake run in the current environment determine the compiler). Of course one could also think about integrating catkin_tools itself better into QtCreator (which is I think what you were talking about), but that is a completely different story and to me catkin_tools is so great from the command line, that I am happy continue using it there in conjunction with editing code in an IDE. Best, Nikolaus On 30.03.2016, at 10:52, Tobias Hunger wrote: > Hi Nikolaus, > > Thanks for the instructions! I just managed to do a successful catkin > build (outside of Qt Creator for now:-) > > My first impression is that it will be tricky to fully integrate the > catkin_tools workflow, even simple things like figuring out how to map > source and build directories or even find the full path to files that > compiler warnings are about. I will need to poke around a bit before I > dare to make suggestions on how to support this though. > > Best Regards, > Tobias > > On Tue, Mar 29, 2016 at 4:06 AM, Nikolaus Demmel > wrote: >> Hi Tobias, >> >> thanks for your reply. >> >>> do you have an easy to follow guide to set up some demo project for ROS? I >>> have been trying to test ROS projects for a while now, but so far failed >>> most of the time with setting up everything so that I could build some >>> not-too-simple project that I can test Qt Creator and ROS together. >> >> I never used Arch Linux before, but I came up with this tutorial that should get you started with a catkin workspace of 77 ros-core packages from source: https://gist.github.com/NikolausDemmel/aeea157e00b2520aedee >> >> Note that if you just want to test catkin itself, there is actually no need to install all the ROS packages. You just would need a few of the python dependencies, and then create a workspace with catkin and some test packages. But you wanted a not-too-simple setup, so here you go :-). >> >> It is maybe worth reiterating that I am talking about the catkin_tools workflow. There is also catkin_make, which has a different approach in that it adds a top-level CMakeLists.txt to your workspace and builds all packages in one giant CMake project. This makes QtCreator integration easier, since there is just one project to import with one build directory. But the catkin-tools workflow has many advantages and is meant to eventually replace catkin_make as the default workflow: https://github.com/catkin/catkin_tools/issues/90 >> >>> I did fix some bugs with ROS and Qt Creator already, but so far have not >>> had the opportunity to test the whole workflow myself. >> >> Yeah I noticed things getting smoother over time. Thanks! With 3.6.1 I actually sill have an issue where `include_directories` is not properly propagated to the code-model such that QtCreator does not find include files. This is in conjunction with catkin. But I haven’t had that issue in 4.0 yet, so I stopped to investigate. >> >> Cheers, >> Niko >> >> P.S.: For your reference, there seem to also be people wanting a feature like this outside of ROS: http://stackoverflow.com/q/3898763/1813258 >> From Tobias.Hunger at theqtcompany.com Thu Mar 31 12:16:10 2016 From: Tobias.Hunger at theqtcompany.com (Hunger Tobias) Date: Thu, 31 Mar 2016 10:16:10 +0000 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: , Message-ID: Hi Nikolaus, On Thu, Mar 31, 2016 at 9:59 AM, Nikolaus Demmel wrote: > Hi Tobias, > > thanks for trying it out. > > I think I made the wrong impression of what feature I am actually asking. It should be much simpler. :-) I agree, it should be simpler. > The catkin_tools workflow is currently supported quite well with qtcreator, given that catkin_tools creates more-or-less standard cmake build directories, one for each ROS package. So go ahead in your new ROS workspace after calling `catkin build`, open the CMakeLists.txt of one of the ROS packages in QtCreator as a project, then select one configuration and point the build directory to /build/. Now you can call make from within QtCreator, rerun cmake, have code completion, click on warnings to get to the source location, use the debugger, etc. That is all great. You actually should not even need to rerun cmake, provided you set up catkin to use the "-GCodeBlocks - Unix Makefiles" generator we need for Qt Creator. > The step that is inconvenient, which I was asking about in this thread, is the fact that for each ROS package, you have to manually select the correct build directory and Kit when opening the project. Since there can be many ROS packages in a workspace (>100), it gets cumbersome. That’s why I am looking for a way to generate the project file (CMakeLists.txt.user) from command line without user interaction. I want to be able to specify the build directory(s), and also the configuration(s)/Kit(s). For the latter it would be great if there was a “environment Kit”, which selects compiler/qmake/cmake in the same way that cmake would on the command line (in fact cmake has already been run from the initial `catkin build` call, so QtCreator could just ask cmake what compiler it configured, which would maybe be a sort of “cmake Kit”, where you let cmake run in the current environment determine the compiler). My original idea was to use the default build directory setting (Tools>Options>Build & Run>General IIRC) and adapt that for catkin use. Opening the source directory should then automatically pick up the right build directory. I am not sure this is an option anymore, considering that there are way more directories in $WS/build than there are in $WS/src. I need to check how that mapping from $WS/src/something/other/CMakeLists.txt to $WS/build/somedirectory works to check if that is an option. If"somedirectory" is just the CMake project name, then that should be easily doable already. Considering that such a simplistic mapping would fail in fun ways when you end up with two projects in $WS/src/* sharing the same project name (but in a different filesystem location), I do not think that is what catkin actually does:-) You would still need to set up the kits yourself then (we need the information there for the code model). For Qt Creator 4.1 I would like to have a "import existing build" functionality (as we have for qmake-based projects), which will pick a kit based on an existing build directory (and create a new kit if none matches). That requires quite a bit of refactoring of the qmake import functionality so that I can re-use (part of) the code from there. That is why this did not make the cut for Qt Creator 4.0: It was just too late to start a big refactoring when I got round to implement that. > Of course one could also think about integrating catkin_tools itself better into QtCreator (which is I think what you were talking about), but that is a completely different story and to me catkin_tools is so great from the command line, that I am happy continue using it there in conjunction with editing code in an IDE. Oh, OK:-) Best Regards, Tobias -- Tobias Hunger, Senior Software Engineer | The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From ekke at ekkes-corner.org Thu Mar 31 18:04:23 2016 From: ekke at ekkes-corner.org (ekke) Date: Thu, 31 Mar 2016 18:04:23 +0200 Subject: [Qt-creator] HowTo start Qt Linguist from External Tools with english language instead of current locale Message-ID: <56FD4A87.3060305@ekkes-corner.org> I added an entry in QtCreator External Tools to start Qt Linguist from there. works well. But QtLinguist start with german locale which is my current locale from OSX. I prefer to use IDE's in english In Qt Creator I can set the language, but I don't know HowTo change the language to en in Linguist - probably adding something to 'arguments' in external tools didn't found a hint in the docs thx helping ekke From cristian.adam at gmail.com Thu Mar 31 18:49:40 2016 From: cristian.adam at gmail.com (Cristian Adam) Date: Thu, 31 Mar 2016 18:49:40 +0200 Subject: [Qt-creator] HowTo start Qt Linguist from External Tools with english language instead of current locale In-Reply-To: <56FD4A87.3060305@ekkes-corner.org> References: <56FD4A87.3060305@ekkes-corner.org> Message-ID: Simply rename / delete linguist_de.qm from Qt5's translations directory. When the file is not found it falls back to English. Cheers, Cristian. On Thu, Mar 31, 2016 at 6:04 PM, ekke wrote: > I added an entry in QtCreator External Tools to start Qt Linguist from > there. > works well. > > But QtLinguist start with german locale which is my current locale from > OSX. > > I prefer to use IDE's in english > In Qt Creator I can set the language, but I don't know HowTo change the > language to en in Linguist - > probably adding something to 'arguments' in external tools > didn't found a hint in the docs > > thx helping > > ekke > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nikolaus at nikolaus-demmel.de Thu Mar 31 19:00:20 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Thu, 31 Mar 2016 19:00:20 +0200 (CEST) Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: , Message-ID: <1128978134.34073.961ea0e5-8785-4a9b-803a-599a405ed7c7.open-xchange@email.1und1.de> An HTML attachment was scrubbed... URL: From ekke at ekkes-corner.org Thu Mar 31 19:06:29 2016 From: ekke at ekkes-corner.org (ekke) Date: Thu, 31 Mar 2016 19:06:29 +0200 Subject: [Qt-creator] HowTo start Qt Linguist from External Tools with english language instead of current locale In-Reply-To: References: <56FD4A87.3060305@ekkes-corner.org> Message-ID: <56FD5915.70703@ekkes-corner.org> Am 31.03.16 um 18:49 schrieb Cristian Adam: > Simply rename / delete linguist_de.qm from Qt5's translations directory. > > When the file is not found it falls back to English. cool this nice trick does it :) ciao ekke > > Cheers, > Cristian. > > On Thu, Mar 31, 2016 at 6:04 PM, ekke > wrote: > > I added an entry in QtCreator External Tools to start Qt Linguist from > there. > works well. > > But QtLinguist start with german locale which is my current locale > from OSX. > > I prefer to use IDE's in english > In Qt Creator I can set the language, but I don't know HowTo > change the > language to en in Linguist - > probably adding something to 'arguments' in external tools > didn't found a hint in the docs > > thx helping > > ekke > _______________________________________________ > Qt-creator mailing list > Qt-creator at qt-project.org > http://lists.qt-project.org/mailman/listinfo/qt-creator > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nikolaus at nikolaus-demmel.de Thu Mar 31 20:23:11 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Thu, 31 Mar 2016 20:23:11 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: , Message-ID: <3FC93313-BD8B-4D57-94EA-050FED046E80@nikolaus-demmel.de> Sorry for the previous HTML contaminated email! I blame it on my hoster’s webmail client. Now I’m back at my machine. Here it comes again hopefully just plaintext: Hi Tobias, On 31.03.2016, at 12:16, Hunger Tobias wrote: > >> The catkin_tools workflow is currently supported quite well with qtcreator, given that catkin_tools creates more-or-less standard cmake build directories, one for each ROS package. So go ahead in your new ROS workspace after calling `catkin build`, open the CMakeLists.txt of one of the ROS packages in QtCreator as a project, then select one configuration and point the build directory to /build/. Now you can call make from within QtCreator, rerun cmake, have code completion, click on warnings to get to the source location, use the debugger, etc. That is all great. > > You actually should not even need to rerun cmake, provided you set up catkin to use the "-GCodeBlocks - Unix Makefiles" generator we need for Qt Creator. Is that enough? That sounds great. I thought that QtCreator might do some more "magic". This can be run easily for all packages with `catkin build --force-cmake --cmake-args "-GCodeBlocks - Unix Makefiles"`. However I noticed that QtCreator currently reruns cmake in any case whenever you open a project. I think this is fine, and that way the user doesn't even have to remember to call `catkin build` with the `"-GCodeBlocks - Unix Makefiles"` argument manually. > >> The step that is inconvenient, which I was asking about in this thread, is the fact that for each ROS package, you have to manually select the correct build directory and Kit when opening the project. Since there can be many ROS packages in a workspace (>100), it gets cumbersome. That’s why I am looking for a way to generate the project file (CMakeLists.txt.user) from command line without user interaction. I want to be able to specify the build directory(s), and also the configuration(s)/Kit(s). For the latter it would be great if there was a “environment Kit”, which selects compiler/qmake/cmake in the same way that cmake would on the command line (in fact cmake has already been run from the initial `catkin build` call, so QtCreator could just ask cmake what compiler it configured, which would maybe be a sort of “cmake Kit”, where you let cmake run in the current environment determine the compiler). > > My original idea was to use the default build directory setting (Tools>Options>Build & Run>General IIRC) and adapt that for catkin use. Opening the source directory should then automatically pick up the right build directory. Your idea is to determine if the opened CMakeLists.txt file corresponds to a catkin package, and if yes, use some special logic to determine the correct build directory? This could be the perfect solution, I like it. > I am not sure this is an option anymore, considering that there are way more directories in $WS/build than there are in $WS/src. I need to check how that mapping from $WS/src/something/other/CMakeLists.txt to $WS/build/somedirectory works to check if that is an option. If"somedirectory" is just the CMake project name, then that should be easily doable already. Considering that such a simplistic mapping would fail in fun ways when you end up with two projects in $WS/src/* sharing the same project name (but in a different filesystem location), I do not think that is what catkin actually does:-) It might actually be even easier than you think. Bear with me :-). First, let me explain shortly what is the layout. In the src directory, there might be arbitrarily many catkin packages, at an arbitrary depth in the file tree and are identified by the presence of a `package.xml` manifest file. This file also specifies the name of the catkin package, i.e. the folder name containing the catkin package in the source space is not relevant. One reason why allowing packages ot be nested in the folder subtree of the src directory is important is because often we want to create one git repository with multiple catkin packages in subfolders, and then clone that repo alongside other repositories into the src of our workspace. Now in the build directory, all the top-level folders correspond to the catkin package names, so the folder layout can be much different. Having multiple packages of the same name in a catkin workspace is an error, so we don't have to worry about that. And yes, the cmake project name is also supposed to match the catkin package name. But maybe you don't have to worry about any of that :-). catkin_tools comes with some built-in commands to query information about the workspace and about packages, meant exactly for tool-support. For example, to determine if the current directory is inside a catkin workspace you can run catkin locate and it will either print the workspace-root or give you an error. Then for example to find the build directory of a package foo_pkg, you just run catkin locate foo_pkg --build This even works together with profiles (which are is catkin's support for multiple configurations in the same workspace), so you can query the build directory of a package for each of the profiles (or to keep it simply initially, just for the currently active profile). This is only a rough sketch, but I think using catkin itself to find the correct folders could be a robust solution. I can certainly help with figuring out the details and edge cases, and also extending `catkin locate` to provide all the informaton we need. > > You would still need to set up the kits yourself then (we need the information there for the code model). This would actually be a big step already even if you need to manually select the Kit. > > For Qt Creator 4.1 I would like to have a "import existing build" functionality (as we have for qmake-based projects), which will pick a kit based on an existing build directory (and create a new kit if none matches). That requires quite a bit of refactoring of the qmake import functionality so that I can re-use (part of) the code from there. That is why this did not make the cut for Qt Creator 4.0: It was just too late to start a big refactoring when I got round to implement that. This sounds perfect. I think this could be the last piece of the puzzle to achieve what I had originally intended. :) How would this work with configurations (Release / Debug / ...). At least with the Makefile backend, a given build directory that you import would correspond to exactly one configuration, namely the one that CMAKE_BUILD_TYPE is set to. Would you then simply only offer this one configuration for that project, or offer to create additional build directories for the other configurations? > >> Of course one could also think about integrating catkin_tools itself better into QtCreator (which is I think what you were talking about), but that is a completely different story and to me catkin_tools is so great from the command line, that I am happy continue using it there in conjunction with editing code in an IDE. > > Oh, OK:-) The solution above is shaping up nicely IMHO, so maybe we can talk about a deeper integration of catkin_tools sooner than expected. E.g. offering to automatically set the build-step to `catkin build --this` for cmake projects that are detected to be a catkin package inside a workspace should be straight forward once the above is implemented. Thanks for your ideas! You got me all excited now :-). Kind regards, Nikolaus From Tobias.Hunger at theqtcompany.com Thu Mar 31 21:27:02 2016 From: Tobias.Hunger at theqtcompany.com (Hunger Tobias) Date: Thu, 31 Mar 2016 19:27:02 +0000 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: <3FC93313-BD8B-4D57-94EA-050FED046E80@nikolaus-demmel.de> References: , , <3FC93313-BD8B-4D57-94EA-050FED046E80@nikolaus-demmel.de> Message-ID: Hi Nikolaus, On Thu, Mar 31, 2016 at 8:23 PM, Nikolaus Demmel wrote: >> You actually should not even need to rerun cmake, provided you set up catkin to use the "-GCodeBlocks - Unix Makefiles" generator we need for Qt Creator. > > Is that enough? That sounds great. I thought that QtCreator might do some more "magic". We try very hard to not do any magic, especially not with the build systems! Ideally you should be able to open any project using one of the supported build systems without any changes and without any special configuration in Qt Creator. That is one of the reasons why I sometimes joke that "Qt Creator is the best IDE for everybody that hates IDEs":-) > This can be run easily for all packages with `catkin build --force-cmake --cmake-args "-GCodeBlocks - Unix Makefiles"`. In fact I did catkin config -DCMAKE_BUILD_TYPE=Release "-GCodeBlocks - Unix Makefiles" You need to experiment a bit to get the spaces quoted (who puts spaces into something that is supposed to be used as a command line argument?!). > However I noticed that QtCreator currently reruns cmake in any case whenever you open a project. I think this is fine, and that way the user doesn't even have to remember to call `catkin build` with the `"-GCodeBlocks - Unix Makefiles"` argument manually. I do not know catkin well enough to comment whether or not that is viable. Qt Creator will rerun cmake though, since it needs to make sure there is no stale data in the configuration it reads. Who knows what the user did to the project since cmake was last run? >>> The step that is inconvenient, which I was asking about in this thread, is the fact that for each ROS package, you have to manually select the correct build directory and Kit when opening the project. Since there can be many ROS packages in a workspace (>100), it gets cumbersome. That’s why I am looking for a way to generate the project file (CMakeLists.txt.user) from command line without user interaction. I want to be able to specify the build directory(s), and also the configuration(s)/Kit(s). For the latter it would be great if there was a “environment Kit”, which selects compiler/qmake/cmake in the same way that cmake would on the command line (in fact cmake has already been run from the initial `catkin build` call, so QtCreator could just ask cmake what compiler it configured, which would maybe be a sort of “cmake Kit”, where you let cmake run in the current environment determine the compiler). >> >> My original idea was to use the default build directory setting (Tools>Options>Build & Run>General IIRC) and adapt that for catkin use. Opening the source directory should then automatically pick up the right build directory. > > Your idea is to determine if the opened CMakeLists.txt file corresponds to a catkin package, and if yes, use some special logic to determine the correct build directory? This could be the perfect solution, I like it. Even simpler: My idea was to just configure the default build directory for all projects to point to wherever catkin will put its build directories. At this point I am mostly thinking about how you can configure Qt Creator to make things more convenient for you right now (or soon, since I assume you will be using Qt Creator 4.0 for the better cmake experience;-). > This is only a rough sketch, but I think using catkin itself to find the correct folders could be a robust solution. I can certainly help with figuring out the details and edge cases, and also extending `catkin locate` to provide all the informaton we need. Sure, something like that would be best. But that of course requires somebody that knows those tools well to actually implement this:-) >> For Qt Creator 4.1 I would like to have a "import existing build" functionality > > This sounds perfect. I think this could be the last piece of the puzzle to achieve what I had originally intended. :) > > How would this work with configurations (Release / Debug / ...). At least with the Makefile backend, a given build directory that you import would correspond to exactly one configuration, namely the one that CMAKE_BUILD_TYPE is set to. Would you then simply only offer this one configuration for that project, or offer to create additional build directories for the other configurations? I want to generalize the code and reuse as much as possible from the qmake implementation. There we basically use the default build directory as configured by the user for the kits, go up one level and scan all the folders found in that directory. If a directory is found, then all the checkboxes for build configurations are unchecked, the existing directories are shown under the kit they use (with kits being set up for builds that do not match any existing kit) and all these directories are checked. So all the configurations should be found (provided you use a somewhat sane directory layout) and clicking on "Configure" should get you rolling with all of them set up. That is not perfect for your use-case, but it should seriously reduce the necessary number of user interactions to open one cmake project that is a part of a bigger catkin project. > The solution above is shaping up nicely IMHO, so maybe we can talk about a deeper integration of catkin_tools sooner than expected. E.g. offering to automatically set the build-step to `catkin build --this` for cmake projects that are detected to be a catkin package inside a workspace should be straight forward once the above is implemented. I will keep your use-case in the back of my mind as I continue to work on Creator's cmake support for 4.1. > Thanks for your ideas! You got me all excited now :-). I am looking forward to discuss ideas as they evolve, and will await your patches;-) Best Regards, Tobias -- Tobias Hunger, Senior Software Engineer | The Qt Company The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B From nikolaus at nikolaus-demmel.de Thu Mar 31 23:41:26 2016 From: nikolaus at nikolaus-demmel.de (Nikolaus Demmel) Date: Thu, 31 Mar 2016 23:41:26 +0200 Subject: [Qt-creator] Generating QtCreator cmake project for given CMakeLists.txt from command line In-Reply-To: References: , , <3FC93313-BD8B-4D57-94EA-050FED046E80@nikolaus-demmel.de> Message-ID: Hi Tobias, On 31.03.2016, at 21:27, Hunger Tobias wrote: > On Thu, Mar 31, 2016 at 8:23 PM, Nikolaus Demmel wrote: >>> You actually should not even need to rerun cmake, provided you set up catkin to use the "-GCodeBlocks - Unix Makefiles" generator we need for Qt Creator. >> >> Is that enough? That sounds great. I thought that QtCreator might do some more "magic". > > We try very hard to not do any magic, especially not with the build systems! Ideally you should be able to open any project using one of the supported build systems without any changes and without any special configuration in Qt Creator. > > That is one of the reasons why I sometimes joke that "Qt Creator is the best IDE for everybody that hates IDEs":-) Yeah I realize this more and more, and it is great. (Maybe my biggest criticism in that regard would be that QtCreator does nothing from command line, like e.g. the project generation that I was initially talking about). This is actually to some extend also true for catkin. It tries very hard to make catkin packages behave like a standard cmake package. That’s the reason it already works great with QtCreator or any other IDE that supports plain Cmake. > >> This can be run easily for all packages with `catkin build --force-cmake --cmake-args "-GCodeBlocks - Unix Makefiles"`. > > In fact I did > catkin config -DCMAKE_BUILD_TYPE=Release "-GCodeBlocks - Unix Makefiles" Yup, that works as well to have it permanently configured in your catkin workspace. But since QtCreator runs it as well every time, you can save a few cycles and not run the additional generator when calling cmake through `catkin build`. It shouldn’t hurt though. > >> However I noticed that QtCreator currently reruns cmake in any case whenever you open a project. I think this is fine, and that way the user doesn't even have to remember to call `catkin build` with the `"-GCodeBlocks - Unix Makefiles"` argument manually. > > I do not know catkin well enough to comment whether or not that is viable. I think this is what people have done so far when using QtCreator with catkin. Since I think few people realize that QtCreator simply calls the CodeBlocks generator under the hood, they would not have run it manually. I think we can just leave everything as is for this point. > > Qt Creator will rerun cmake though, since it needs to make sure there is no stale data in the configuration it reads. Who knows what the user did to the project since cmake was last run? True, but most of the time cmake itself can tell if it needs to rerun or not, i.e. calling `make` will check if it should first re-run cmake, since e.g. one modified the CMakeLists.txt. But in any case, I think QtCreator does fine here, no need to change IMHO. > >>>> The step that is inconvenient, which I was asking about in this thread, is the fact that for each ROS package, you have to manually select the correct build directory and Kit when opening the project. Since there can be many ROS packages in a workspace (>100), it gets cumbersome. That’s why I am looking for a way to generate the project file (CMakeLists.txt.user) from command line without user interaction. I want to be able to specify the build directory(s), and also the configuration(s)/Kit(s). For the latter it would be great if there was a “environment Kit”, which selects compiler/qmake/cmake in the same way that cmake would on the command line (in fact cmake has already been run from the initial `catkin build` call, so QtCreator could just ask cmake what compiler it configured, which would maybe be a sort of “cmake Kit”, where you let cmake run in the current environment determine the compiler). >>> >>> My original idea was to use the default build directory setting (Tools>Options>Build & Run>General IIRC) and adapt that for catkin use. Opening the source directory should then automatically pick up the right build directory. >> >> Your idea is to determine if the opened CMakeLists.txt file corresponds to a catkin package, and if yes, use some special logic to determine the correct build directory? This could be the perfect solution, I like it. > > Even simpler: My idea was to just configure the default build directory for all projects to point to wherever catkin will put its build directories. At this point I am mostly thinking about how you can configure Qt Creator to make things more convenient for you right now (or soon, since I assume you will be using Qt Creator 4.0 for the better cmake experience;-). Aha, I didn’t get that. That should work (as long as I only open catkin packages, not other cmake packages as well). Thanks for pointing that out again, I didn’t realize this was a setting. It is somewhat limited though, since a relative path will not work with all packages (as I tried to explain they can be nested arbitrarily deep inside the src folder, i.e. different packages might need different number of “../“ in the relative path. Using an absolute path only works for a single workspace and a single profile. Still I think this could be beneficial to many people, since I assume that working with multiple workspaces and multiple profiles is the exception, not the rule. Ah, and yes, I am using 4.0 beta ;-). > > > >> This is only a rough sketch, but I think using catkin itself to find the correct folders could be a robust solution. I can certainly help with figuring out the details and edge cases, and also extending `catkin locate` to provide all the informaton we need. > > Sure, something like that would be best. But that of course requires somebody that knows those tools well to actually implement this:-) I would feel confident to figure things out on the catkin side of things. I would need some pointers to where additions in QtCreator should be made, though. Where would I for example add the conditional code that checks if a cmake project that is about to be opened should be considered a catkin package and then adapt the build directory according to some output of a subprocess call to `catkin locate`? Is there a good example of how to best do subprocess calls in QtCreator (I guess I could look for the calls to `cmake`)? > > >>> For Qt Creator 4.1 I would like to have a "import existing build" functionality >> >> This sounds perfect. I think this could be the last piece of the puzzle to achieve what I had originally intended. :) >> >> How would this work with configurations (Release / Debug / ...). At least with the Makefile backend, a given build directory that you import would correspond to exactly one configuration, namely the one that CMAKE_BUILD_TYPE is set to. Would you then simply only offer this one configuration for that project, or offer to create additional build directories for the other configurations? > > I want to generalize the code and reuse as much as possible from the qmake implementation. There we basically use the default build directory as configured by the user for the kits, go up one level and scan all the folders found in that directory. If a directory is found, then all the checkboxes for build configurations are unchecked, the existing directories are shown under the kit they use (with kits being set up for builds that do not match any existing kit) and all these directories are checked. > > So all the configurations should be found (provided you use a somewhat sane directory layout) and clicking on "Configure" should get you rolling with all of them set up. > > That is not perfect for your use-case, but it should seriously reduce the necessary number of user interactions to open one cmake project that is a part of a bigger catkin project. With the catkin layout the build directories of multiple configuration of a single catkin package will not be siblings in the directory tree. But once you have this working as you describe it for the standard single-package cmake workflow, maybe I can investigate addition conditional logic depending on if a project is detected to be a catkin package or not. I will wait to hear more about your progress on this. > >> The solution above is shaping up nicely IMHO, so maybe we can talk about a deeper integration of catkin_tools sooner than expected. E.g. offering to automatically set the build-step to `catkin build --this` for cmake projects that are detected to be a catkin package inside a workspace should be straight forward once the above is implemented. > > I will keep your use-case in the back of my mind as I continue to work on Creator's cmake support for 4.1. > >> Thanks for your ideas! You got me all excited now :-). > > I am looking forward to discuss ideas as they evolve, and will await your patches;-) I hope I can get in there with some guidance on where/how to add stuff — as time allows. I have started by downloading and compiling the source. It’s a start ;-). The first thing I would like to have a look at is the conditional default-build-directory that I describe above. Specific questions requesting guidance are inline in the discussion. Kind regards, Nikolaus