We have built a large-ish program using QT Creator 5 on Fedora (easy install). We now have to compile for older distro's including RH5 and RH6.
I realize that QT Creator doesn't support the older versions of libstdc++ which ship with RH5 and RH6. However, can we still use make & qmake from the command line to compile our project under other distros?
Qt Creator gives you the commands.
Open Qt Creator and your project (on Fedora). Then go to the "Projects" tab (the one with a folder icon at the left of your window). The window should show you the compiling steps it uses for building your program (and the cleaning steps too). Most of the time. Copy those steps in a shell script, adapt it to your targeted plateform (RHEL in your case) and then compile the program by launching your script on the targeted plateform (i.e. RHEL). Otherwise you can using the traditional qmake && make on Red Hat.
Related
I have a strange error with qmake: To compile octomap, first I need to compile octovis code, accordingly with documentation I need to use:
cd octovis/src/extern/QGLViewer
qmake
mingw32-make
But when I use qmake, they just do nothing, seeing the task monitor the amount of memory used is constant and the amount of processor used is 0.
I tried to uninstall/reinstall QT and MinGW, and didn't work, I have change the version of QT (4.8.5, 4.8.6, 5.5) and MinGW and didn't work, I'm using windows 7 over 64 bits. I appreciate any solution.
I was able to generate the qmake files with no problem. make failed, but my environment is QT 5.5.1, which I doubt is compatible. I can tell you how my environment is set up, and that should get you going.
Go ahead and download QT Creator for whatever version of QT you require. This includes the appropriate mingw version, so they don't need to be downloaded separately. http://download.qt.io/archive/qt/
Find out what version of QT is recommended. It looks like 4.X, but I'm not 100 percent certain.
Add X:/QTCREATOR/QTVERSION/mingw/bin to your system path, with QTCREATOR being the root directory for QT Creator and QTVERSION being the version. These are included with the QT Creator install.
Also add X:/QTCREATOR/Tools/mingw/bin to your system path.
I'm assuming you have CMake already. That's obviously required.
This should have your build environment set up properly. You're sure to encounter other problems along the way.
The best option appears to just run it under Linux. There are pre-compiled binaries.
I am using Qt Creator to deploy my Qt application. On Mac, I'd like to include the required Qt libraries in the .app bundle. Is there any way to do it automatically using Qt Creator? Should I do it using the command-line? In that case, how should I do it?
The macdeployqt command line tool will add all the necessary Qt libraries that your Qt project references.
If you require any other, 3rd party libraries, you'll need to copy these manually and set the paths to them using the install_name_tool command.
You can check which libraries your application references using the otool command. For example: -
otool -L MyApplication.app/Contents/MacOS/MyApplication
For Qt Creator, I tend to write a script that adds the necessary libraries and calls macdeployqt and then under Projects, add a build step which calls the script.
An example script that would just add the Qt libraries would look something like this: -
#!/bin/bash
pwd
echo Copying QT libraries...
macdeployqt ./MyApplication.app
You can simply run macdeployqt foo.app. Qt Creator does not support this feature off-hand either. However, you can inject custom commands into your process in the QtCreator project settings.
It does not support QML just yet though. There are patches under codereview where it is coming. See the following link for details:
https://codereview.qt-project.org/#q,status:open+project:qt/qttools,n,z
Note: macdeployqt should not be used for usual development and debug! It should be only used when deploying. Otherwise, it is executed each time for building even if you just recompile the code due to a minor change for testing. This can slow down that process, but as for deploying, it should be alright.
On QT6 I was able to do it entirely within QT Creator:
In Projects/Build, add a custom build step after 'make' (probably only want to do this for your 'release' configuration):
Command: %{Qt:QT_HOST_PREFIX}/bin/macdeployqt
Arguments: %{ActiveProject:BuildConfig:Path}/%{ActiveProject:Name}.app - qmldir=%{ActiveProject:NativePath}
Working Directory: %{buildDir}
I was able to test it by airdropping the resulting .app onto my test machine.
reference: https://doc.qt.io/qt-6/macos-deployment.html#macdeploy
On ubuntu 12.10 I've been using Qt Creator for a while. After updating to ubuntu 13.04 I've encountered the problem, that I can no longer see my src-files in the project editor. I only see the .pro and .pri file.
I've checked that the build path is correct and re-installed QT 5. As well, I can compile and run the project as usual, but the src-files just won't appear in the editor. I used to run the project with QT4 (qmake-qt4 makefile.pro), might this interfere?
I found a workaround by adding the required source and header files manually via right-click on the project. But this is rather inconvenient as the source files are spread across several folders. I still couldn't figure out why QT creator does not add them automatically after I've upgraded to ubuntu 13.04 and QT 5 (downgrade to QT 2.7 was in vain, too).
I am doing C development using Netbeans on OS X and my project fails to build, stating "...this installation of Open MPI was not compiled with Fortran 90 support"
I have installed a newer gcc and Open MPI (along side the default versions), and I can build using them via make on a command line. This leads me to believe that Netbeans is using the default Open MPI installation (which did not have fortran support). If I am correct, how do I get it to use the new installation? I told Netbeans about the other compilers via the Tool Collection Manager (File->Project Properties->Build->Tool Collection->[...]). However, I do not know of a way to tell it about Open MPI.
I have a working solution. This solution exists in two parts.
1) I reran configure on the command line for my project and specified full paths for MPICC and MPIFC. This solved the problem of getting Netbeans to use the right mpicc compiler. However, it created another issue: the mpif90 wrapper could not find gfortran.
2) I altered the 'GUI environment' PATH variable to put gfortran in my path using the /etc/launchd.conf method found here (http://stackoverflow.com/questions/135688/setting-environment-variables-in-os-x).
After a reboot, Netbeans compiles my project. So, I'm claiming success.
I'm having some trouble installing and configuring qt on my vista laptop.
I'm trying to setup a development environment on my laptop where I compile from the command line, because that's how the environment is setup on my university's linux machines, so I don't want to tie myself to some IDE .. (plus, real programmers use the command line!)
I haven't used the command line before for C++ development, it was all MSVC, so now I'm having a bit of trouble.
I'm still using MSVC, but from the command line. I practically have no idea what's going on, I just know that I have to run:
qmake
nmake
to compile my code!
I downloaded the opensource version of qt, and did the configuration, and tried a simple qt application (from a tutorial) and it worked, it compiled and executed pretty much as expected.
Now, when I decided to run another project that uses opengl, I got the following error:
fatal error C1083: Cannot open include file: 'qgl.h': No such file
or directory
I'm not sure where does the compiler look for header files, and I didn't copy any header files anywhere, I assume that configure.exe worked its magic somehow and added the include directory to one or more enviroment variables or to some registery location or whatever other peculier places that the MSVC compiler searches for to find include directories.
However, what I did was search my C:\qt\include\ folder to make sure that qgl.h exists, and sure enough there it was. so why can't nmake find it?
I think the actual solution to this is in your pro file:
QT += opengl
If you want to stay with the command line anyway (plus use it on a linux box later / parallel) I'd suggest at least trying out the MinGW version of Qt. I'm using it regularly, and besides of the non-existance of a GUI it works pretty well. Using MinGW also has the advantage that you can simply download and install the MinGW edition of Qt and don't need to reconfigure or recompile anything.
Also, trying out QtCreator might be interesting. It's still beta and requires the beta Qt 4.5 but it's a nice small IDE that integrates nicely with gcc.
Two potential solutions (they solved issues at my workplace)
Do you have qt include and bin folders in the PATH variable? I think the doc says only one of these is needed, but one of the students had Vista and putting the other in the PATH variable solved a "Cannot open include file" problem.
If you're using MSVC did you run configure and nmake from the Visual Studio command prompt? We had problems when using the bare windows Command Prompt because the VS one adds a lot of temporary environment variables to the configure process.
Good luck
Install the complete Qt SDK for Windows which includes Qt 4.6 SDK, Qt Creator 1.3, and MinGW.
It will also install "Qt Command Prompt" launcher that you can use to build Qt apps from the command line.
I'm sure you're more familiar with MSVC than MinGW, as I do too (I've been using MSVC 6.0 to MSVC# 2008 for developing .NET apps).
But try MinGW with Qt and I think it's better for long term. I do some C++ development on Linux too so getting familiar with MinGW will be beneficial for you in cross-platform C++/Qt development.
For more info, see Installation of Qt 4.6 SDK for Windows.
Qmake generates Makefile from *.pro file located in current directory. It has qt path compiled in. Type "qmake -v" to see it. You can't move qt's dir after compiling it. If You haven't moved it, first maybe try to install Qt following instruction from INSTALL file. Good luck.
The opensource version of Qt does not provide profiles (mkspecs in qt terms) so qmake can generate nmake (msvc) compatible makefiles.
You have to use mingw/gcc.