after setting up KDevelop under Kubuntu 19.10 with CMake, the following error occurs whenever I open a project:
Failed to configure the project (error message: Configuration failed.).
As a result, KDevelop's code understanding will likely be broken.
To fix this issue, please ensure that the project's CMakeLists.txt files are correct, and KDevelop is configured to use the correct CMake version and settings. Then right-click the project item in the projects tool view and click 'Reload'.*
If I ignore the message, I can compile my project without any issues, the CMakeLists.txt is set up correctly. I have no idea where to solve this error, I haven't found anything for this specific case on google.
EDIT: Additional information
KDevelop Version: 5.4.2
CMake Version: 3.13.4
Upon creating a new KDevelop project, I create "new from template", choose the KDevelop Category and the standard "Simple KDevelop Plugin" as project type. After creation, it instantly prompts me with said error.
In the picture below, you can see the CMake configuration for the newly added project.
CMake project config
Make sure you have:
base-devel
extra-cmake-modules
To install:
apt install extra-cmake-modules
If that doesn't work, you might need to add some repos. Check this discussion
Explanation:
I had a similar problem and got:
Failed to configure project 'me_first_proj' (error message: Exit code 1). As a result, KDevelop's code understanding will likely be broken.
Inside CMakeLists.txt I found it looking for $ECM_MODULE_PATH which turned out to be empty.
This, was related to extra-cmake-modules which I didn't have.
For Arch: pacman -Syu extra-cmake-modules
The name might differ for your distro.
I already had base-devel but required ECM
Related
I need to build Gstreamer1.0(1.14.2) from its source files using meson & ninja. First, I cloned the repository, added a custom plugin to support OpenMAX. Then, in the gst-build executed the,
meson build --buildtype=debug -Dcustom_subprojects=gst-omx -Ddisable_gst_libav=true .
This doesn't generate any error messages and ninja works like a charm as well. However, for all plugins I get
Message: Could not create symlink to <plugin_name>
message, which is not good because I cannot reach them using terminal. My host device is ARM64 and linaro-aarch, and running a custom operating system. I appreciate any kind of comment & help to solve. Thanks in advance.
I have trouble using toolchains in Docker from CLION. My Docker container is based on ubuntu 18.04 and CMake 3.17 is installed. The image definition is based on this reference implementation.
When I reload the CMake project with CLion, the following error is shown.
-- Build files have been written to: /tmp/tmp.vmmD6opEtD/cmake-build-debug-local-docker
CMake File API: /path/to/glow/cmake-build-debug-local-docker: no reply dir found
When I configure the CMake project in the Docker container directly, it succeeded. I believe there is nothing wrong with the CMake project itself.
What does "no reply dir found" mean? There is no clue for this error message even with googling.
I have this error sometimes when CLion is using rsync to sync files. Disabling rsync gets rid of this error:
The problem is that the file sync is slower, but slow is better than non-functional :)
I had the exact same error, in my case setting up SSH to avoid asking for passphrase, that is on a terminal run ssh-add then try to reload the CMake project
Just had the same problem.
Your build dir in clion cmake project settings is other than the one appear in the cmake options specified there.
So I'm using WebStorm and trying to put a project on GitHub, but every time I select "Share Project on GitHub" from the VCS menu, I get the following error message:
Errors while executing git --version. exitCode=1 errors: xcrun: error: invalid active developer path (/Library/Developer/CommandLineTools), missing xcrun at: /Library/Developer/CommandLineTools/usr/bin/xcrun
Now, at the time of writing this I'm not even positive that I linked my WebStorm to a GitHub account. Would this message be related to that if I don't? If so, where would I go to check if I linked it?
Please open try running
xcode-select --install
In your OS terminal - this will download and install xcode developer tools and fix the problem. The problem likely is that one needs to explicitly agree to the license agreement.
see Command Line Tools not working - OS X El Capitan/macOS Sierra/macOS High Sierra:
I have never used qt before, but I got a task to build an already existing application for Windows and Mac using the newest version of qt. What I have done so far is:
Installed Visual Studio Professional 2017.
Installed the latest Qt 5 Open Source.
Note: when installing the QT 5 I installed all of the components under QT/ QT 5.10.1 which were: qt 5.10.1 components and under QT/Tools I installed QT Creator 4.6.0 CDB Debugger Support and MinGW 5.3.0 : qt 5 Tools
I cloned the branch from GitHub that I wanted to build to C:/Projects
Then in Git Bash I cd to the project and tried to run qmake. The first time I had this problem:
"bash: qmake: command not found"
After searching into the problem I found out that I haven't added "Qt/5.10.1/msvc2017_64/bin" to PATH and I added it. Now when I echo $PATH I can see the directory added.
Now after bash recognises qmake and I try to run it it says:
Project ERROR: Cannot run target compiler 'cl'. Output:
===================
Maybe you forgot to setup the environment?
I have no idea how to fix the problem due to the fact that I have never used QT to build anything before. Can you help me?
In the read me file for the application I am trying to build it says:
Step-by-step instructions
1. Clone this branch
2. cd into the project
3. Run qmake - this wil generate Makefiles for all of the project's modules
4.Run make (Use the -j4 of -j flags - this greatly speeds up the build process - see the manual page for make for more info). You can also use the -s flag to silence the output - this also wins you a few seconds.
I am a bit puzzled by the instructions, it does not say to run qmake -project anywhere, but there is a qt project file in the directory that I cd to. Does this mean that I don't have to run qmake -project?
Also I have no idea how to run make by using the -j4 and -j flags and how to run flags in general. Could someone help me?
I believe that for the Cannot Run Compiler 'cl' issue I have some kind of a problem with the set up. Did I install too many components or is it to do with the VS setup?
I highly recommend you to install Qt from the Qt installer, and to use Qt Creator.
However, if you want to use it from source, you first need to get a compiler (ex: MinGW). Then you need to execute the instructions, but you need to use a MinGW (or whatever compiler you use) command prompt, which will automatically set the environment variables needed.
Note that this is Windows specific. On mac and linux, the compiler's environment variables are already set in the default command prompt (if there is a compiler, obviously).
I'm currently struggling to build OpenCV for the UWP.
I already googled quite a lot and found Microsoft's OpenCV Github Repo which tecnically should do the magic. However, this repo and pretty much everything else I found in this regard are outdated (Visual Studio 2015, old OpenCV versions etc.). I need to use OpenCV 3.3 because it's a cross platform project and I don't want to recompile everything else solely because of an outdated git repo.
Can anyone explain the process of building OpenCV using CMake from the official repo for the UWP?
My first attempt was to simply use the Windows Dlls, however my application than shouts "Failed to load module" at me. Then, I tried to build OpenCV the Visual Studio Project files as stated here and here. However, this doesn't do it for me as it throws the following error:
CMake Error at cmake/OpenCVUtils.cmake:440 (if):
if given arguments:
"(" "OR" "OFF" ")"
Unknown arguments specified
Call Stack (most recent call first):
CMakeLists.txt:317 (OCV_OPTION)
It also tells me to check the CMakeOutput.log but there it's 0 errors everywhere.
If someone could tell me either how to get rid of this error or what to change in the CMake GUI to build OpenCV for UWP, that'd be great!
Edit:
I also tried to compile OpenCV with a platform specific toolchain for WinRT as found in platforms/winrt. It doesn't really change anything though (I'm not even sure if it should), I still get "Unable to load Dll: The specified module could not be found". Maybe GPPK is right in his assumption that it's more an UWP problem than an OpenCV one. Anyways, if anyone knows anything about this, I'd really love to get some help!
Ok, I didn't get an answer yet, so I managed to do it myself (more or less).
Here's how I did it:
Prerequisites
Visual Studio (2017)
UWP SDK
CMake
OpenCV from official repo
Steps
Create a build folder which you want the UWP libraries to built in.
Start PowerShell in this folder and execute the following command (exchange C:\OpenCV with your local OpenCV path): cmake -G "Visual Studio 15 2017" -DCMAKE_SYSTEM_NAME=WindowsStore -DCMAKE_SYSTEM_VERS
ION="10.0" -DCMAKE_VS_EFFECTIVE_PLATFORMS=x86 -DBUILD_TESTS=OFF -DBUILD_PERF_TESTS=OFF C:\OpenCV
If you get the same error message as I did (see question for more details), go to the root CMakeLists.txt in the OpenCV folder and add a # in front of line 317 (OCV_OPTION(ENABLE_PYLINT)). This should fix it, to check, execute the command from above once more.
Ensure that the output claims "Windows RT support YES" and "Building for Microsoft platform: Windows".
Open OpenCV.sln in the build folder. Ensure that all projects are either of type "Windows Store" or "Universal Windows" and build the ALL_BUILD Project twice: Once in Release configuration, and once in Debug. Then build the INSTALL Project.
That's it, you should find the built Dlls in the install folder. Step 3 is probably a bug in OpenCV and will hopefully be fixed in upcoming versions.