When I develop a project that output is a shared library or executable, normally, I will use include_directories in the CMakeLists.txt of root directory. In such case, before I include a header within a translate unit, I can't use angle brackets to let headers include each other. I think it is because files in include_directories is not a part of project. I can fix this by using relative path, but when I have to import external project, this can not fixed easily.
Assume that I have a external project with a root CMakeLists.txt, which has this snippet
add_library(mylib INTERFACE)
target_include_directories(mylib INTERFACE .)
and the directory seems like this
└──external
├───mylib.hpp
└───CMakeLists.txt
I want to import this project, and I'm writing a header-only project. The file tree seems like this
├──external (above)
├──include
| ├──header.hpp
| └──CMakeLists.txt
└──CMakeLists.txt
The CMakeLists.txt in root directory is simply add_subdirectory(include), and here is the CMakeLists.txt in folder include
add_library(Header INTERFACE)
target_include_directories(Header INTERFACE .)
target_link_libraries(Header INTERFACE mylib)
Theoretically, I can now include mylib.hpp by #include <mylib.hpp>, but it is not the case. VSCode tells me it can't find this header file. As much as I know, the only way to solve this is add a translate unit and let the implementation code include the header.hpp. I have tested this several time. VSCode won't report this error as long as I add a cpp file and include that header, and once I remove the cpp file, the error appears again.
By the way, I have set "C_Cpp.default.configurationProvider": "ms-vscode.cmake-tools" in configuration file of vscode.
Please tell me how to fix this without a translate unit.
Related
Goal
I'd like to launch a CLI program closely related to my flutter project, i.e., the program is saved somewhere near the flutter project folder.
My end goal here, is so that I could release a separate problem outside of the flutter app bundle at a fixed location relative to the bundle, e.g., same parent folder, while flutter-built exe can still find the program automatically. The solution targets Windows/macOS.
Expectation
I expect that I could retrieve a standard project path, such as the path to main.dart, and go from there using relative paths. I was so spoiled by Python's __file__ and wish to see something similar. This is also fairly easy to do with Windows/macOS native API, like this
For example, say I created a project under this folder
/path/to/my/flutter_project
I expect to call a Dart API to get the path of main.dart like this
/path/to/my/flutter_project/lib/main.dart
Observation
According to this answer The closest thing I got with flutter, is
import 'dart:io' as io;
Uri filePath = io.Platform.script.resolve('.');
however, puts me to a prescribed location:
// macOS
~/Library/Containers/com.example.flutterRelpath/Data/
This is the package data folder instead of the project folder.
If I query the script itself using
io.Platform.script.path
I get
~/Library/Containers/com.example.flutterRelpath/Data/main.dart
which is not the physical location of the script.
Question
Does it mean that I would need an installer to install the CLI program there or prepare a UI for the user to specify the location before I could use it? This seems a lot of trouble.
There is no reason to obtain the path of the script, nor does that make sense to do in a compiled application as the source files are not directly used at runtime.
You can simply use a relative path to reference whatever file/executable you want.
final uri = Uri.file('relative/file/path');
This will give you a Uri to the path file in the file folder in the relative folder, which would be at the same level as your executable.
├── executable.exe
├── relative
│ └── file
│ └── path//The Uri will refer to HERE
In order for this to be a relative, the passed path must not start with a path separator. So it should not be:
final uri = Uri.file('/relative/file/path');
Have a look at the dcli package and the DartScript and DartProject classes. they have methods designed to provide exactly this type of path information.
E.g.
DartProject.self.pathToProjectRoot
I'm having more trouble than I'd like to admit to structure a simple project in Python to develop using Visual Studio Code.
How should I structure in my file system a project that is a simple Python package with a few modules? Just a bunch of *.py files together. My requisites are:
I must be able to step debug it in vscode.
It has a bunch of unit tests using pytest.
I can select to debug a specific test from vscode tab and it must stop in breakpoints.
pylint must not show any false positives.
The test files must be in a different directory of the main module files.
I must be able to run all the tests from the console.
The module is executed inside a virtual environment using python standard lib module venv
The code will use type hints
I may use another linter, even another test framework.
Nothing fancy, but I'm really having trouble to get it right. I want to know:
How should I organize my subdirectory: a folder with the main files and a sibling folder with the tests? Or a subfolder with the code and a subsubfolder with the tests?
Which dirs must have a init.py file?
How the tests should import the files from the module? Should I use relative imports?
Should I create a pytest.ini file?
Should I create a .env file?
What's the content of my launch.json the debugger file config in vscode?
Common dir structure:
app
__init__.py
yourappcode.py
tests (pytest looks for this)
__init__.py
test_yourunittests.py
server.py if you have one
.env
.coveragerc
README.md
Pipfile
.gitignore
pyproject.toml if you want
.vscode (helpful)
launch.json
settings.json
Or you could do one better. Ignore my structure and look at the some of famous python projects github page. Like fastAPI, Flask, asgi, aiohttp are some that I can think of right now
Also:
I think absolute imports are easier to work with compared to relative imports, I could be wrong though
vscode is able to use pytest. Make sure you have a testing extension. Vscode has a built in one im pretty sure. You can configure it to pytest and specify your test dir. You can also run your test from command line. If youre at the root, just running ‘pytest’ will recognise your tests dir if it’s named that by default. Also your actual test files need to start with prefix test_ i think.
The launch.json doesn’t need to be anything special. When you click on the settings button next to play button in the debug panel. Vscode will ask what kind of app is it. I.e If its a flask app, select python then select flask and it will auto generate a settings file which you can tweak however you want in order to get your app to run. I.e maybe you want to expose a different port or the commands to run your app are different
It sounds to me like you just need to spend a bit of time configuring vscode to your specific python needs. For example, you can use a virtualenv and linting in whichever way you want. You just need to have a settings.json file in the .vscode folder in your repo where you specify your settings. Configurations to specify python virtualenv and linting methods can be found online
I am working with veins. I needed to change the default behavior of BaseWaveApplLayer. Rather doing in file changes, I created a second application file with the name MyApp. I added MyApp.cc, MyApp.h, MyApp.ned files. Then I duplicated TraCIDemoRSU11p, and added files for TraCIDemoRSU11pEnhanced. Now in .ini file when I changed *.rsu[*].applType to point to "TraCIDemoRSU11pEnhanced". It compiles fine, but when I try to run it, generates the following error:
Submodule appl: no module type named `TraCIDemo11pEnhanced' found that implements module interface org.car2x.veins.base.modules.IBaseApplLayer (not in the loaded NED files?).
In MyApp.ned file it is declared to be of type IBaseApplLayer. I tried setting ned-path parameter in ini file, didn't resolve the problem. In veins project properties, NED source folder is checked. Do I have to add the ned path for the newly created application file, somewhere?
Try using the cookiecutter package (https://veins.car2x.org/tutorial/#newprojects) to create new applications that use veins instead of copying and pasting files:
Use pip to install the package (https://github.com/cookiecutter/cookiecutter)
Then follow the steps in (https://github.com/veins/cookiecutter-veins-project/blob/master/README.md) to create the new project; following the prompts that appear.
This will create a new project that uses Veins and has the application layer-related files for you to update according to your application with no errors.
I imported one project to another project by following this link tutorial
successfully build that two project. But when I added
#import <SampleSubproject/SampleSubproject.h> this to my main ViewController.m file error be occur.
Error : "SampleSubproject/SampleSubproject.h" file is not found.
Why is that error occurred ??
Is there any error ?
Please help me guys!
You have Headers search path settings at project configuration. Let's say there're directories A and B. With "SampleSubproject/SampleSubproject.h" Xcode looks for SampleSubproject.h at A/SampleSubproject/SampleSubproject.h and B/SampleSubproject/SampleSubproject.h. If there's a checked recursive flag for any directories, the search is performed at this directories and their subdirectories.
Now you'll need to check what is the headers search path for the target you are building keeping in mind that you might have several targets and the file with the import might not be included into all the targets.
As you also have a red library, you'll want to check the Libraries search path as well, it works in the same way.
I want to run my working pydev project python code by double clicking the main module (outside of eclipse): xxx.py
The problem is that due to my imports being in different packages:
from src.apackage.amodule import obj
when xxx.py is double clicked it complains it doesn't know where the imports are (even though when I run xxx.py in pydev it magically knows what I'm importing).
A simple workaround is to remove all of the packages and move all of the modules into one directory (that obviously works but is very inconvenient)
How can I run my code in the file system without doing that work around?
This page answers my question excellently:
http://blog.habnab.it/blog/2013/07/21/python-packages-and-you/
Bottom line is always execute your code from the top, highest level, root directory (e.g. using a minimal main.py file that executes the main script of your program). Then, use absolute imports always and you never have a missing module issue since you start the program from the top directory and all imports are based off that 'home' path.
The problem you encountered is the natural behavior of most languages. A programm only knows about its working path (the path it is started in), the paths which are registered in the environment variables and at least relative paths.
The "magic" of the executable you created is therefore: It collects all scripts/modules needed, and copies/combines them next to/in the executable. The Executable then runs within the directory where all other scripts also reside and voila ...
If you are not happy with your workaround of creating an executable every time you want to run your project without PyDev there are two alternatives.
First but not the one I would suggest is registering the working path into in the environment variables.
Second and the one I think is much better: Create a link to the python executable and alter the calling string of the textfield "Target:". Append the path to your script you would like to run. Then alter the textfield "Start in:" and enter the project directory. After you did this you should be able to start your project with a simple double click.
(If you rely on external libraries which are neither on the path nor in you project you could search for appending paths temporarily to the pythonpath via the sys module.)
I hope I could help a bit.