I have a project in XCode that uses some C++ code (openCV to be exact). One of my files, AdjustVC.mm importing this file (it's called image_processing.mm). Everything works fine. But when I import image_processing.mm in another file, TestVC.mm, I get errors:
ld: 14 duplicate symbols for architecture armv7
It's not a problem about your code base, ld is the tool for the linking phase, most likely you are using your libraries in the wrong way.
you should add more details about the libraries involved in your project or just verify them.
Related
I am looking for a way to suppress/satisfy the linker when a symbol is undefined at compile time.
To illustrate what I am trying to accomplish I will put up a simple example:
CoreCpp.h file:
#include <stdio.h>
int calculate();
CoreCpp.cpp file:
#include "CoreCPP.h"
int calculate(){
return 0;
}
With these two files above I will create a .a static library.
Then in Xcode I will create a simple command line project as you can see in the image below:
If I link the static library, the project will compile without any issue of course:
However, if I don't link the static library, it will throw an undefined symbol error.
Undefined symbols for architecture x86_64:
"calculate()", referenced from:
_main in main.o
ld: symbol(s) not found for architecture x86_64
Question: Is there a way that I can suppress the error or somehow make the linker happy at compile time? Can we apply the concept of weak linking here?
I'm not familiar with Xcode, but its compilation/linking process should be similar to Eclipse. So first you should understand the differences between compilation, linking, and building. Then you should also realize that there are 2 ways of building your project: either compile your source code files separately and then link them together to get machine-executable instructions, OR to compile and link in one single step/command. In the step of compilation only, you can either get an output of a static or dynamic library, or a .o object file, which will be used later in the linking step. With separate steps of compile/link you have more control over your build process. What Eclipse (and supposedly Xcode) is doing when you build your project, is either of those 2 ways, depending on your project settings. If you'd like to take a full control over the process of building in Eclipse or whatever IDE you have, you should take care of your project settings, or even better to set it up to work with a Makefile, where you can define exactly how your source code should be compiled and linked and in which steps/order. In Eclipse you need to create a Makefile-project and edit your Makefile alongside the project files.
In this your question above, what you were trying to do (or asking for help) is forcing the IDE to build your project without specifying the library, for which obviously your linker was complaining about. Instead, if you specify the library it doesn't complain any more. But be careful, you should make sure your IDE settings are fine to (re)compile your library from its source upon changes, otherwise you may end-up linking an old version of the compiled library file with your main.cpp and other files.
I've tried literally everything on the planet and am about to pull my hair out...
I'm trying to write and build a static library in C++, and then include and call it from an app in Objective C. Simple enough. I get the following linker error when trying to run the app:
Undefined symbols for architecture armv7:
"spmEstimatorMobile::spmCalculate(double*, double*, double*)", referenced from:
-[RecordViewController outputAccelerationData:] in RecordViewController.o
ld: symbol(s) not found for architecture armv7
clang: error: linker command failed with exit code 1 (use -v to see invocation)
I've tried EVERYTHING. The app calling file has the .mm extension, I've made sure the compilers that are set for both the library and the app are the same, I've ensured that I build the library for the same architecture (armv7) that I'm building the app for, I've included the library by adding it in the build phase settings page, I'm compiling all sources when building the library as well, I've tried running lipo etc. to make a universal binary just in case ... I just can't figure this out...
If it helps, this is on XCode 6.2., I'm trying to deploy to my iPhone 5 running iOS 8.1. I've double checked all the deployment targets too.
Any ideas??? Anyone!
EDIT: I'm using STL stuff in the C++ static lib (i.e. vectors, etc.) .. not sure if that makes a difference...
I'm running on a Mac Mini, late 2012 (The latest), 64 bit, Intel Core i7 2.3 gHz HyperThreading x86_64, or so says my computer. I'm also using Code::blocks for the time being because help in Xcode (which I would prefer) is even more sparse because I am using C++. That said, programming language of choice is C++, and will eventually expand into Objective-C++, but that's for another day.
The problem is this: I am trying to compile a program that will do I'm not quite sure what yet (it's a template, let's just say it's a game for now.) using SDL 1.2.15, which I have compiled from source on my Mac. (I did try using SDLmain.m and SDL.h, but that was full of Objective-C code, and as I could find no good information on whether I could code in C++ with these on the internet, I gave up.) The source I compiled seems to work just fine, Code::Blocks can compile it's tutorial just fine and it runs ok. However, I cannot seem to get my game template to work. It compiles fine, with no errors or warnings, but the linker does not do the same. Now, I'm really not sure what ld does, I just know that it links together code. (Anyone want to elaborate on that?) When Code::Blocks gets to it's linking step it fails with this message:
-------------- Clean: Debug in PyroLauncher ---------------
Cleaned "PyroLauncher - Debug"
-------------- Build: Debug in PyroLauncher ---------------
Compiling: App.cpp
Compiling: OnCleanup.cpp
Compiling: OnEvent.cpp
Compiling: OnInit.cpp
Compiling: OnLoop.cpp
Compiling: OnRender.cpp
Linking console executable: bin/Debug/PyroLauncher
Undefined symbols for architecture x86_64:
"_SDL_GetError", referenced from:
App::OnInit() in OnInit.o
"_SDL_Init", referenced from:
App::OnInit() in OnInit.o
"_SDL_PollEvent", referenced from:
App::OnExecute() in App.o
"_SDL_Quit", referenced from:
App::OnCleanup() in OnCleanup.o
"_SDL_SetVideoMode", referenced from:
App::OnInit() in OnInit.o
"_main", referenced from:
start in crt1.10.6.o
(maybe you meant: _SDL_main)
ld: symbol(s) not found for architecture x86_64
collect2: ld returned 1 exit status
Process terminated with status 1 (0 minutes, 0 seconds)
0 errors, 0 warnings
Which has me seriously confused. I just know that ld can't find for my native architecture x86_64. If anyone can help me, it would be greatly appreciated.
What I think the solutions may be: (I don't know how to do them, however.)
SDL is not compiled for 64 bit, and therefore cannot be compiled against with 64 bit code.
SDL is compiled for Objective-C, and therefore cannot be used with code in C++.
My code is 32 bit, SDL is 64 bit, and they are not compatible.
Do you understand what is happening here? Your code is calling functions which the linker (ld) can't find.
The two main probabilities are that you either need to include some more files in your project, or that you need to define some macros.
Search through the project for the definition of SDL_GetError() (and the others). If you don't find it, then you need to find the file which declares it and include that.
If you do find it, then it will probably be surrounded by #ifdef XXXX and you will have to define xxxx in your project options (not in your code). Most good IDEs (I use NetBeans, with Cygwin for C+ development) will grey out the then contents of an ifdef if it isn't defined, so you can imemdiately see if that is the case.
Btw, have a look around the SDL website for installation instructions. Alternatively, your download might have a file called "Install" or "Install.txt", and may also have a Makefile. If it does, then in your IDE look for an option Project/New ... from existing makefile.
Hope this helps.
You need to add the files SDLMain.m and SDLMain.h to your project. The SDLMain files contain glue code so SDL code can run on Mac OS X. You also need to link the Cocoa framework to your project. The Mac version of SDL is written in Cocoa so you need to link to Cocoa. Linking the Cocoa framework does not force you to use Objective-C. You can write your game in C++.
If you're looking for information on setting up SDL with Xcode 4, read the following article:
Using SDL with Xcode 4
this question has gone back and forth a bit as I have learnt some things about g++ about unix systems (sorry if I messed anyone about).
For a project I am currently trying to finish I would like to get twitcurl running with Xcode and OpenFrameworks. If anyone has managed to do this please share with me this arcane feat.
Below is some of the things I am stuck on:
Regurgitating some of the instructions from their wiki it says to
download the libraries source
build twitcurl library using visual C++ or make (their is also cmake in their).
Include twitcurl.h and curl headers in your twitter application and link to twitcurl.lib and libcurl.lib/libcurl.dll.
This seems simple enough but when I make the library, there is no twitcurl.lib. A different file libtwit.a is produced, which I assume must be what OSX needs (as installing puts it in usr/local/ along with the headers). However then if I try to include the twitcurl.h in a header and make a twitCurl object I then get more undefined symbols
Undefined symbols for architecture i386:
"twitCurl::~twitCurl()", referenced from:
testApp::setup() in testApp.o
"twitCurl::twitCurl()", referenced from:
testApp::setup() in testApp.o
ld: symbol(s) not found for architecture i386
I now am assuming twitcurl makes for a 64bit (I have tried adding CFLAGS=-m32, but it fails), and to my knowledge Openframeworks is only 32 bit currently. This means twitcurl may not be an option for me, but again, if anyone knows how please let me know!
Thanks in advance.
You'd probably do well to recompile twitcurl, last I looked (3 months ago), it was compiled with VC6, which has its own way of mangling C++ symbols, so the lib won't link correctly for g++.
you could then just import it directly as well and then not need to worry about linking to the libs, however it would be a good idea to test in a small demo app, before integrating into an already build system, that way you know if the error is on your side or somewhere else.
Also, when it comes to this type of problem, the exact compiler errors are more helpful than just describing the problem.
The error message you've posted is the linker's longwinded way of telling you that its looking for a definition of the curl_easy_setopt function, which is part of libcurl. Adding -lcurl to your g++ line should fix this (by telling the linker to link in libcurl).
However, twitcurl comes with a Makefile, which already does this. In addition, if you are trying to build a shared library there, you're doing it wrong...
Maybe you can try ofxTwitter, an openFrameworks addon for twitter, to get your twitter functionality. get it here: http://forum.openframeworks.cc/index.php/topic,2750.0.html
I used CMake and Visual C++ to build the HyDE library. Then, still in VC++, I was able to successfully create code and build an executable that links into HyDE.lib and the HyDE header files.
I then discovered that in order to work with others at my company, it would be preferable to develop in Eclipse CDT. Knowing very little about Eclipse CDT, I created a default hello world project, deleted the code and then dumped in all of my code into the src folder. Then I attempted to change the includes and lib path and libs to mirror what had worked in VC++. At this point everything seems to compile, but I get an error in linking:
/cygdrive/c/EclipseWorkspace/425HyDE/Debug/../src/FS5HyDE.cpp:16: undefined reference to `HyDEAPI::HyDE::HyDE(HyDESystemModel::SystemModel*, bool)'
(There are many more errors like this, all referring to HyDE methods.) Here is what is being run at the command line:
g++ -L"C:\Progra~1\boost\boost_1_42\lib" -L"C:\EclipseWorkspace\HyDE" -o"425HyDE.exe" ./src/Adapter_FS5HyDE.o ./src/EPSCommands.o ./src/EPSCurrentSensor.o ./src/EPSFault.o ./src/FS5HyDE.o ./src/HyDEObservation.o ./src/MCDH.o ./src/MCDH_Module.o ./src/PDBComponent.o ./src/PowerSystem.o ./src/Program.o ./src/SSPCComponent.o ./src/Telemetry.o ./src/TelemetryReport.o -l:libboost_thread-vc90-mt-gd-1_42.lib -lHyDE
This is definitely not a library ordering problem because I've the other ordering as well (there are only two). Is it possible that there is a problem with compiling HyDE.lib in VC++ (which uses a Windows compiler) and compiling my program with g++? Could there be a problem in the way that Eclipse CDT is autogen'ing the makefiles? Any other ideas?
(Note: there appear to be plenty of others questions on SO with similar problems, but after reading through them I have yet to find one that addresses my problem.)
Classic missing symbol error. Which source file defines:
HyDEAPI::HyDE::HyDE(HyDESystemModel::SystemModel*, bool)' ?
Was this file added to the compilation? Can you spot it on the command line you pasted?
If this symbol belongs to an external library, after adding the directory path with -L , you could add the name of the specific library you want to link with your program using -l.
I'm going to suggest that you try to add to the compilation command the directory path to HyDE.lib, followed immediately by the library name, like this:
-L"C:\path_to_hyde_library" -l:HyDE.lib
and then tell us what happened.
Solution: Since the HyDE library was compiled with the Visual Studios compiler and I'm attempting to build the code that links to it with the Cygwin toolchain the two compilers use different name mangling schemes so that the latter linker can not find the expected symbols in the HyDE library. The only solution that I've found is to recompile the HyDE library with the Cygwin toolchain or compile the new code with whatever compiler Visual Studios is using. (grumble grumble)
./src/FS5HyDE.o and ./src/HyDEObservation.o should be the latest parameter if other object files (*.o files) need them, it means that the most needed object files should be appeared as last as possible in the parameters list.