Sorry about the short title, but I honestly can't get a better description of what is happening because I don't know enough...
Some background first, I am "converting" a multi-byte application to support unicode and I've made the standard char/string wchar_t/wstring changes and the my code is building without problems.
What happens is that when the application is being initialized it hits an assert when it registers the applications's document templates. The code is the standard
CMultiDocTemplate* pRepDocTemplate = NULL;
pRepDocTemplate = new CMultiDocTemplate(IDR_DIAGNOSTIC_REPORT_TYPE,
RUNTIME_CLASS(CDiagnosticReportDoc),
RUNTIME_CLASS(CChildFrame), // custom MDI child frame
RUNTIME_CLASS(CDiagnosticReportView));
and CDiagnosticsReportView has the standard DECLARE_DYNCREATE and IMPLEMENT_DYNCREATE in the header and source.
The assert is at doctmpl.cpp line 29 (mfc120ud.dll - at least is using the correct dll), but I can't find the source code anywhere to actually know what is happening.
The inheritance tree is pretty straightforward:
CDiagnosticReportView
\->CReportViewBase
\->CXTPReportView
\->CView
CXTPReportView is part of a framework that we are using which is provided by Codejock (Codejock extreme toolkitPro). From the build pane I know that it's linking against it's unicode debug dll (ToolkitPro1631vc120UD.dll)
Suffice to say that in the multibyte configuration this problem doesn't occur.
The project is configured to use the UNICODE character set (Project properties->Configuration Properties->General->Character Set).
Any help would be appreciated!
Thanks in advance!
It all had to do with the fact that I was linking against the non UNICODE build of Codejock. Even though I saw a reference to the UNICODE dll, the lib wasn't the correct one!
Problem was solved when I opened my eyes ;)
Related
Hello there back in 1999 early 2000 a lot of my apps used this MFC Export dialog extension which added filters to file extensions, because MFC didn't yet have that (at least that's what I remember to be the reason, maybe I was drunk I don't now), the library can be downloaded here: https://www.codeproject.com/Articles/54/Adding-filters-to-the-Open-File-dialog
Now I know that this functionality is built in already, but its easier to fix one library then to fix all of my apps
Now if I compile this with VC++ 6 everything is fine, the problem is that I would like to use a bit more modern compiler
Here are my changes that I added to this lib over the years (its a big file so I will post a link instead): https://pastebin.com/pj0j2nL4
The example code that I use to test this library can be found here: https://www.mediafire.com/file/xz5mxwy2rcmb9h3/FileExportDialog_demo_upg.zip/file
Just click on File->Export and you should see the problem
Now the problem happens on the library's void CFileExportDialog::OnTypeChange() method
is this part
CWnd *fileNameBox = GetParent()->GetDlgItem(edt1);
if (fileNameBox == NULL)
fileNameBox = GetParent()->GetDlgItem(cmb13);
For some reason, GetParent() returns NULL (I cannot even get to GetDlgItem() at this point)) pointer, I have no idea why does it do that, and the strangest thing is that this used to work on VC++ 6 (it also worked on MFC7 as well), why doesn't it work on VC++ 17 (MFC140) anymore ?
Hopefully someone has an idea what changed and how can I fix my library
Thanks for Anwsering and Best Regards
I'm trying to write a plugin for 3ds max, I went through the entire sdk installation process to the letter as described in the help files.
The problem I'm facing though is intellisence complaining about an invalid macro definition
"IntelliSense: command-line error: invalid macro definition:_CRT_SECURE_CPP_OVERLOAD_STANDARD_NAMES_COUNT =1"
I found the definition in project settigs -> c/c++ -> preprocessor definitions as inherited from parent or project default.
I tried disabling the inherited definitions and re-entered them, this time without the space between the name and the = and all works fine so I'm guessing its a typo on their part?
Anyway, I want to change the default project or whatever to not repeat it every time i start a new project. The project is created with a wizard which required me to copy over some files to appear and after which I had to enter the sdk path.
The files I copied are plain text with some fancy extensions and not much in them so I'm guessing the defaults are described in the sdk directory.. somewhere. Does anybody know what kind of a file I'm looking for?
EDIT: I found a file called root.vcxproj_template and it has a section for preprocessor definitions but all it contains is
<PreprocessorDefinitions>_USRDLL;%(PreprocessorDefinitions)</PreprocessorDefinitions>
and no mention of the broken one
EDIT2: in another part of the file there was a path to a property sheet (maxsdk\ProjectSettings\propertySheets\3dsmax.common.tools.settings) which included the faulty definition. I fixed it an no more complaints from VS.
_CRT_SECURE_CPP_OVERLOAD_STANDARD_NAMES_COUNT = 1 means that compiler should replace all old C run-time routines such as sprintf, strcpy, strtok with new versions such as strprintf_s, strcpy_s, strtok_s and similar. It goes in pair with following definition _CRT_SECURE_CPP_OVERLOAD_STANDARD_NAMES = 1.
More you can find here: (MSDN) https://msdn.microsoft.com/en-us/library/ms175759.aspx. However I tried to use this but without success. It says that you can use this only for statically allocated buffers like char buffer[32], but compilers was still complaining bout unsecure strcpy.
I'm having some real difficulties porting a really old Visual Studio 97 C++ project to Visual Studio 2010. Let me begin by first giving a little background on the errors I was getting immediately prior to this new LNK1224 error because they may be related, but I'm not sure.
Prior to my new error I was receiving this error:
error LNK2005: "void __cdecl operator delete(void *)" (??3#YAXPAX#Z) already defined in LIBCMT.lib(delete.obj) nafxcw.lib(afxmem.obj)
Through some digging I found that the reason for this error was because both the MFC and CRT libraries contain definitions for "new" and "delete" so they were colliding. Microsoft provides 2 solutions for this detailed in http://support.microsoft.com/kb/q148652/ . One of them was to make sure that in all your files you always include the MFC headers (afx stuff) first. Well there are about 100 files in this project and I just got tired of trying to find the files that were including resources in the wrong order. So I went with the other solution which is basically forcing libraries to load in a particular order. Basically you have to tell the compiler to ignore a particular library so that you can load it explicitly your self in the order that you choose. In my case, it was nafxcw.lib.
So under Project Properties --> Linker --> Input, I explicitly ignored nafxcw.lib and then explicitly included it at the front of the list.
So after doing this, my LNK2005 errors went away. But they were replaced with one single link error.
error LNK1224: invalid image base 0x287600000
I don't know if I fixed my previous link errors correctly and this new link error is in fact the next thing I have to deal with, or I simply created a more critical link error that is basically stopping the linking process before it gets to my original LNK2005 errors. In either case, there isn't much information I could find on this error. Microsoft doesn't say much in this link about it http://msdn.microsoft.com/en-us/library/3ya3f8wz%28v=vs.80%29.aspx
You specified an invalid base address for the image. Base addresses must be 64KB aligned (the last four hex digits must be zero) and image base must fit within a 32-bit signed or unsigned value.
This isn't all that helpful to me and there seems to be no other clues as to where this problem is coming from. I don't know what the next step is.
OK, so it looks like I have solved my own problem. Here is what I did. I needed to know where the heck this number was coming from so I simply used Notepad++ to do a word search through all the project files looking for "2876" which I got from the error message "LINK : fatal error LNK1224: invalid image base 0x287600000". I found that in the project file (.vcxproj) had the following entry in it:
<BaseAddress>0x287600000</BaseAddress>
So I opened it up and sat there wondering how this number was wrong. I mean I don't even know what this field is for. I didn't even generate this file, M$ made it. Why would the IDE create it's own input file incorrectly? Anyways, as I was trying to google this "BaseAddress" item to figure out what it was, it dawned on me that it looked like there were too many zero's. So I went back and counted and sure enough, this wasn't a 32-bit number, it was a 36-bit number. Deleted one of the zero's, recompiled, and boom it worked. Low and behold, that's kind of what the defintion I looked up, mentioned in the problem statement, hinted at looked up earlier on MSDN but it didn't click.
I don't make a habit to rummage through auto generated files very often so I never questioned that this may be the problem.
I have a game which uses std::wstring as its basic string type in thousand of places as well as doing operations with wchar_t and its functions: wcsicmp() wcslen() vsprintf(), etc.
The problem is wstring is not supported in R5c (latest ndk at the time of this writting).
I can't change the code to use std::string because of internationalization and I would be breaking the game engine which is used by many games ...
Which options do I have?
1 - Replace string and wstring with my own string classes
This would give me better platform independency, but it is ridiculous to reimplement the wheel.
I've already started with a COW implementation of strings. I need it to be COW because I use them as keys in hash_maps.
This is of course lots of work and error prone ... but it seems it is something I can do.
2 - Try to fix the NDK recompiling the STLPort with my own implementations of the wide char string functions of the C standart library (wcslen, mbstowcs ... )
This would be the preferable way ... but I have no idea how to do it :(
How do I replace a function (lets say wcslen) in the libstdc++.a or libstlport_static.a? (not sure where they are :()
And as well I'm not sure which functions I need to reimplement, I know wcslen is not working so I guess they should be all ...
3 - Do you have any other idea?
I can't wait for an official fix for this and I will have to go with option #1 if I can't realize how to do #2.
I've read somewhere that if you target 2.3 you can use wstrings, but I ought to target Android 2.1.
PS: Forgot to say I need to use STL of course, but no RTTI and I can live without exceptions.
Thanks in advance!
Try out CrystaX's NDK. It has had stl support long before the official google one. The current version (r5), which is based off the of the official ndk r5, is still beta 3, but it does have wchar_t support.
http://www.crystax.net/android/ndk-r5.php
I'm suffering from the same problem as you, but my only other thought is to load the strings via the JNI (as jstring* in native land), then convert them to UTF characters as necessary. Take a look at the available JNI string functions here:
http://download.oracle.com/javase/1.5.0/docs/guide/jni/spec/functions.html#string_operations
Qt provides an excellent copy-on-write, international-friendly string implementation, QString, that is LGPLed.
You could, in theory extract it from the Qt source and use it in your own project. You will find the QString implementation in src/corelib/tools/qstring.h and .cpp in a Qt source download. You would also need the QChar, QByteArray, QAtomic, and QNamespace includes/classes (all under the corelib folder,) and you should define QT_NO_STL_WCHAR when compiling. (For this I would compile by hand or using my own script/Makefile.) Not simple, but once you get it up and running your life will be a lot simpler. It's better than reinventing the wheel, because it comes with loads of convenience functions and features.
Rather than stripping out just QString, you could also just use the QtCore module as a whole. See the android-lighthouse project for a Qt port to Android. (Also, it might be better to get your sources from there than from the above "vanilla" link, regardless of what you do.)
I'm trying to use the new OpenCV 2.0 API in MS Visual C++ 2008 and wrote this simple program:
cv::Mat img1 = cv::imread("image.jpg",1);
cv::namedWindow("My Window", CV_WINDOW_AUTOSIZE);
cv::imshow("My Window", img1);
Visual Studio returnes an unhandled exception and the Console returns:
OpenCV Error: bad flag (parameter or structure field)
(Unrecognized or unsupported array type) in unknown function,
file ..\..\..\..\ocv\opencv\src\cxcore\cxarray.cpp, line 2376
The image is not displayed. Furthermore the window "My Window" has a strange caption: "ÌÌÌÌMy Window", which is not dependent on the name.
The "old" C API using commands like cvLoadImage, cvNamedWindow or cvShowImage works without any problem for the same image file. I tried a lot of different stuff without success.
I appreciate any help here.
Konrad
As I just commented, imread isn't working for me either. A little googling shows other people having the same problem; I guess it's a bug in the library code. For now, here's a hacky workaround:
IplImage* img = cvLoadImage("lena.jpg");
cv::Mat lena(img);
cvReleaseImage(&img);
This way, you can at least use the C++ API for the rest of your stuff.
There's help for this issue.
The solution is, that the usual proposed opencv library files in the linker are not working properly. Instead try to use the debug library files by this:
In Visual C++:
go to Project->Properties (or Alt-F7)
Configuration Properties->Linker->Input->Additional Dependencies
replace the usual
" cv210.lib cxcore210.lib highgui210.lib" by
" cv210d.lib cxcore210d.lib highgui210d.lib" - which are the debugging libraries.
The OpenCv 2.0 API commands should work now.
I had the same problem described above which turns out to be caused by the settings of the linker.
I found the answer in another thread,
OpenCV 2.3 and Visual Studio 2010.
To repeat it here:
Properties of your project (right click on it)
C/C++
General
include directory add the < your directory >\OpenCV2.3\include\opencv2, < your directory >\OpenCV2.3\include\opencv and < your directory >\OpenCV2.3\include
Linker
General
List item
Input
Add all the libs like opencv_core230d.lib opencv_highgui230d.lib and so on...
Once I've done the above, I can run imshow and imread + all other cpp functions seamlessly! OP's problem has probably already been resolved, but hopefully this will be useful to other people who are led here looking for the same solution.
Are you sure you added the whole path starting from /home/.... I had the same problem as you but when I added the whole path, things work out pretty well. The whole path had to be added despite the fact the path exists in the include files.
imread in openCV unlike Matlab does not return an error when file/folder is not found - instead it returns a null matrix, which in turn is reflected as an error during imshow.
Also, imread does not look for image files in the included folders or the workspace. So, specify the entire path whenever possible.
Please take a note of this for future references.
Firstly, you'd better compile your own version OpenCV.
I had the same error with the build (I got from Sourceforge), and solved by compiling my own version in debug and release versions.
And make sure you change the original system env variable PATH to the new build folder build/bin, then you could build and run the imshow() in Debug mode.
I believe this might be related to unicode.
Try the macro _TEXT()
For example:
cv::Mat img1 = cv::imread(_TEXT("image.jpg"),1);
Unicode in Visual C++ 2