C / C++ / MFC

FirstPrevNext

Re: Why I can't use the compiler option /GH in VC6.0 ? jiangry11-Jul-03 3:14 jiangry 11-Jul-03 3:14  thanks, first. but it's not same with you. /Gh is with _penter() and /GH is with _pexit(), isn't it? I know /Gh is ok , but /GH can't be used. Re: Why I can't use the compiler option /GH in VC6.0 ? Mike Dimmick11-Jul-03 3:19 Mike Dimmick 11-Jul-03 3:19  Sorry, the option was added in VS.NET 7.0 (i.e. compiler version 13.x). VC 6.0 doesn't have this option. [edit] I just found an article on simulating a _pexit()[^] that might help you. The basic trick is to modify the return address of the function on the stack within the _penter() function to point to your _pexit(), then store the true return address elsewhere. When the function returns, it returns to _pexit(), which should then jump to the true return address when done. [/edit] -- Mike Dimmick Re: Why I can't use the compiler option /GH in VC6.0 ? jiangry11-Jul-03 3:35 jiangry 11-Jul-03 3:35  Oh, thanks a lot. and now , I want to call a hook function when a fuction return, but i will not add some codes in it, what shall i do? which function is similar with _pexit()? MSDN 2003 working with Visual studio 6.0a LizardKingSchwing11-Jul-03 2:31 LizardKingSchwing 11-Jul-03 2:31  Hi I have installed MSDN 2003 which was recently released free by microsoft however I cannot link it into my visual studio (like when I press F1 on a class e.g CString it does not search the MSDN) ; It pops up a message about MSDN not being installed properly. Does anyone know of a tweak to get this back working .. cheers LK--< Re: MSDN 2003 working with Visual studio 6.0a Mike Dimmick11-Jul-03 2:37 Mike Dimmick 11-Jul-03 2:37  The new 'document viewer' isn't compatible with the old one. VC 6.0 doesn't know how to launch the new one. If you have an MSDN subscription, you should have received a separate disc marked 'Installs with Visual Studio 6 only' - at least, our MSDN Universal does. If you don't have this disc, for context-sensitive help in VS 6 you'll need to keep a previous version of the documentation around, i.e. either that which came with Visual Studio 6, or a release before April 2002 (I believe). I wasn't aware that MS had ever officially released a free version. -- Mike Dimmick Re: MSDN 2003 working with Visual studio 6.0a LizardKingSchwing11-Jul-03 2:48 LizardKingSchwing 11-Jul-03 2:48  Thanks for the reply Its listed on the microsoft site for download heres the links http://download.microsoft.com/download/b/d/c/bdccea1b-96d9-4ad9-8045-56619af95835/qtr40enud1.img http://download.microsoft.com/download/b/d/c/bdccea1b-96d9-4ad9-8045-56619af95835/qtr40enud2.img http://download.microsoft.com/download/b/d/c/bdccea1b-96d9-4ad9-8045-56619af95835/qtr40enud3.img LK--< Re: MSDN 2003 working with Visual studio 6.0a David Crow11-Jul-03 2:55 David Crow 11-Jul-03 2:55  Mike Dimmick wrote: ...or a release before April 2002 (I believe). October 2001, to be exact. break pointer failed El'Cachubrey11-Jul-03 2:29 El'Cachubrey 11-Jul-03 2:29  I have trouble with break pointers. Sometime one dont wont work and my VC++ 6.0 notice mee what "Break points cannot be set and will be disabled." Why this occur .This became really nasty problem for me, because sometime this work fine and suddenly this fail by another reason.I dont change settings of my project (suppose enable to set break pointer spend on only from settings).My project its console application, which run in debug mode.And any time again same trouble arised, but i continued work with one, but after any changest in source code break pointers work fine once again. Thanks Re: break pointer failed Mike Dimmick11-Jul-03 2:52 Mike Dimmick 11-Jul-03 2:52  When a breakpoint is on an illegal line, VC will try to move the breakpoint to the next legal line. If there's no legal next line, the breakpoint is disabled. A line can be considered illegal if it doesn't contain a complete statement, for example, you have a multi-line if condition, or an assignment that is formatted into multiple lines, and you try to set the breakpoint on anything other than the first line. It will also be considered illegal if the source file isn't included in any of the modules loaded on startup. To add modules to debug on startup, you can go to Project > Settings > Debug tab > Additional DLLs category, and add the modules you want loaded at startup. VC 7.x is better in this respect, IIRC; if there's a breakpoint set in a module that isn't listed to be loaded at startup, the debugger simply marks the breakpoint with a '?' symbol in the source, then tries to set it when the module loads. I can't remember what happens if this fails. Another reason this can fail is if the debugger crashes before the workspace is saved. Breakpoint locations are saved in the workspace options file (Workspace.opt in VC6, or .vco in eMbedded Visual C++), which is saved when you save the workspace. If the debugger crashes between saving the workspace and making changes to the source, or you change the source outside the IDE, the breakpoints remain on the line number that they were set on, even if the code's moved, when you next open the workspace. Breakpoints set in the Disassembly window are saved relative to the module base address, IIRC. For more information on debugging with VC 6.0, read John Robbins' Debugging Applications[^]. This is the first edition; the new book only covers VS.NET. Unfortunately it's out of print. -- Mike Dimmick Re: break pointer failed Cedric Moonen11-Jul-03 2:52 Cedric Moonen 11-Jul-03 2:52  Break points must be set on lines that "will produce" some assembly code. For example, you can't set a breakpoint on a variable declaration: int Value; //You cannot set a break point here But if you assign some value to this variable, it will work (in assembly code, you will have an instruction that loads some registers) int Value = 0; //You can set a breakpoint here So, look if you breakpoint is set on a valid line ! Hope this helps Re: break pointer failed Ryan Binns11-Jul-03 3:07 Ryan Binns 11-Jul-03 3:07  The other two have mentioned about setting a breakpoint on an invalid line, but if you've put a condition on the breakpoint that the debugger can not understand, then it will disable the breakpoint as well. Check this too. Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact" Re: break pointer failed El'Cachubrey11-Jul-03 3:54 El'Cachubrey 11-Jul-03 3:54  Thanks all very much But i want give more restrict trouble reference, since all suggest not helped. When i begin execute my progect, to appear message ...exe file doesnot contain debugging information and further break pointers failed. My project have Win32Debug flag. But i also have custom make file, maby trouble here. Also i want underline what break pointers failed wherever i placed it in my source file althought from recent time all work fine. Sorry for bad eanglesh. CASE tool Jerome Conus11-Jul-03 2:28 Jerome Conus 11-Jul-03 2:28  Hi ! Where I was working before, we had Together as CASE Tool. Where I'm working now, they cannot afford it, so I'm wondering if someone here knows about a good CASE Tool (I'm don't need a full-featured app), which is either free or cheap ! Thank you for your suggestions ! Jerome Re: CASE tool Ryan Binns11-Jul-03 2:33 Ryan Binns 11-Jul-03 2:33  There's one called ProxyDesigner somewhere on CP (in the free tools section I think), but it may not be good enough for what you want. Poseidon[^] can be downloaded free, but it does run under Java, so may be a bit slow. I'm sure there are others. Just google for them Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact" Re: CASE tool Jonathan Gilligan11-Jul-03 6:05 Jonathan Gilligan 11-Jul-03 6:05  Together is written in Java too, so speed should not be the issue. The bigger problem with Poseidon is that its UML model is too strictly tied to the Java language, so it doesn't always fit C++ programming very smoothly. Also, CASE is useless without round-trips and Poseidon only supports this for Java. Re: CASE tool Ryan Binns11-Jul-03 6:14 Ryan Binns 11-Jul-03 6:14  Cool. I was just giving a couple of examples to get him started Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact" Re: CASE tool RK_200011-Jul-03 8:14 RK_2000 11-Jul-03 8:14  Enterprise Architect. It's built by an Australian company. I have used it both in my current and previous job and it is a very solid tool and quite affordable too (only $95 US for a single licence): http://www.sparxsystems.com.au/ images in column headings si_6911-Jul-03 2:14 si_69 11-Jul-03 2:14  Hi Does any1 know how to display images in the column headers of a CListView thanks simon Re: images in column headings Ryan Binns11-Jul-03 2:30 Ryan Binns 11-Jul-03 2:30  Similar to item images for a list control. Set an image list for the header control, and set the image index using CHeaderCtrl::SetItem(). Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact" Serialization of COLEDATETIME - STRANGE THING! giorgos11-Jul-03 2:01 giorgos 11-Jul-03 2:01  there is a strange thing happening when trying to serialize a COleDateTime member variable of a class. In fact, i 'm writting: if (ar.IsStoring()) { // storing ar << start_date.m_dt; ar << start_date.m_status; } else { // loading ar >> start_date.m_dt; ar >> start_date.m_status; // this line does not compile!!!! } the commented line does not compile!!! any ideas why? thanks for your time Re: Serialization of COLEDATETIME - STRANGE THING! Ryan Binns11-Jul-03 2:25 Ryan Binns 11-Jul-03 2:25  You can't serialize an enumerated type like that. You can just write the whole COleDateTime object in one go: ar << start_date; Hope this helps, Ryan "Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact" compiler option pranavamhari11-Jul-03 1:43 pranavamhari 11-Jul-03 1:43  hi, What is your opinion about the expand 'any suitable' function inline ? Will that cause any problem ? why does the compiler issue warnings while compiling with that option ? regards..... Hari Krishnan Re: compiler option Mike Dimmick11-Jul-03 3:02 Mike Dimmick 11-Jul-03 3:02  'Any suitable' basically means that the compiler gets to look at every function you wrote and decide whether to inline it. Basically it tries to see whether the cost of inlining the function is more or less than the cost of not doing so - if less, it will probably inline it. The problem will likely be that your code size will be very large. Most references actually recommend trying to minimize code size in the general case, because more code = more working set = more page faults, typically. A page fault can swamp any benefit you might get from code that would be faster if it had all been in RAM at the same time. As for the warnings, I have no idea, since I've never used this option. Post a list of the warnings! -- Mike Dimmick Re: compiler option pranavamhari11-Jul-03 3:36 pranavamhari 11-Jul-03 3:36  thanks. i better try to minimize code size Hari Krishnan Indicate the sort order in list view si_6911-Jul-03 1:33 si_69 11-Jul-03 1:33  Hi All i have been given the following code which is used to Indicate the sort order in list control control, by the use of up and down arrows (common windows feature) this works fine for a list control, but i cant get it to work with a list view, first off i get an error with CHeaderCtrl* HeaderCtrl = GetHeaderCtrl(); saying its an undeclared identifier has ne1 ne ideas thanks si private: int m_CurrentSortItem; bool m_SortAscending; Constructor - m_CurrentSortItem = 0; m_SortAscending = true; Override Create then after Create is called - HDITEM HeaderItem; HeaderItem.mask = HDI_FORMAT | HDI_BITMAP; CHeaderCtrl* HeaderCtrl = GetHeaderCtrl(); HeaderCtrl->GetItem(m_CurrentSortItem, &HeaderItem); if (HeaderItem.hbm != 0) { DeleteObject(HeaderItem.hbm); HeaderItem.hbm = 0; } HeaderItem.fmt |= HDF_BITMAP | HDF_BITMAP_ON_RIGHT; HeaderItem.hbm = (HBITMAP)LoadImage(AfxGetInstanceHandle(), MAKEINTRESOURCE(m_SortAscending ? IDB_UP : IDB_DOWN), IMAGE_BITMAP, 0, 0, LR_LOADMAP3DCOLORS); HeaderCtrl->SetItem(m_CurrentSortItem, &HeaderItem); Finally - void OnLvnColumnclick(NMHDR *pNMHDR, LRESULT *pResult) { LPNMLISTVIEW pNMLV = reinterpret_cast<lpnmlistview>(pNMHDR); HDITEM HeaderItem; HeaderItem.mask = HDI_FORMAT | HDI_BITMAP; CHeaderCtrl* HeaderCtrl = GetHeaderCtrl(); HeaderCtrl->GetItem(pNMLV->iSubItem, &HeaderItem); if (HeaderItem.hbm != 0) { DeleteObject(HeaderItem.hbm); HeaderItem.hbm = 0; } HeaderItem.fmt |= HDF_BITMAP | HDF_BITMAP_ON_RIGHT; m_SortAscending = (m_CurrentSortItem != pNMLV->iSubItem) ? true : !m_SortAscending; HeaderItem.hbm = (HBITMAP)LoadImage(AfxGetInstanceHandle(), MAKEINTRESOURCE(m_SortAscending ? IDB_UP : IDB_DOWN), IMAGE_BITMAP, 0, 0, LR_LOADMAP3DCOLORS); HeaderCtrl->SetItem(pNMLV->iSubItem, &HeaderItem); if (m_CurrentSortItem != pNMLV->iSubItem) { HeaderCtrl->GetItem(m_CurrentSortItem, &HeaderItem); HeaderItem.fmt &= ~(HDF_BITMAP | HDF_BITMAP_ON_RIGHT); if (HeaderItem.hbm != 0) { DeleteObject(HeaderItem.hbm); HeaderItem.hbm = 0; } HeaderCtrl->SetItem(m_CurrentSortItem, &HeaderItem); m_CurrentSortItem = pNMLV->iSubItem; } SortItems(SortFunc, (DWORD_PTR)this); *pResult = 0; }

Last Visit: 31-Dec-99 18:00     Last Update: 13-Jun-24 21:42 Refresh ᐊ Prev1...14639146401464114642146431464414645146461464714648 Next ᐅ

General    News    Suggestion    Question    Bug    Answer    Joke    Praise    Rant    Admin   

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.