C / C++ / MFC

FirstPrevNext

Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE emrosa4-Jul-02 4:18 emrosa 4-Jul-02 4:18  Thanks for your help....I tried that but it did not work...however, you gave an idea and I put myself to think deeply and I just found this: Let be a multidimensional array MArray[max1stdim][max2nddim]...[maxnthdim], the indexes referencing certain location be i1,i2,...,in. If MArray is declared as int MArray[max1stdim*max2nddim*...*maxnthdim]; then the index referencing the location (i1,i2,...in-1,in) will be: index=(i1*max2nddim*max3rddim*...*maxnthdim)+(i2*max3rddim*...*maxnthdim)+...+(in-1)*maxnthdim+in Then for an array MArray[4][3][2] declared as MArray[3*3*3], the single index to reference (i,j,k)will be: index=(i*4*3)+(j*2)+k I have tried this and it works! Thanks again..... Eric Manuel Rosales Pena Alfaro PhD student Unversity of Essex Wivenhoe Park Colchester, CO4 3SQ Essex, Uk email: emrosa@essex.ac.uk tel: +44-01206-87311 Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE Alan Chambers4-Jul-02 5:56 Alan Chambers 4-Jul-02 5:56  Can ask one question? how can you have "an array MArray[4][3][2] declared as MArray[3*3*3]"? If you are trying to access the 3*3*3 MArray at position [4][3][2] using a single index then that is impossible, because the rows and columns will be indexed from 0 through to 2 for each x,y and z column, so 4 and 3 are invalid index positions. If you were trying to access the last record in a 3*3*3 matrix, the single index value you require is 26 (0 to 8, 9 to 17, 18 to 26). Using my equation earlier, if you were trying to index the last position in the 3d array, so XPos = 2, YPos = 2 and ZPos = 2, you would get: (2 + (2 * 3)) + (2 * 3 * 3)) = 26. the index position you desire. Martin has implemented the coded version of the equation (but in the opposite direction to the way I wrote my equation) : b[i*(d2*d3) + j*d3 + k] = x; i = ZPos j = YPos k = XPos d2 = Height d3 = Width *Please note I made a small error before when I wrote (yPos + Height), this is probably why you couldn't get it to work, it should have read (yPos + Width), which I have now changed it to. Without this correction the formula would only *work* if the 3d array was a cube. It should now work regardless of dimensions. Alan. "When I left you I was but the learner, now I am the master" - Darth Vader Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE 4-Jul-02 11:25 4-Jul-02 11:25  Sorry for the mistake...the dimensions of the array MArray[4][3][2] declared as MAarray[4*3*2]...and trying to access element Marray(3,3,2).... The rest of the formula is correct and your correction as well... Thanks, Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE Martin Ziacek4-Jul-02 3:43 Martin Ziacek 4-Jul-02 3:43  Try this: int d1 = 3; int d2 = 4; int d3 = 5; BYTE x = 0; BYTE *b = new BYTE[d1*d2*d3]; for (int i = 0; i < d1; i++) { for (int j = 0; j < d2; j++) { for (int k = 0; k < d3; k++) { b[i*(d2*d3) + j*d3 + k] = x; x++; } } } I hope it does what you need. Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE emrosa4-Jul-02 4:20 emrosa 4-Jul-02 4:20  Thanks for your help....I have tried that and it did work...therefore, making a generalization for this, I just found this: Let be a multidimensional array MArray[max1stdim][max2nddim]...[maxnthdim], the indexes referencing certain location be i1,i2,...,in. If MArray is declared as int MArray[max1stdim*max2nddim*...*maxnthdim]; then the index referencing the location (i1,i2,...in-1,in) will be: index=(i1*max2nddim*max3rddim*...*maxnthdim)+(i2*max3rddim*...*maxnthdim)+...+(in-1)*maxnthdim+in Then for an array MArray[4][3][2] declared as MArray[3*3*3], the single index to reference (i,j,k)will be: index=(i*4*3)+(j*2)+k I have tried this and it works! Thanks again..... Eric Manuel Rosales Pena Alfaro PhD student Unversity of Essex Wivenhoe Park Colchester, CO4 3SQ Essex, Uk email: emrosa@essex.ac.uk tel: +44-01206-87311 Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE Martin Ziacek4-Jul-02 4:41 Martin Ziacek 4-Jul-02 4:41  Would be nice if you will write article about it and compare performance between different types of allocations and accesseing that memory. Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE emrosa4-Jul-02 4:46 emrosa 4-Jul-02 4:46  How will I do that? What allocation and accesing methods?. Where will I submit the article?...Would you like to help me? Thanks, Eric Manuel Rosales Pena Alfaro PhD student Unversity of Essex Wivenhoe Park Colchester, CO4 3SQ Essex, Uk email: emrosa@essex.ac.uk tel: +44-01206-87311 Re: MULTI-DIMENSIONAL ARRAYS SEEN AS ONE-DIMENSIONAL ONE Martin Ziacek4-Jul-02 4:53 Martin Ziacek 4-Jul-02 4:53  I understood you think addressing multi-dimensional array with one index you are computing by yourself is a faster method than addressing multi-dimensional array with pointers. So what I think is a demonstration program using two arrays of the same size but different access methods and allocation and comparison between speeds. And obviously, you will submit article with that demo program here at CP. Link before compile (?) in MSDev Cristian Amarie4-Jul-02 1:47 Cristian Amarie 4-Jul-02 1:47  Can anyone let me know why I see 'Linking...' and after this 'Compiling...' ? Thank you. [Sample] --------------------Configuration: Shared - Win32 Release-------------------- Compiling... EnumString.cpp Linking... Creating library .\Release/Shared.lib and object .\Release/Shared.exp LINK : warning LNK4089: all references to "IMAGEHLP.dll" discarded by /OPT:REF Compiling... Report.cpp Report2.cpp ReportPreviewWnd.cpp Shared.cpp Generating Code... Shared.dll - 0 error(s), 3 warning(s) Re: Link before compile (?) in MSDev Jeremy Pullicino4-Jul-02 1:50 Jeremy Pullicino 4-Jul-02 1:50  Your monitor might be upside down... "Hey man, Taliban, Tali me Banana." Re: Link before compile (?) in MSDev Tibor Blazko4-Jul-02 1:51 Tibor Blazko 4-Jul-02 1:51  i guess slow refresh rate t! Re: Link before compile (?) in MSDev Tibor Blazko4-Jul-02 3:56 Tibor Blazko 4-Jul-02 3:56  big sorry, i did not realize current forum hm, it me already happened too (no your action after build run?), do not worry t! Re: Link before compile (?) in MSDev Martin Ziacek4-Jul-02 4:22 Martin Ziacek 4-Jul-02 4:22  Isn't it that feature of Pentium when it computes some instructions before it needs them? ActiveX events from automation server (.EXE) ? 4-Jul-02 1:36 4-Jul-02 1:36  I'm writing a automation EXE server in VC++ 6.0 and I want to send COM/ActiveX events to the calling program... is it possible and how ? The project was created as a MFC application with Automation support and works fine as a Automation server for my client application. Now I just need to notify the client of events in my automation EXE application. Please help.... thanks Re: ActiveX events from automation server (.EXE) ? MrBean4-Jul-02 22:21 MrBean 4-Jul-02 22:21  *BUMP* I really need help here, guys Re: ActiveX events from automation server (.EXE) ? MrBean7-Jul-02 22:00 MrBean 7-Jul-02 22:00  *bump* (again) Question about Spy++ User 66584-Jul-02 1:19 User 6658 4-Jul-02 1:19  Hi! I use Spy++ to watch the messages a windows sends. For example if the user clicks on a menu item Spy++ loggs the following messages: WM_MENUSELECT WM_COMMAND wNotifyCode: 0 (sent from a menu) wID:7224 If I doubleclick on the WM_COMMAND log a new windows opens show the parameters (wParam and lParams) sent by WM_COMMAND. Often these values have a wParam like 00001C38. Now I want to emulate this message. I tried: SendMessage(wnd,WM_COMMAND,00001C38,0); But this doesn't compile. So how do I need to write the wParam/lParam parameter shown in Spy++ in order to use it in my own SendMessage function? thanks in advance modified 12-Sep-18 21:01pm. Re: Question about Spy++ Nish Nishant4-Jul-02 1:49 Nish Nishant 4-Jul-02 1:49  Instead of 00001C38 use 0x00001C38 Nish Author of the romantic comedy Summer Love and Some more Cricket [New Win] Review by Shog9 Click here for review[NW] Re: Question about Spy++ User 66584-Jul-02 8:16 User 6658 4-Jul-02 8:16  Thanks a lot! modified 12-Sep-18 21:01pm. #pragma once and #ifndef -Dy4-Jul-02 0:35 -Dy 4-Jul-02 0:35  Is there a difference between #pragma once and #ifndef _THISDEF_ #define _THISDEF_ // class declaration ... #endif ? I've always defaulted to using the #pragma once but have noticed that others tend not to use it and favour the #ifndef instead, why is this? Am I missing something very simple? Dylan Kenneally London, UK Re: #pragma once and #ifndef Alan Chambers4-Jul-02 1:16 Alan Chambers 4-Jul-02 1:16  Good point. The #ifndef statement is only included in VC++ for backwards compatibility on older compilers and (according to MSDN) the #if !defined statement is supposed to be used instead (though it gives no reason as to why? surely the #ifndef code would be more portable because of this?). The #pragma statments are platform independent compiler directives, but I'm not sure of the difference between that and #ifndef or #if !defined, if there is any at all. I would like to know for sure though . Alan. "When I left you I was but the learner, now I am the master" - Darth Vader Re: #pragma once and #ifndef Alexandru Savescu4-Jul-02 8:51 Alexandru Savescu 4-Jul-02 8:51  There is a difference between #ifndef and #if !defined. You can have some operations in the latter case. For instance: // in header.h #define A #define B // in anotherheader.h // you can use #if !defined A AND !defined B #include "header.h" #else #include "secondheader.h" #endif Best regards, Alexandru Savescu Re: #pragma once and #ifndef markkuk4-Jul-02 1:21 markkuk 4-Jul-02 1:21  The #ifndef block is portable, #pragma is non-portable by definition. Re: #pragma once and #ifndef Michael Dunn4-Jul-02 7:32 Michael Dunn 4-Jul-02 7:32  The #ifndef version has a nice advantage in that you can test whether a header has been included by checking #ifdef _THISDEF_. You can't do that with the #pragma version. And, as has been mentioned, the #ifndef way works on every compiler (for those who have to worry about portability). --Mike-- Just released - RightClick-Encrypt - Adds fast & easy file encryption to Explorer My really out-of-date homepage Sonork-100.19012 Acid_Helm OnUpdate(CCmdUI* pCmdUI) isnt called John Oliver3-Jul-02 23:47 John Oliver 3-Jul-02 23:47  Hello I have two CDialog based classes, both with toolbars, using the technique used in the MSDN MFC sample dlgcbr32 which describes how to use toolbars and status bars in dialogs by overriding these particular control classes I created the first class a few years ago and it works. I cannot get the toolbar button states to update in the 2nd class which I recently created, although the class is similar to the first. I cannot get OnUpdateButtonX(CCmdUI* pCmdUI) to be called - CDlgToolBar::OnIdleUpdateCmdUI isn't being called. Can anybody tell me what methods or messages need to be present for OnIdleUpdateCmdUI to get called. Thanks, John

Last Visit: 31-Dec-99 18:00     Last Update: 15-Jun-24 0:41 Refresh ᐊ Prev1...16642166431664416645166461664716648166491665016651 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.