Easy question on numeric limits

TheMSsForum.com: The Microsoft Software Forum

The MSS Forum ‹ VC
- Archive
  - Biz
  - MCSE
  - CRM
  - Drivers
  - Framework
  - ADO
  - ASP
  - Compact
  - Forms
  - Dotnet
  - C#
  - VB
  - FontpageGen
  - Excel
  - WorkSheet
  - Exchange
  - Setup
  - Fox
  - Fontpage
  - ASP
  - IIS
  - Entourage
  - Money
  - Messanger
  - PocketPC
  - Powerpoint
  - Project
  - Publisher
  - Excel
  - VB
  - Security
  - Portal
  - Services
  - SQLServerDev
  - SVCS
  - SQLServer
  - VB
  - VC
  - MFC
  - ExcelGen

- Previous
  - 1
    - resource leak in destructor? Hello everyone, I understand there should be no exception thrown from destructor. But exception will block the remaining code to be executed until catch block. In my destructor, I free resources step by step. But at some first step, there may be exception in destructor, so I suspect the remaining resources can not be freed correctly. In my class, I use member variable to hold resources. My questions are, - My current solution is to declare each member as smart pointer, so that to ensure the destructor will be called. - I am not sure whether my concern is correct or no such issue -- e.g. member's destructor will always be called or not, even if I do not wrap them with auto_ptr? thanks in advance, George Tag: Easy question on numeric limits Tag: 295638
  - 2
    - Calling VB.net from C++ In the past I have used VB 6 programs that I called from VS 2003 C++ by using the #import "Mydll.dll" no_namespace in my c++ program I had the dll located in the c++ project where the c++ program resided. This works for me. However I can not seem to do this for a VB.net program Here is my code // VBCallfromC.cpp : Defines the entry point for the DLL application. #include "stdafx.h" //#include "iostream.h" #include <comdef.h> #import "VbCalendar.dll" no_namespace clsclsVbcalPtr ptrclsVbcal(__uuidof(clsVbcal)); SAFEARRAY FAR* RTNGArray = NULL; When I compile this I get the following message fatal error C1083: Cannot open type library file: 'c:\testtheory \vbcallfromc\vbcallfromc\vbcalendar.dll': Error loading type library/ DLL. The dll is in that directory The 2005 VB.net program is a com interop program and is registered automatically when I compile the dll. I tried my calling code in VS c + + 2003 and in VS c++ 2005 and nether compiles. I also tried playing around with the around with the CLS but no luck. Any ideas? thanks Josh Tag: Easy question on numeric limits Tag: 295637
  - 3
    - different result on x86 and x64 Hello everyone, For the following code, on x86 build, the output is 000000C8 000000C8 and on x64 build, the output is CCCCCCCC000000C8 00000000000000C8 I think the reason is, for x64, both void* and void** are 64bit, for for x86 void* and void** are 32 bit, and for int, always 32bit for both x86 and x64. This line of code "void* b2 = *b1", dereference b1 and makes it into 64-bit on x64, which concatenate value of a 0x000000C8 and previous DWORD 0xcccccccc, so forms the result 0xCCCCCCCC000000C8. But on x86, when dereference b1 and makes it into 32-bit on x86, and recovers to original value 0x000000C8. In my understanding correct? [Code] #include <iostream> using namespace std; int main() { int a = 200; void** b1 = (void**)&a; void* b2 = *b1; int b3 = (int)*b1; cout << (void*)b2 << endl; cout << (void*)b3 << endl; return 0; } [/Code] thanks in advance, George Tag: Easy question on numeric limits Tag: 295623
  - 4
    - video capture Anyone know the best way to do video capture from a webcam in VC++? Thanks. Tag: Easy question on numeric limits Tag: 295622
  - 5
    - Partial template specialization with a reference as a parameter - Hi, I'm having a problem while using references as template parameters and partial specialization. Here's a simplified case where it doesn't work (with VC++ Express 2008) : ---- #include <iostream> int C; template<class A, int& B> class D { D() { std::clog << "Base." << std::endl; } }; template<int& B> class D<char, B> { D() { std::clog <<"Special: char." << std::endl; } }; int main(int argc, char** argv) { D<int, C> d1; D<char, C> d2; return 0; } ---- This should produce two lines: "Base." and "Special: char.". However, I get "Base." two times, which means the specializing didn't get picked up. If I replace the reference by a pointer (or anything else), it works correctly. While using a pointer would be an OK workaround, I would really prefer to keep using references, since it compile correctly in GCC and is our main compiler. I couldn't test it with other VC++ versions. Anybody has any clues on how to fix this? Is this a recognized bug? I looked around but couldn't find anything. Thanks, Fran=E7ois Ferland Tag: Easy question on numeric limits Tag: 295611
  - 6
    - About duplicate variable names in macros. There is a problem when using macros. Let assume that some macro is defined #define SomeMacro( memvar1 ) \ { \ int *memvar2 = NULL; \ ... \ memvar2 = memvar1; \ SomeFunction( memvar2 ); \ } Someone may call the macro specifying as an argument a variable with the name memvar2. In this case instead of memvar2 = memvar1; /* from the macro definition */ we will get memvar2 = memvar2; It is not what we wanted to get. At first I decided to use a special prefix for local macro variable names such as m_. For example m_memvar2. Well, the code works. However one macro may call another macro and the same problem can arise. Is any trick in C which allows to escape the problem? Vladimir Grigoriev Tag: Easy question on numeric limits Tag: 295608
  - 7
    - Passing a structure to a function by its address? Hello, Something else about structures I don't know or forgot I guess! Please help! Please consider the following code: ==========================================.h struct tagMsg{ long HWND; long MSG; long lPARAM; long wPARAM; long TIME; }; long ULC_KERNEL_get_msg (struct tagMsg *S) ; ==============================================.c int main() { struct tagMsg S; ULC_KERNEL_get_msg(&S); return 0; } long ULC_KERNEL_get_msg (struct tagMsg *S); { //some code! } ================================================= The above works! So why won't the following compile? ================================================h typedef struct tagMsg{ long HWND; long MSG; long lPARAM; long wPARAM; long TIME; } KM_MSG, *pKM_MSG; long ULC_KERNEL_get_msg (struct KM_MSG *S); ==============================================.c int main() { struct KM_MSG S; ULC_KERNEL_get_msg(&S); return 0; } long ULC_KERNEL_get_msg (struct KM_MSG *S) { //some code! } ================================================= I don't get it, if we create a new type by typedefing the tagMsg structure and call it KM_MSG, then this means that S is an object of KM_MSG type, which is of the tagMsg type right? So if we pass the address of S to a function that takes a KM_MSG type, why won't it work? I just want to pass the address of the structure because I want to modify its contents while in the function. So that when I return from this function the data that was modified is still available. This is slightly different from how I am used to modifying a structure's content where I usually create the strucutre object in the function using malloc and then pass the pointers around! Once again, all help is sincerely appreciated! -- Best regards Roberto Tag: Easy question on numeric limits Tag: 295604
  - 8
    - No RTTI tutorials? Hello everyone, I have searched for RTTI tutorials for 15 minutes, but find none. I want to see in depth how RTTI is implemented, for VC environment. All I found is either forum discussion without any definite result or any non-free web sites. Could anyone suggest any free online tutorial for this topic please? :-) thanks in advance, George Tag: Easy question on numeric limits Tag: 295595
  - 9
    - Linking error in VS2008(vc++) I'm making a project in vc9 in visual studio 2008. When i compile my project there are some linking errors like: 1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _sprintf already defined in libcmtd.lib(sprintf.obj) 1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _memchr already defined in libcmtd.lib(memchr.obj) To fix above errors i add libcmtd.lib in "ignore specific library" under project setting in linker tab.. But now i'm getting following 2 errors and dont way the way to fix them. 1>nafxcwd.lib(appcore.obj) : error LNK2001: unresolved external symbol ___argv 1>nafxcwd.lib(appcore.obj) : error LNK2001: unresolved external symbol ___argc Please suggest how to fix them. And mfc42d.lib is part of VC6. This library is not present in VS2008. What's the replacement of this library in VS2008. Tag: Easy question on numeric limits Tag: 295592
  - 10
    - command window of VC Hello everyone, I am using VSTS 2008, there are 4 different command windows under "Visual Studio Tools" from Programs of Start Menu. They are, Visual Studio 2008 Command Prompt Visual Studio 2008 x64 Cross Tools Command Prompt Visual Studio 2008 x64 Win64 Command Prompt Visual Studio 2008 Itanium Cross Tools Command Prompt I am using x64 machine and always use x64 binary, sometimes I will use x86 binary. What are the differences between the 4 command window and for my requirement which one(s) should I use? thanks in advance, George Tag: Easy question on numeric limits Tag: 295583
  - 11
    - product validation i need a basic product activation box and keygen for deploying my shareware project. (i can do encryption myself. i want to keep things as basic as possible. i need a validation box and registry codeing.) any links/locations for source codes? regards, marc Tag: Easy question on numeric limits Tag: 295567
  - 12
    - Antialiasing.... General question! Hello, I am using a microchip MCU which is connected to a color LCD screen (176 x 132 pixels) with a 16 bit per pixel color depth. I am displaying images, icons and text. However, each character in my text font is made from plain black and white pixels. It looks okay, but if you look carefully you can see the imperfections of characters such as "X" or "C" around their rounding edges. All my fonts are stored in bmp files, and converted to 16 bit per pixel depth format and stored in external flash. As I require, I load the data from the flash onto a bus which is common to the MCU and LCD allowing data to either be loaded in the MCU or the LCD. Upon display operations, this data is directed from the flash to the LCD so data can be displayed. however, for a clearer blend of the characters, it would be required for me to load my character in a portion of memory(for example a scratch pad area) in my processor and perform some rasterization (antialiasing) operations of my character and then output it to my LCD. The problem is that, rasterization is made up of a pretty complex algorithm and wondering if anyone knows if a rasterization module or library exists in C language. OR if anyone can tell me where I can find some information on how to process the colors of my pixels. For example if I have two pixels *not* aligned one on top of the other but touch each other at their conners. Assuming the top left corner is (0,0), suppose the first one is at coordinate (2,2) and the second one is at coordinate (3,1), and if the backgound is white what color should be pixel coordinate at (2,1) be .... 50% between black and white (like a shade of grey)????? I know this stuff can get pretty complicated, but I was wondering if there can be a simple rule of thumb that I can follow in order to create antialiasing process myself for this particular miniature color screen! All feedback is appreciated! Best regards Roberto -- Best regards Robert Tag: Easy question on numeric limits Tag: 295566
  - 13
    - compile error for function InternetOpen Hello everyone, What is wrong with my simple code below? I have tried to figure it out in 40 minutes, but failed. [Code] 1>main.cpp 1>c:\program files\microsoft sdks\windows\v6.0a\include\wininet.h(58) : error C2146: syntax error : missing ';' before identifier 'HINTERNET' 1>c:\program files\microsoft sdks\windows\v6.0a\include\wininet.h(58) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int 1>c:\program files\microsoft sdks\windows\v6.0a\include\wininet.h(58) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int 1>c:\program files\microsoft sdks\windows\v6.0a\include\wininet.h(59) : error C2143: syntax error : missing ';' before '*' 1>c:\program files\microsoft sdks\windows\v6.0a\include\wininet.h(59) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int 1>c:\program files\microsoft sdks\windows\v6.0a\include\wininet.h(59) : error C2378: 'HINTERNET' : redefinition; symbol cannot be overloaded with a typedef #include <wininet.h> int main() { InternetOpen ("Test Agent", INTERNET_OPEN_TYPE_PRECONFIG, NULL, NULL, 0); return 0; } [/Code] thanks in advance, George Tag: Easy question on numeric limits Tag: 295552
  - 14
    - Virtual execution environment? Hello, I'm fairly sure this is not the appropriate discussion group for my question, so any links to appropriate groups or information would be greatly appreciated. I'd like to set up an execution environment in which I can run unmanaged C & C++ programs written by my students as homework exercises without risking the destruction of my machine should they inadvertently (or intentionally) write code that does something malicious such as attempting to delete all files from the root on down, installing a virus, or performing some other obnoxious act. From my limited knowledge in this area I've been told that a Virtual Machine might be what I am looking for, but that's not much to go on given that I really know nothing about the specifics of the issue at this point. I'm currently using WinXP Pro. Thanks, Ray Mitchell Tag: Easy question on numeric limits Tag: 295544
  - 15
    - name decoration issue in DLL Hello everyone, I think extern C is enough to prevent name from decorated, but when we export function names from a DLL using dllexport, even if we declare at the same time with extern C, but without a DEF file to define un-decorated exported function names, the name is still decorated. I do not know why DEF file is needed? extern C is not enough to control name decoration for both DLL exported names and current DLL model linking? regards, George Tag: Easy question on numeric limits Tag: 295538
  - 16
    - debugging release version build issue in managed code Hello everyone, I asked question about debugging native code before -- mode details it is about optimized release mode x64 code which will use register to store variable which will block debugger from monitoring the variable value. Today, I debugged a managed program, also in release build for the managed program, but I do not compile it with x64, and it is for "Any CPU". I met with the same issue and when see the assembly code, it has the same pattern that putting some variable in register prevents debugger to see its value. My question is, I am not 100% confident enough (since the build is not for x64 release, but for "Any CPU" release, different build option from the native code issue before) and I want to confirm with you the same issue happens not only in native code in x64 release mode, but also in managed code in release "Any CPU" mode? thanks in advance, George Tag: Easy question on numeric limits Tag: 295535
  - 17
    - broken samples.... Most C++ samples (OLE DB Providers & Tools) would not even compile with 2005 and 2008, mostly for embarrasing (defaulting to int) reasons like: 1. untyped const 2. even undeclared variables (also assumed to be int) like for(i = 0; blah, blah blah) Now that your compiler got better (and all these "wonders" are gone) it would be nice to ask different teams to make sure that their samples still compile... Practical example, in the Platform SDK, samples, dataaccess, oledb, rowsetviewer... This application is also (binary) delivered as a tool... So, this is like telling customers: we didn't even bother to rebuild our tools... :-) Regards, Daniel Tag: Easy question on numeric limits Tag: 295525
  - 18
    - VS2008 Beta vs Release Version Is VS2008 out of Beta testing yet? Tag: Easy question on numeric limits Tag: 295502
  - 19
    - Stl map and CString trouble Hi, I'm trying to keep MFC out of this project and so I'm using the stl std::map on the heap. I was originally using a char * as the key, but I ran into the "duh, obvious -- it's comparing pointers" issue, and switched to the CString (<atlstr.h> version, NOT the CLR version) use. I thought this would solve the pointer issue since the operators for CString are overridden. Now, I declare the map as: std::map<CString, byte*>* g_VarMap; //byte is typdef'd unsigned char. also on the heap. Now, I add strings to the map and this works entirely as expected. However, later on, I try to do a find() to grab the value. Now, digging into the guts, the find() is doing an == to detect if the objects are the same. CString has overrided this operator so that it compares the guts of the strings, not the pointers. I still get find failing. Here's an example of a little dummy test. Now, as you can see here, my map has the strings in it and pointers to memory locations. - g_VarMap [5](("AA",0x003a9e0b ""),("BB",0x003a9e1b ""),("XX", 0x003a9c95 ""),("YAR",0x003a9c96 ""),("YY",0x003a9c94 "")) find returned (<Bad Ptr>,0xcdcdcdcd <Bad Ptr>) when I do a find on YAR. The code is a bit long to put here, but I'm sure this is something you know the answer to. Any ideas? To me, the == operator should return that there is a match. Tag: Easy question on numeric limits Tag: 295498
  - 20
    - VS2005 Profiler generating very huge .vsp file Development Environment : Microsoft Visual Studio 2005, Version 8.0.50727.42, Microsoft Visual C++ OS : Microsoft Windows Server 2003, Service Pack 2 Hello, While profiling a project using Linker /PROFILE option, the .vsp the Report file which gets generated is about 2GB. On 1GB RAM machine the .vsp through VS 2005 IDE would not even open up; even after 2 hours. On 4GB RAM machine it opens after 30 minutes. The report did not specify the % of total time it takes (it specifies the time it take but not percentage time). What could be the reason for such a huge report file getting generated? Is there some way(some setting) to reduce the size of the profiler report so that I am able to use it? I am also keen to get the % of time consumed by the functions called during an action in the applciation, How can I achieve it? Any kind of help or pointers would be highly appreciated. Thanks and Regards Ramesh Singh Tag: Easy question on numeric limits Tag: 295487
  - 21
    - what message is sent by login services when user logs in to system. Hi All, Can you please let me know what ,message is sent by windows logon services when a User logs into the system. I have services running in the system. I need to detect a Logged in user and launch an application into his account. Thanks Anup Tag: Easy question on numeric limits Tag: 295486
  - 22
    - Change BIOS using C/C++? Dear All, Is it possible to change BIOS using a C/C++ programm? For example I must activate Hyperthreading in BIOS for all machine. It would be nice to have a script or program, which will do it automatically. Thanks les Tag: Easy question on numeric limits Tag: 295456
  - 23
    - Passing an array of structures from a pointer [3] Hello, First and foremost, I want to thank everyone that is helping me with my posts! Okay, so I have finally been able to enter some small C programs in VC++ so that I can test them before I actually say that they compile without errors :) The subject of this post has the same name as my two (2) previous posts which indicates that this post is a sequel to the previous posts! On that note, reccomendations have been given to me where I should not define any arrays of structures in my headers and that structures should only be declared in headers. Now ! I am trying to accomplish the seperation of my structure's declaration from its definition of arrays of structures. In other words, I am trying to declare my structure globally (or in a header) and its definition in a function. To keep the program short I did not show a header. Therefore whatever we see as glogal, it is presumed that I would include it in a header. Some questions and notes before we begin: Q1) Is setting a pointer to 0 before using it a proper thing to do? Ex: obj_DDLB = NULL; Q2) Is it proper to declare and define a structure in a header if it is only used in one place in the whole project? This means that this structure would only have one purpose for one particular .c file. Or shopuld I still just declare it in a header and define it in a function of that particular .c file. Q3) What about enums, can they be defined in a header. I do lots of that!!! Ex: enum enum_BUTT_ACTUATE { e_ACT_UD_ARWS=1, e_ACT_EXIT, e_ACT_NEXT, e_ACT_CONFIG}; Q4) This question involves a #define macro. Basically, if these types of pre-processor directives should be in a header *and* if all definitions of structures should be in a .c function, then how is one to find the size of an array structures using: //CALCULATES THE SIZE OF THE DDLB INFORMATIONS-15 #define DDLB_ITEM_LIST1 (long) ((sizeof (COLL1) / sizeof (struct ddlbColl1))) I mean if the above statement is global (in a header) and my array of structures are defined in a function, this will generate an error! I can always put the macro in the particular function where the array of structures is defined.... but would this be proper? Anyways this is not a proiority right now, but is an important consideration for me, because exactly for this problem, I put macros all over the place.... in functions in headers..... not very practical when viewing code I guess! NOTE 1) For now, please disregard the global pointers issue and why they should not be defined globally. I will look at this later. Please consider the following program bywhich compiles and obtains the correct result of 200! This is the original version where I declare and define the structure globally and then we will look at the second code sample which declares the structure in a header and defines it in a function. However the latter has a problem! ============================================ #include <stdlib.h> #include <malloc.h> typedef struct ddlbColl1{ long LINK_ID; int LIST_NUMBER; } COLL1; COLL1 DDLB_COLL1[] = {200, 4, 201, 8}; typedef struct ddListBox{ int x; long y; struct ddlbColl1 *xxx; } DDLB; DDLB *obj_DDLB; // Eventually Won't be global! DDLB* ULC_DDLB_config_ddlb (struct ddlbColl1 DDLB_COLL1[], int x, int y); void ULC_DDLB_ddlb ( DDLB *obj_DDLB); int main(void) { obj_DDLB = ULC_DDLB_config_ddlb (DDLB_COLL1, 10,5); ULC_DDLB_ddlb (obj_DDLB); return 0; } DDLB* ULC_DDLB_config_ddlb ( struct ddlbColl1 DDLB_COLL1[], int x, int y) { obj_DDLB = NULL; obj_DDLB = (DDLB*) malloc (sizeof (struct ddListBox)); obj_DDLB->x = x; obj_DDLB->y = y; obj_DDLB->xxx = DDLB_COLL1; //Assign the address of the DDLB_COLL1[] array! return obj_DDLB; } void ULC_DDLB_ddlb ( DDLB *obj_DDLB) { // implementation file which uses the infos from DDLB_COLL1 array! long x; x = obj_DDLB->xxx[0].LINK_ID; //x = 200!!! Correct result! } ================================================ Now lets consider the following code sample where I delare the structure gloabally (or in my header) and define the array of this structure in a function. This version has the wrong result for x. please view code: ============================================ #include <stdlib.h> #include <malloc.h> typedef struct ddlbColl1{ long LINK_ID; int LIST_NUMBER; } COLL1; typedef struct ddListBox{ int x; long y; struct ddlbColl1 *xxx; } DDLB; DDLB *obj_DDLB; // Eventually Won't be global! DDLB* ULC_DDLB_config_ddlb (int x, int y); void ULC_DDLB_ddlb ( DDLB *obj_DDLB); int main(void) { obj_DDLB = ULC_DDLB_config_ddlb (10,5); ULC_DDLB_ddlb (obj_DDLB); return 0; } DDLB* ULC_DDLB_config_ddlb (int x, int y) { COLL1 DDLB_COLL1[] = { 200, 4, 201, 8}; obj_DDLB = NULL; obj_DDLB = (DDLB*) malloc (sizeof (struct ddListBox)); obj_DDLB->x = x; obj_DDLB->y = y; obj_DDLB->xxx = DDLB_COLL1; //Assign the address of the DDLB_COLL1[] array! return obj_DDLB; } void ULC_DDLB_ddlb ( DDLB *obj_DDLB) { // implementation file which uses the infos from DDLB_COLL1 array! long x; x = obj_DDLB->xxx[0].LINK_ID; // x = -858993460 ????? } ================================================ All help appreciated! -- Best regards Roberto Tag: Easy question on numeric limits Tag: 295447
  - 24
    - Run-Time Check Failure Hello everyone, I am writing a C++ client to dynamically load a DLL and call its function, using LoadLibrary and GetProcAddress. In the C++ client, when executing statement -- "if(!StoreData(GetCurrentThreadId()))", after it returned, there is error message, any ideas? (I showed along with the source codes for C++ client and the DLL itself) [Code] Run-Time Check Failure #0 - The value of ESP was not properly saved across a function call. This is usually a result of calling a function declared with one calling convention with a function pointer declared with a different calling convention. DLL Client code: // The process code #include <windows.h> #include <stdio.h> #define THREADCOUNT 1 #define DLL_NAME TEXT("testdll.dll") VOID ErrorExit(LPSTR); typedef BOOL(*StoreDataFp)(DWORD); typedef BOOL (*GetDataFp)(DWORD*); DWORD WINAPI ThreadFunc(VOID) { int i; HMODULE hMod; hMod = LoadLibrary("testdll.dll"); StoreDataFp StoreData = (StoreDataFp)GetProcAddress(hMod,"StoreData"); GetDataFp GetData = (GetDataFp)GetProcAddress(hMod, "GetData"); if(!StoreData(GetCurrentThreadId())) ErrorExit("StoreData error"); for(i=0; i<5; i++) { DWORD dwOut; if(!GetData(&dwOut)) ErrorExit("GetData error"); if( dwOut != GetCurrentThreadId()) printf("thread %d: data is incorrect (%d)\n", GetCurrentThreadId(), dwOut); else printf("thread %d: data is correct\n", GetCurrentThreadId()); Sleep(0); } return 0; } int main(VOID) { DWORD IDThread; HANDLE hThread[THREADCOUNT]; int i; // Load the DLL LoadLibrary(DLL_NAME); // Create multiple threads. for (i = 0; i < THREADCOUNT; i++) { hThread[i] = CreateThread(NULL, // default security attributes 0, // use default stack size (LPTHREAD_START_ROUTINE) ThreadFunc, // thread function NULL, // no thread function argument 0, // use default creation flags &IDThread); // returns thread identifier // Check the return value for success. if (hThread[i] == NULL) ErrorExit("CreateThread error\n"); } WaitForMultipleObjects(THREADCOUNT, hThread, TRUE, INFINITE); return 0; } VOID ErrorExit (LPSTR lpszMessage) { fprintf(stderr, "%s\n", lpszMessage); ExitProcess(0); } DLL code: // The DLL code #include <windows.h> static DWORD dwTlsIndex; // address of shared memory // DllMain() is the entry-point function for this DLL. BOOL WINAPI DllMain(HINSTANCE hinstDLL, // DLL module handle DWORD fdwReason, // reason called LPVOID lpvReserved) // reserved { LPVOID lpvData; BOOL fIgnore; switch (fdwReason) { // The DLL is loading due to process // initialization or a call to LoadLibrary. case DLL_PROCESS_ATTACH: // Allocate a TLS index. if ((dwTlsIndex = TlsAlloc()) == TLS_OUT_OF_INDEXES) return FALSE; // No break: Initialize the index for first thread. // The attached process creates a new thread. case DLL_THREAD_ATTACH: // Initialize the TLS index for this thread. lpvData = (LPVOID) LocalAlloc(LPTR, 256); if (lpvData != NULL) fIgnore = TlsSetValue(dwTlsIndex, lpvData); break; // The thread of the attached process terminates. case DLL_THREAD_DETACH: // Release the allocated memory for this thread. lpvData = TlsGetValue(dwTlsIndex); if (lpvData != NULL) LocalFree((HLOCAL) lpvData); break; // DLL unload due to process termination or FreeLibrary. case DLL_PROCESS_DETACH: // Release the allocated memory for this thread. lpvData = TlsGetValue(dwTlsIndex); if (lpvData != NULL) LocalFree((HLOCAL) lpvData); // Release the TLS index. TlsFree(dwTlsIndex); break; default: break; } return TRUE; UNREFERENCED_PARAMETER(hinstDLL); UNREFERENCED_PARAMETER(lpvReserved); } extern "C" __declspec(dllexport) BOOL __stdcall WINAPI StoreData(DWORD dw) { LPVOID lpvData; DWORD * pData; // The stored memory pointer // Retrieve a data pointer for the current thread lpvData = TlsGetValue(dwTlsIndex); // If NULL, allocate memory for the TLS slot for this thread if (lpvData == NULL) { lpvData = (LPVOID) LocalAlloc(LPTR, 256); if (lpvData == NULL) return FALSE; if (!TlsSetValue(dwTlsIndex, lpvData)) return FALSE; } pData = (DWORD *) lpvData; // Cast to my data type. // In this example, it is only a pointer to a DWORD // but it can be a structure pointer to contain more complicated data. (*pData) = dw; return TRUE; } extern "C" __declspec(dllexport) BOOL __stdcall WINAPI GetData(DWORD *pdw) { LPVOID lpvData; DWORD * pData; // The stored memory pointer lpvData = TlsGetValue(dwTlsIndex); if (lpvData == NULL) return FALSE; pData = (DWORD *) lpvData; (*pdw) = (*pData); return TRUE; } [/Code] thanks in advance, George Tag: Easy question on numeric limits Tag: 295435
  - 25
    - digital camera Is there a good way to retrieve the images from a digital camera in VC+ +? I'm hoping for a method that will be compatible with pretty much all cameras. Thanks. Tag: Easy question on numeric limits Tag: 295430
- Next
  - 1
    - Segfault on new? Okay, I have a really simple program that illustrates a problem I'm having. I'm using VC++6.0 (yes, an upgrade is in the works). Anywho, I have this problem: int main (int argc, char * argv[]) { int iNumFuncs = 1; int * hey = new int [iNumFuncs]; <<--Segfaults. ?????? return 0; } Is my installation just gone out the window, or am I so incredibly tired that I can't even do a dynamic allocation anymore? Tag: Easy question on numeric limits Tag: 295424
  - 2
    - Passing an array of structures from a pointer [2] Hello, I am trying to test some C code in VC++. Here is the code: ======================================= #include <iostream> #include <stdio.h> #include <string.h> #include <conio.h> #include <stdlib.h> #include <malloc.h> int main(void) { int x; x++; return 0; } ============================================= I am getting 102 errors!!!! The first error is: c:\Program Files\Microsoft Visual Studio .NET 2003\Vc7\include\cerrno(18): error C2143: syntax error : missing '{' before ':' and when I douvbe click on it, it takes me to a line in a file called "cerno" ???? using ::errno; ///<<<<< Its been a while I haven't used VC++! Can someone help me as to why this is happening! Thanks in advance! -- Best regards Roberto Tag: Easy question on numeric limits Tag: 295420
  - 3
    - Forcedly kill console session terminates service Hi All, I have a service that is registered to a windows 2003 server SP2 with option "Allow service to interact with Desktop" enable, and the Startup type is set as Automatic. Two process was spawned by the service. The Service is running under SYSTEM account, One of the spawned process that requires COM interface object is running under an administrator C account. The other one is running under SYSTEM account. What's happening is: - administrator A logs into the w2k3 server using "mstsc / console" the service is working fine and all 3 processes are running. - administrator B logs into the w2k3 server using "mstsc / console", and administrator A's console session is killed forcedly - then COM object is destroyed, the process running in the C account is terminated, as a result the service is terminated. However, if administrator B logs into the w2k3 server after administrator A is successfully log off, the service will be working and none of the related processes will be terminated. Does any one has any idea how killing the console session forcedly by second user impacts the COM object and the service that has option "Allow service to interact with Desktop" enable? Thanks Tag: Easy question on numeric limits Tag: 295418
  - 4
    - How to open 2GB .vsp file quickly? It takes 30 minutes to open on 4GB RAM machine Development Environment Microsoft Visual Studio 2005, Version 8.0.50727.42(RTM.050727-4200),Microsoft Visual C++ 2005 77642-113-3000004-41639 Microsoft .NET Framework Version 2.0.50727 SP2 Microsoft Windows Server 2003 Enterprise Editions, Service Pack 2 Intel® Pentium® D CPU, 2.80GHz, 1 GB/4 GB of RAM While profiling a project using Linker /PROFILE option, the .vsp Report file which gets generated is about 2GB. On 1GB RAM machine the .vsp through VS 2005 IDE would not even open up; even after 2 hours. On 4GB RAM machine it opened after 30 minutes. The report did not specify the % of total time it takes (it specifies the time it take but not percentage time). Is there a way to open up the .vsp quickly and also to display the % time taken by each function. Thanks, Ramesh Kr Singh Tag: Easy question on numeric limits Tag: 295398
  - 5
    - Resource string count How to get number of resource strings in resource file. If any .rc file contains some strings in String Table resource then which function returns the total number of strings and their name,ID. Tag: Easy question on numeric limits Tag: 295394
  - 6
    - Did Microsoft slightly change C++ language in 2008? The textbook for the traditional C++ class that I had in Fall of 2001 at communnity college for the Microsoft Visual C++ 6.0 C++ compiler had examples like cout << endl; that worked with Microsoft Visual C++ 6.0 In 2008 the new Microsoft Visual Studio 2008 Express Edition C++ no longer supports traditional cout by itself. Microsoft might have attempted to change the C++ programming language. On the Microsoft website they had an example similar to std::cout << std::endl Then after changing this old C++ calculator program by using a std:: before every member of the iostream class it started working. This new version of C++ calculator program is called cs085dww.cpp On start menu programs Microsoft Visual Studio 2008 C++ tools then Microsoft Visual Studio command prompt copy a:\cs085dww.cpp c: Fix it with notepad if Internet header and trailer are not remmed out or deleted then cl cs085dww.cpp If it works dir cs085dww.* should show cs085dww.cpp cs085dww.obj cs085dww.exe to run it on Windows use cs085dww.exe after compiling. <iostream.h> was also changed to <iostream> cs085dww.cpp is next */ // beginning of program // cs085dww.cpp version of cs085dwy.cpp #include<iostream> // void mackhw(char* margret,int pegwid,void* peggyv); // char ruth[80]; // char lionel[80]; // int fixit[14]; // -------------------------------------------------- // This cs085dww.cpp is written by Eric Matteson // permission is granted to copy this source code // file cs085dww.cpp and to publish it on the // internet and to use it at least for non profit use. // cs085dww.cpp main is reverse Polish notation calcultor // to test class dewey (128 bit floating point numbers). // To compile and run on Linux // c++ cs085dww.cpp -o cs085dww.out // ./cs085dww.out // --------------------------------------------------- // zackint is class of 256 bit integer arithmetic // in slow motion on a 32-bit CPU class zackint { private: int zisa[8]; public: zackint() { zisa[0]=0; zisa[1]=0; zisa[2]=0; zisa[3]=0; zisa[4]=0; zisa[5]=0; zisa[6]=0; zisa[7]=0; } void zenlarge(int zarger[]); void zeshrink(int zlower[]); zackint zacopyr(); int zaddsub(zackint zatiny,const char* zacksa); int zishifr(const char* zirse); void zacopyo(zackint zcrumb); int zandor(zackint zoright,const char* zaocho); void zacopyt(int zcrumi); // -------------------------- void operator = (zackint); void operator = (int); zackint operator + (zackint); zackint operator - (zackint); zackint operator >> (int); bool operator << (zackint); bool operator < (zackint); bool operator <= (zackint sleright); bool operator > (zackint sgtright); bool operator >= (zackint sgeright); bool operator == (zackint); bool operator != (zackint); zackint operator & (zackint zanright); zackint operator | (zackint oriright); zackint operator * (zackint zmulsra); zackint operator / (zackint zzadenom); void vswap(int zswap[]); void fromfour(int wafour[]); int lengthenexp(int shortdew[]); int isaddsubover(); // ------------------------------------ // end of zackint declarations }; // ------------------------- void zackint::zenlarge(int zarger[]) { int zargsub,zenlctr,zargterml,zargtermh; zargsub=0; zenlctr=0; while(zenlctr < 8) { zargterml=((*this).zisa[zenlctr]) & 65535; zargtermh=(((*this).zisa[zenlctr]) >> 16) & 65535; zarger[zargsub]=zargterml; zarger[zargsub+1]=zargtermh; zargsub += 2; zenlctr ++; } } void zackint::zeshrink(int zlower[]) { int zlowsub,zesctr,zlowl,zlowh; zlowsub=0; zesctr=0; while(zesctr < 8) { zlowh=((zlower[zlowsub+1]) & 65535) << 16; zlowl=(zlower[zlowsub]) & 65535; (*this).zisa[zesctr]=zlowh | zlowl; zlowsub += 2; zesctr ++; } } zackint zackint::zacopyr() { int zcactr,zcaterm; zackint zpyisa; zcactr=0; while(zcactr < 8) { zcaterm=(*this).zisa[zcactr]; zpyisa.zisa[zcactr]=zcaterm; zcactr ++; } return zpyisa; } int zackint::zaddsub(zackint zatiny,const char* zacksa) { int saflags,sacarry,sazctr,sabigrz; // zacksa is s for subtract a for add int saresu[16]; int sabig[16]; int satiny[16]; // returns integer flag code oring 80h sign // 40h zero 01h carry char sazizsa; sazizsa = *(zacksa + 0); saflags = 64 + 4 + 2; sacarry = 0; (*this).zenlarge(sabig); zatiny.zenlarge(satiny); sazctr=0; while(sazctr < 16) { sabigrz=0; if(sazizsa == 's') { sabigrz=sabig[sazctr]-satiny[sazctr]; sabigrz=sabigrz-sacarry; } if(sazizsa == 'a') { sabigrz=sabig[sazctr]+satiny[sazctr]; sabigrz=sabigrz+sacarry; } saresu[sazctr] = sabigrz & 65535; sacarry=0; if((sabigrz & 65536) != 0)sacarry=1; if((saresu[sazctr]) != 0)saflags=6; sazctr ++; } if((sabigrz & 32768) != 0)saflags=saflags+128; if(sacarry != 0)saflags=saflags+1; (*this).zeshrink(saresu); return saflags; } int zackint::zishifr(const char* zisrse) { int shifrctr,shifrcyo,shifrcyi,shifrcyd,shifzz; int destrr[16]; int srcrr[16]; char zlsrse,zrsrse; zlsrse = *(zisrse + 0); zrsrse = *(zisrse + 1); shifzz=70; shifrctr=15; shifrcyo=0; if((*this).zisa[7] < 0) { if((zlsrse == 's')&&(zrsrse == 'e')) { shifrcyo=32768; } } (*this).zenlarge(srcrr); while(shifrctr >= 0) { shifrcyi=shifrcyo; shifrcyd=srcrr[shifrctr]; shifrcyo=0; if((shifrcyd & 1) != 0)shifrcyo=32768; shifrcyd = shifrcyd >> 1; shifrcyd = shifrcyd + shifrcyi; destrr[shifrctr]=shifrcyd; if(shifrcyd != 0)shifzz=6; shifrctr --; } (*this).zeshrink(destrr); return shifzz; } void zackint::zacopyo(zackint zcrumb) { int bcactr,bcaterm; bcactr=0; while(bcactr < 8) { bcaterm = zcrumb.zisa[bcactr]; (*this).zisa[bcactr]=bcaterm; bcactr ++; } } int zackint::zandor(zackint zoright,const char* zaocho) { int ctroa,aoflag,aoterm; int lashenoa[8]; int rashenoa[8]; char baocho; baocho = *(zaocho + 0); ctroa=0; while(ctroa < 8) { lashenoa[ctroa]=(*this).zisa[ctroa]; rashenoa[ctroa]=zoright.zisa[ctroa]; ctroa ++; } ctroa=0; aoflag=70; while(ctroa < 8) { aoterm=0; if(baocho == 'a') { aoterm=lashenoa[ctroa] & rashenoa[ctroa]; } if(baocho == 'o') { aoterm=lashenoa[ctroa] | rashenoa[ctroa]; } (*this).zisa[ctroa]=aoterm; if(aoterm != 0)aoflag=6; ctroa ++; } return aoflag; } void zackint::zacopyt(int zcrumi) { int bcaitr,bcabrut; bcaitr=1; bcabrut=0; if(((*this).zisa[7]) < 0)bcabrut=0-1; while(bcaitr < 8) { (*this).zisa[bcaitr]=bcabrut; bcaitr ++; } (*this).zisa[0]=zcrumi; } void zackint::operator = (zackint zeqright) { int eiqctr; int eiqleft[8]; eiqctr=0; while(eiqctr < 8) { eiqleft[eiqctr]=zeqright.zisa[eiqctr]; eiqctr = eiqctr + 1; } eiqctr=0; while(eiqctr < 8) { (*this).zisa[eiqctr]=eiqleft[eiqctr]; eiqctr = eiqctr + 1; } } void zackint::operator = (int ieqright) { int eiqleft; eiqleft=ieqright; (*this).zacopyt(eiqleft); } zackint zackint::operator + (zackint zadright) { zackint zadleft; zadleft=(*this).zacopyr(); zadleft.zaddsub(zadright,"a"); return zadleft; } zackint zackint::operator - (zackint zsuright) { zackint zsuleft; zsuleft=(*this).zacopyr(); zsuleft.zaddsub(zsuright,"s"); return zsuleft; } zackint zackint::operator >> (int issrmany) { zackint srzleft; int issrctr; issrctr=issrmany; srzleft=(*this).zacopyr(); while(issrctr > 0) { srzleft.zishifr("zl"); issrctr --; } return srzleft; } bool zackint::operator << (zackint mltright) { zackint mltleft; int mltcy; bool mltres; // zackint << zackint is not shift left // << is much less than using borrow instead of sign mltres=false; mltleft=(*this).zacopyr(); mltcy=mltleft.zaddsub(mltright,"s"); if((mltcy & 1) != 0)mltres=true; return mltres; } bool zackint::operator < (zackint sltright) { zackint sltleft; int sltsf; bool sltres; sltres=false; sltleft=(*this).zacopyr(); sltsf=sltleft.zaddsub(sltright,"s"); if((sltsf & 128) != 0)sltres=true; return sltres; } bool zackint::operator <= (zackint sleright) { zackint sleleft; int slesf; bool sleres; sleres=false; sleleft=(*this).zacopyr(); slesf=sleleft.zaddsub(sleright,"s"); if((slesf & 128) != 0)sleres=true; if((slesf & 64) != 0)sleres=true; return sleres; } bool zackint::operator > (zackint sgtright) { zackint sgtleft; int sgtsf; bool sgtres; sgtres=false; sgtleft=(*this).zacopyr(); sgtsf=sgtleft.zaddsub(sgtright,"s"); if(((sgtsf & 128)==0)&&((sgtsf & 64)==0))sgtres=true; return sgtres; } bool zackint::operator >= (zackint sgeright) { zackint sgeleft; int sgesf; bool sgeres; sgeres=false; sgeleft=(*this).zacopyr(); sgesf=sgeleft.zaddsub(sgeright,"s"); if((sgesf & 128) == 0)sgeres=true; return sgeres; } bool zackint::operator == (zackint yeqright) { zackint yeqleft; int yeqzf; bool yeqres; yeqres=false; yeqleft=(*this).zacopyr(); yeqzf=yeqleft.zaddsub(yeqright,"s"); if((yeqzf & 64) != 0)yeqres=true; return yeqres; } bool zackint::operator != (zackint zneright) { zackint zneleft; int znezf; bool zneres; zneres=false; zneleft=(*this).zacopyr(); znezf=zneleft.zaddsub(zneright,"s"); if((znezf & 64) == 0)zneres=true; return zneres; } zackint zackint::operator & (zackint zanright) { zackint zandleft; zandleft=(*this).zacopyr(); zandleft.zandor(zanright,"a"); return zandleft; } zackint zackint::operator | (zackint oriright) { zackint orileft; orileft=(*this).zacopyr(); orileft.zandor(oriright,"o"); return orileft; } zackint zackint::operator * (zackint zmulsra) { zackint slfacaz,srfacaz,zaprod,zampyone; zackint zmpyzero,azmpyres; int mpyictr; mpyictr=6; slfacaz=(*this).zacopyr(); srfacaz=zmulsra.zacopyr(); zmpyzero=0; zampyone=1; zaprod=zmpyzero; while((mpyictr & 64) == 0) { azmpyres = srfacaz & zampyone; if(azmpyres != zmpyzero) { zaprod=zaprod+slfacaz; } slfacaz=slfacaz+slfacaz; srfacaz = srfacaz >> 1; if(srfacaz == zmpyzero)mpyictr=70; } return zaprod; } zackint zackint::operator / (zackint zzadenom) { int zdivctr,znegctr,dshlcy; zackint donstzero,donstone,divsubtrest; zackint numerdd,denomdd,bignumber; zackint zfrac; zdivctr=0; znegctr=0; numerdd=(*this).zacopyr(); denomdd=zzadenom.zacopyr(); donstzero=0; donstone=1; bignumber=donstzero; zfrac=donstzero; if(numerdd < donstzero) { numerdd=donstzero-numerdd; znegctr ++; } if(zzadenom < donstzero) { denomdd=donstzero-zzadenom; znegctr ++; } while(zdivctr < 256) { bignumber=bignumber+bignumber; zfrac=zfrac+zfrac; if(numerdd < donstzero) { bignumber=bignumber+donstone; } numerdd=numerdd+numerdd; dshlcy=0; divsubtrest=bignumber-denomdd; if(bignumber << denomdd)dshlcy=1; if((dshlcy & 1) == 0) { bignumber=divsubtrest; zfrac=zfrac+donstone; } zdivctr ++; } if((znegctr & 1) != 0) { zfrac=donstzero-zfrac; } return zfrac; } // bottom of operator / void zackint::vswap(int zswap[]) { int vswctr; int vswtemp[8]; vswctr=0; while(vswctr < 8) { vswtemp[vswctr]=(*this).zisa[vswctr]; vswctr ++; } vswctr=0; while(vswctr < 8) { (*this).zisa[vswctr]=zswap[vswctr]; vswctr ++; } vswctr=0; while(vswctr < 8) { zswap[vswctr]=vswtemp[vswctr]; vswctr ++; } } void zackint::fromfour(int wafour[]) { int targfrvs[8]; int frfvsctr,frfvsone; frfvsctr=4; frfvsone=0; if((wafour[3])<0)frfvsone=0-1; while(frfvsctr < 8) { targfrvs[frfvsctr]=frfvsone; frfvsctr ++; } frfvsctr=0; while(frfvsctr < 4) { targfrvs[frfvsctr]=wafour[frfvsctr]; frfvsctr ++; } (*this).vswap(targfrvs); } int zackint::lengthenexp(int shortdew[]) { int targlexp[8]; zackint longexpr; int lexprez,lantpar,lnextexp,ltwenty,lrextexp; int lnonzero; longexpr.fromfour(shortdew); longexpr.vswap(targlexp); lnonzero=0; lexprez=0; while(lexprez < 8) { if(targlexp[lexprez] != 0)lnonzero=1; lexprez=lexprez+1; } lrextexp=targlexp[3]; ltwenty=1024*1024; lexprez=ltwenty-1; lantpar=lexprez & lrextexp; lexprez=(lrextexp >> 20) & 2047; if(lrextexp < 0)lexprez=(2048-lexprez) & 2047; lnextexp=ltwenty; if(lrextexp < 0)lnextexp=0-(ltwenty + ltwenty); if(lnonzero == 0)lnextexp=0; if(lnonzero == 0)lexprez=0; targlexp[3]=lnextexp+lantpar; (*this).vswap(targlexp); return lexprez; } // 116 96+20 opposite of sign bit 0 o.k. // 117 96+21 simple overflow int zackint::isaddsubover() { zackint iaozb; int iaxzb[8]; int isaddoverind,addovermaskl,adoleft,adoright; int adofarright,addovermaskr,addoverterm; iaozb=(*this).zacopyr(); iaozb.vswap(iaxzb); addoverterm=iaxzb[3]; isaddoverind=0; // returns 0 for normal long value addovermaskl=1024*2048; addovermaskr=1024*1024; adoleft=0; adoright=0; adofarright=0; if(addoverterm < 0)adoleft=1; // adoleft if negative number if((addoverterm & addovermaskl) != 0)adoright=1; // adoright if overflow bit 21 set if((addoverterm & addovermaskr) != 0)adofarright=1; // adofarright should be opposite of sign if(adoleft != adoright)isaddoverind=1; if((adoleft == adoright)&&(adofarright == adoright)) { isaddoverind=0-1; // underflow result too small magnitude } return isaddoverind; } // ---------------------------------------------------- // ---------------------------------------------------- // zackint 8 32 bit integers // 16 16 bit half integers // and 65536 carrys // and 32768 leftmost sign // and 65535 half integers values // ------ // actual end of class zackint // xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx // class dewey list mantissia without leftmost is // mantissia 20+96 -> 116 bits 115-0 // exponent in short form 126-116 // ---- // lengthened form for floating point // 10-0 biased seperate exponent. is negative if sign bit. // 255-128 high sign // 31 127 regular sign bit // 31-22 127-118 low sign // 21 117 overflow bit for addition same as sign // 20 116 opposite of sign. implied leftmost one bit. // 19-0 115-0 mantissia // class dewey calls class zackint class dewey { private: int hmgreg[4]; public: dewey() { hmgreg[0]=0; hmgreg[1]=0; hmgreg[2]=0; hmgreg[3]=0; } void wswap(int wxswap[]); void dewcopyo(dewey dcrumb); dewey dewcopyr(); void shortenf(zackint largedz,int expdf); dewey fpaddsub(dewey fpstiny,char* lp4345); dewey fpmpysub(dewey dsrfac); dewey fpdiv(dewey bdenom); void dfromint(int src20); int dewcmp(dewey dcmprr); void operator = (dewey deqrighh); bool operator < (dewey dewLTrr); bool operator > (dewey dewGTrr); bool operator >= (dewey dewGErr); bool operator <= (dewey dewLErr); bool operator != (dewey dewNErr); bool operator == (dewey dewEQrr); dewey operator + (dewey dewPLrr); dewey operator - (dewey dewMIrr); dewey operator * (dewey dewPRrr); dewey operator / (dewey dewDLrr); void ctyping(int ctydpoz,const char* ctydchr); int fcpread(int frpoz,int frwid,int frbase,char* frydc); int fcsimpwr(int fwpoz,int fwwid,int fwbase,char* fwrdc); int dprecise(int dprebase,int dprelimit); int ifromdub(); int decompwr(int dwpoz,int dwmax,int dwbase, char* dwyyc); void shove(dewey* dove,int jove); void unshove(dewey* undove,int uove); // -- // -- // -- dewey // end of dewey declarations 592 }; void dewey::wswap(int wxswap[]) { int wdswap[4]; int wdswctr; wdswctr=0; while(wdswctr < 4) { wdswap[wdswctr]=wxswap[wdswctr]; wdswctr ++; } wdswctr=0; while(wdswctr < 4) { wxswap[wdswctr]=(*this).hmgreg[wdswctr]; wdswctr ++; } wdswctr=0; while(wdswctr < 4) { (*this).hmgreg[wdswctr]=wdswap[wdswctr]; wdswctr ++; } } void dewey::dewcopyo(dewey dcrumb) { int dcrrctr; dcrrctr=0; while(dcrrctr < 4) { (*this).hmgreg[dcrrctr]=dcrumb.hmgreg[dcrrctr]; dcrrctr ++; } } dewey dewey::dewcopyr() { dewey dddcres; int dwrrctr; dwrrctr=0; while(dwrrctr < 4) { dddcres.hmgreg[dwrrctr]=(*this).hmgreg[dwrrctr]; dwrrctr ++; } return dddcres; } void dewey::shortenf(zackint largedz,int expdf) { zackint largetz; int largedf[8]; int shortarg[4]; int cinda,ciexp,ciparj; int cidfmask,citemp; cinda=16384; cidfmask=(cinda * 64)-1; cinda=(cinda << 17); largetz=largedz; largetz.vswap(largedf); if(*(largedf + 7) >= 0)cinda=0; ciexp=expdf & 2047; if(expdf > 2047)ciexp=2040; if(*(largedf + 7)< 0)ciexp=2048-ciexp; ciexp=((ciexp & 2047) << 20); citemp = *(largedf + 3); ciparj=citemp & cidfmask; citemp=cinda+ciexp; citemp=citemp+ciparj; *(shortarg + 3) = citemp; *(shortarg + 2) = *(largedf + 2); *(shortarg + 1) = *(largedf + 1); *(shortarg + 0) = *(largedf + 0); if((ciexp == 0)||(expdf < 2)) { *(shortarg + 3)=0; *(shortarg + 2)=0; *(shortarg + 1)=0; *(shortarg + 0)=0; } (*this).wswap(shortarg); } dewey dewey::fpaddsub(dewey fpstiny,char* lp4345) { zackint bpasres; zackint bpabig; zackint bpstiny; zackint bpszero; dewey fbcbig; dewey fpasres; int fpabig[4]; int fhstiny[4]; int expleft,expright,lrsel,expres; int saflags,faslimctr; char cp4345; fbcbig=(*this).dewcopyr(); fbcbig.wswap(fpabig); expleft=bpabig.lengthenexp(fpabig); fbcbig=fpstiny.dewcopyr(); fbcbig.wswap(fhstiny); expright=bpstiny.lengthenexp(fhstiny); cp4345 = *(lp4345 + 0); // shift smallest exponented number right // increasing exponent expres=expleft; lrsel=2; faslimctr=0; if(expright > expres) { expres=expright; lrsel=3; } if((lrsel == 2)&&(expright > 0)) { while((expright < expres)&&(faslimctr < 4096)) { saflags=bpstiny.zishifr("se"); expright=expright+1; faslimctr=faslimctr + 1; } } if((lrsel == 3)&&(expleft > 0)) { // middle of fpaddsub while((expleft < expres)&&(faslimctr < 4096)) { saflags=bpabig.zishifr("se"); expleft=expleft+1; faslimctr = faslimctr + 1; } } if(cp4345 == '+') { bpasres = bpabig + bpstiny; } if(cp4345 == '-') { bpasres = bpabig - bpstiny; } // result bpasres too big shift right once // incrementing exponent once // ----------------------------------------- // ---------------------------------------- // bpasres too small magnitude shift left // many times decreasing exponent. // ------------------------------------------- // ----------------------------------------- // if biasedexponent < 1 mantissia=0 exponent=0 bpszero=0; if(bpasres == bpszero)expres=0; faslimctr=0; if((bpasres.isaddsubover() > 0)&&(expres > 1)) { saflags=bpasres.zishifr("se"); if(expres < 2047)expres=expres+1; } if((bpasres.isaddsubover() < 0)&&(expres > 1)) { while((bpasres.isaddsubover() < 0)&&(expres > 1)) { bpasres=bpasres+bpasres; expres=expres-1; faslimctr=faslimctr+1; if(faslimctr > 511)expres=0; } } if(expres < 2) { bpasres=0; expres=0; } fpasres.shortenf(bpasres,expres); return fpasres; } // bottom of fpaddsub dewey dewey::fpmpysub(dewey dsrfac) { zackint bpaprod; zackint tpaprod; zackint bslfac; zackint bsrfac; dewey bscflc; dewey fpadrod; int aslfac[4]; int ipaprod[8]; int dshfac[4]; int bslexp,bsrexp,mpyexp,w7cpy,fpmpyctr; int sc1024,ss512,is1024,is512,fpmdis; int is256,dfpmctr; bscflc = (*this).dewcopyr(); bscflc.wswap(aslfac); bslexp=bslfac.lengthenexp(aslfac); bscflc=dsrfac.dewcopyr(); bscflc.wswap(dshfac); bsrexp=bsrfac.lengthenexp(dshfac); mpyexp=(bslexp + bsrexp)-(1024 + 1); if(bslexp < 2)mpyexp=0; if(bsrexp < 2)mpyexp=0; if(mpyexp < 2)mpyexp=0; // word 7 of bpaprod ands // 255 8 bits data // 256 1 bit implied one not equal to sign // 512 1 bit unstable should be same as sign // 1024 1 bit fixed sign check bpaprod = bslfac * bsrfac; tpaprod=bpaprod; tpaprod.vswap(ipaprod); w7cpy=ipaprod[7]; is1024=0; is512=0; is256=0; // fpmpysub middle if((w7cpy & 1024) > 0)is1024=1; if((w7cpy & 512) > 0)is512=1; if((w7cpy & 256) > 0)is256=1; if(is1024 != is512) { fpmdis=bpaprod.zishifr("se"); mpyexp=mpyexp+1; } dfpmctr=0; while(dfpmctr < 2) { tpaprod=bpaprod; tpaprod.vswap(ipaprod); w7cpy=ipaprod[7]; is1024=0; is512=0; is256=0; if((w7cpy & 1024) > 0)is1024=1; if((w7cpy & 512) > 0)is512=1; if((w7cpy & 256) > 0)is256=1; if(is1024 == is256) { bpaprod=bpaprod+bpaprod; mpyexp=mpyexp-1; } dfpmctr ++; } fpmpyctr=0; while(fpmpyctr < 12) { bpaprod=bpaprod+bpaprod; fpmpyctr=fpmpyctr+1; } fpmpyctr=0; bpaprod.vswap(ipaprod); dfpmctr=0; if((ipaprod[7])<0)dfpmctr=0-1; while(fpmpyctr < 4) { ipaprod[fpmpyctr]=ipaprod[fpmpyctr + 4]; ipaprod[fpmpyctr + 4]=dfpmctr; fpmpyctr=fpmpyctr + 1; } bpaprod.vswap(ipaprod); fpadrod.shortenf(bpaprod,mpyexp); return fpadrod; } // end of fpmpysub dewey dewey::fpdiv(dewey bdenom) { zackint bfrac,bnumer,cdenom,hrumer; dewey bdvcare,fdvrez; int cnumer[4]; int sdenom[4]; int irumer[8]; int fracexp,numerexp,denomexp,fdivctr; int divignore,dqover; bdvcare=(*this).dewcopyr(); bdvcare.wswap(cnumer); numerexp=bnumer.lengthenexp(cnumer); bdvcare=bdenom.dewcopyr(); bdvcare.wswap(sdenom); denomexp=cdenom.lengthenexp(sdenom); fracexp=(numerexp+1024+1)-denomexp; if(numerexp < 2)fracexp=0; bnumer.vswap(irumer); fdivctr=3; while(fdivctr >= 0) { irumer[fdivctr + 4]=irumer[fdivctr]; irumer[fdivctr]=0; fdivctr=fdivctr-1; } bnumer.vswap(irumer); fdivctr=0; while(fdivctr < 12) { divignore=bnumer.zishifr("se"); fdivctr=fdivctr+1; } bfrac = bnumer / cdenom; dqover=bfrac.isaddsubover(); if(dqover > 0) { divignore=bfrac.zishifr("se"); fracexp=fracexp+1; } if(dqover < 0) { bfrac=bfrac+bfrac; fracexp=fracexp-1; } fdvrez.shortenf(bfrac,fracexp); return fdvrez; } // bottom of fpdiv void dewey::dfromint(int src20) { // dfromint is designed to convert small positive // integers to floating point numbers. // preferably one digit. int cdieight[8]; int mchaleneg,mchalesrc; int cdiexp,cdiror,cdicount,cdiprobe; int cdjcount,cdjtemp,cdjallow; dewey dfiirez; dewey mchalezer; zackint dfjrez; cdiexp=0; while(cdiexp < 4) { cdieight[cdiexp]=0; cdiexp=cdiexp+1; } mchalezer.wswap(cdieight); cdiexp=0; while(cdiexp < 8) { cdieight[cdiexp]=0; cdiexp=cdiexp+1; } cdiexp=0; mchalesrc=src20; mchaleneg=0; if(src20 < 0) { mchalesrc=0-src20; mchaleneg=1; } if(mchalesrc > 0) { cdiror=1; cdicount=0; cdiprobe=mchalesrc; cdjallow=1; while(cdjallow > 0) { cdjtemp=cdiprobe >> 1; cdiprobe=cdjtemp; cdicount=cdicount+1; if(cdiprobe < 1)cdjallow=0; } cdjcount=20-cdicount; cdiprobe=mchalesrc; cdjallow=1; while(cdjallow > 0) { cdjtemp=cdiprobe+cdiprobe; cdiprobe=cdjtemp; cdjcount=cdjcount-1; if(cdjcount < 0)cdjallow=0; } cdieight[3]=cdiprobe; cdiexp=(1024+cdicount)+0; } dfjrez.vswap(cdieight); dfiirez.shortenf(dfjrez,cdiexp); if(mchaleneg != 0) { dfiirez = mchalezer.fpaddsub(dfiirez,"-"); } (*this).dewcopyo(dfiirez); } // end of dfromint int dewey::dewcmp(dewey dcmprr) { zackint zcmpt; zackint rcmpz; dewey dcmpt; dewey dcmpl; int rewcmp[4]; int cewcmp; cewcmp=0; dcmpl=(*this).dewcopyr(); dcmpt=dcmpl.fpaddsub(dcmprr,"-"); dcmpt.wswap(rewcmp); zcmpt.fromfour(rewcmp); rcmpz=0; if(zcmpt < rcmpz)cewcmp=0-1; if(zcmpt > rcmpz)cewcmp=1; return cewcmp; } void dewey::operator = (dewey deqrighh) { int ddehleft[4]; int ddehctr; ddehctr=0; while(ddehctr < 4) { ddehleft[ddehctr]=deqrighh.hmgreg[ddehctr]; ddehctr = ddehctr + 1; } ddehctr=0; while(ddehctr < 4) { (*this).hmgreg[ddehctr]=ddehleft[ddehctr]; ddehctr = ddehctr + 1; } } bool dewey::operator < (dewey dewLTrr) { dewey dewLTlf; int dewLTce; bool dewLTrz; dewLTlf=(*this).dewcopyr(); dewLTrz=false; dewLTce=dewLTlf.dewcmp(dewLTrr); if(dewLTce < 0)dewLTrz=true; return dewLTrz; } bool dewey::operator > (dewey dewGTrr) { dewey dewGTlf; int dewGTce; bool dewGTrz; dewGTlf=(*this).dewcopyr(); dewGTrz=false; dewGTce=dewGTlf.dewcmp(dewGTrr); if(dewGTce > 0)dewGTrz=true; return dewGTrz; } bool dewey::operator >= (dewey dewGErr) { dewey dewGElf; int dewGEce; bool dewGErz; dewGElf=(*this).dewcopyr(); dewGErz=false; dewGEce=dewGElf.dewcmp(dewGErr); if(dewGEce > 0)dewGErz=true; if(dewGEce == 0)dewGErz=true; return dewGErz; } bool dewey::operator <= (dewey dewLErr) { dewey dewLElf; int dewLEce; bool dewLErz; dewLElf=(*this).dewcopyr(); dewLErz=false; dewLEce=dewLElf.dewcmp(dewLErr); if(dewLEce < 0)dewLErz=true; if(dewLEce == 0)dewLErz=true; return dewLErz; } bool dewey::operator != (dewey dewNErr) { dewey dewNElf; int dewNEce; bool dewNErz; dewNElf=(*this).dewcopyr(); dewNErz=false; dewNEce=dewNElf.dewcmp(dewNErr); if(dewNEce > 0)dewNErz=true; if(dewNEce < 0)dewNErz=true; return dewNErz; } bool dewey::operator == (dewey dewEQrr) { dewey dewEQlf; int dewEQce; bool dewEQrz; dewEQlf=(*this).dewcopyr(); dewEQrz=true; dewEQce=dewEQlf.dewcmp(dewEQrr); if(dewEQce > 0)dewEQrz=false; if(dewEQce < 0)dewEQrz=false; return dewEQrz; } dewey dewey::operator + (dewey dewPLrr) { dewey dewPLlf; dewey dewPLrz; dewPLlf=(*this).dewcopyr(); dewPLrz=dewPLlf.fpaddsub(dewPLrr,"+"); return dewPLrz; } // -------------------------------------------- dewey dewey::operator - (dewey dewMIrr) { dewey dewMIlf; dewey dewMIrz; dewMIlf=(*this).dewcopyr(); dewMIrz=dewMIlf.fpaddsub(dewMIrr,"-"); return dewMIrz; } dewey dewey::operator * (dewey dewPRrr) { dewey dewPRlf; dewey dewPRrz; dewPRlf=(*this).dewcopyr(); dewPRrz=dewPRlf.fpmpysub(dewPRrr); return dewPRrz; } dewey dewey::operator / (dewey dewDLrr) { dewey dewDLlf; dewey dewDLrz; dewDLlf=(*this).dewcopyr(); dewDLrz=dewDLlf.fpdiv(dewDLrr); return dewDLrz; } void dewey::ctyping(int ctydpoz,const char* ctydchr) { dewey ctypeez; int itypete,itypezz; char itypess; itypess = *(ctydchr + ctydpoz); itypete = (int)itypess; if(itypete < 0)itypete=itypete+256; itypezz=47; if((itypete >= 48)&&(itypete <= 57))itypezz=itypete-48; if((itypete >= 65)&&(itypete <= 70))itypezz=itypete-55; if((itypete >= 97)&&(itypete <=102))itypezz=itypete-87; if((itypete >= 43)&&(itypete <= 46))itypezz=itypete; // 43 + 44 , 45 - 46 . ctypeez.dfromint(itypezz); (*this).dewcopyo(ctypeez); } int dewey::fcpread(int frpoz,int frwid,int frbase,char* frydc) { dewey fcprez,fcpt,fcpri,fcprase,fcrudase,fcrcst01; dewey fcrcst45,fcrcst46,fcrcst47,fcrcst00; dewey dcprase,fcyurm; int fcprctr,fcprsub,fcprlen,fcprllow,fcprlate; int frnewpoz,fcrneg; fcrneg=0; fcprsub=0; fcrcst00.dfromint(fcprsub); fcprsub=1; fcrcst01.dfromint(fcprsub); fcprsub=45; fcrcst45.dfromint(fcprsub); fcprsub=46; fcrcst46.dfromint(fcprsub); fcprsub=47; fcrcst47.dfromint(fcprsub); fcprlen=1; fcprllow=1; fcprlate=1; fcprase.dfromint(frbase); fcrudase = fcrcst01 / fcprase; fcprctr=0; fcprez=fcrcst00; while(fcprllow > 0) { fcprsub=fcprctr+(frpoz-1); fcpt.ctyping(fcprsub,frydc); if(fcpt < fcrcst47)fcprllow=0; if(fcprllow > 0) { fcprctr ++; if(fcprctr >= frwid) { fcprllow=0; fcprlen=0; fcprlate=0; } } } while((fcprlen > 0)&&(fcprlate > 0)) { fcprsub=fcprctr+(frpoz-1); fcpt.ctyping(fcprsub,frydc); if(fcpt >= fcrcst46)fcprlen=0; if(fcpt > fcrcst46)fcprlate=0; if(fcpt == fcrcst46)fcprctr ++; if(fcpt == fcrcst45)fcrneg=1; if(fcprlen > 0) { if(fcpt < fcprase) { fcprez = fcprez * fcprase; fcprez = fcprez + fcpt; } fcprctr ++; if(fcprctr >= frwid)fcprlen=0; } } if(fcprctr >= frwid)fcprlate=0; dcprase=fcrudase; while(fcprlate > 0) { fcprsub=fcprctr+(frpoz-1); fcpt.ctyping(fcprsub,frydc); if(fcpt >= fcprase)fcprlate=0; if(fcprlate > 0) { fcyurm = dcprase * fcpt; fcprez = fcprez + fcyurm; dcprase = dcprase * fcrudase; fcprctr ++; if(fcprctr >= frwid)fcprlate=0; } } if(fcrneg > 0) { fcyurm = fcrcst00 - fcprez; fcprez = fcyurm; } // ---- (*this).dewcopyo(fcprez); frnewpoz=fcprctr+1; return frnewpoz; } // bottom of fcpread int dewey::fcsimpwr(int fwpoz,int fwwid,int fwbase,char* fwrdc) { dewey wrthiz,dwrbase,wrudbase; dewey wrcst00,wrcst01; dewey wrilop,wriprod,writry; int dwrexp,dwrsub,dwrctr,dwrsgn; int dwrilop,dwriver,dwriallow; // -- dwrbase.dfromint(fwbase); dwrsgn=43; dwrsub=0; wrcst00.dfromint(dwrsub); dwrsub=1; wrcst01.dfromint(dwrsub); wrthiz=(*this).dewcopyr(); if(wrthiz < wrcst00) { wrthiz=wrcst00-wrthiz; dwrsgn=45; } dwrexp=0; wrudbase=wrcst01 / dwrbase; if((wrthiz < wrudbase)&&(wrthiz > wrcst00)) { while((wrthiz < wrudbase)&&(dwrexp > (1-2048))) { wrthiz=wrthiz*dwrbase; dwrexp=dwrexp-1; } } // middle of fcsimpwr if(wrthiz >= wrcst01) { while((wrthiz >= wrcst01)&&(dwrexp < 2048)) { wrthiz=wrthiz/dwrbase; dwrexp=dwrexp+1; } } // cut here to partially round number up if(wrthiz > wrcst00) { dwrctr=0-1; writry=wrudbase; while(dwrctr < fwwid) { writry=writry*wrudbase; dwrctr=dwrctr+1; } wrthiz=wrthiz+writry; } // end of partial rounding up dwrctr=0; if(dwrctr < (fwwid+1)) { dwrsub=dwrctr+(fwpoz-1); *(fwrdc + dwrsub)=(char)dwrsgn; *(fwrdc +(dwrsub+1))=(char)46; } dwrctr=2; while(dwrctr < (fwwid+1)) { dwrilop=0; dwriver=0; dwriallow=1; while(dwriallow > 0) { wrilop.dfromint(dwrilop); wriprod=wrilop*wrudbase; writry=wrthiz-wriprod; if(writry >= wrcst00)dwriver=dwrilop; if(writry < wrcst00)dwriallow=0; if(dwriallow > 0) { dwrilop=dwrilop+1; if(dwrilop >= fwbase)dwriallow=0; } } wrilop.dfromint(dwriver); wriprod=wrilop*wrudbase; writry=wrthiz-wriprod; wrthiz=writry*dwrbase; dwriallow=dwriver+48; if(dwriver > 9)dwriallow=dwriver+87; dwrsub=dwrctr+(fwpoz-1); *(fwrdc+dwrsub)=(char)dwriallow; dwrctr=dwrctr+1; } dwrsub=fwwid+(fwpoz-1)+1; // change to dwrsub=dwrexp; return dwrsub; } // end of fcsimpwr int dewey::dprecise(int dprebase,int dprelimit) { dewey lprecise,rprecise,tprecise; dewey tprebase,udprebase,dpreone; dewey dprezero; int dpreactr,dpreallow; dpreactr=0; dprezero.dfromint(dpreactr); lprecise=(*this).dewcopyr(); tprebase.dfromint(dprebase); dpreactr=1; dpreone.dfromint(dpreactr); udprebase=dpreone/tprebase; dpreactr=0; dpreallow=1; if(lprecise < dprezero)lprecise=dprezero-lprecise; rprecise=lprecise; while(dpreallow > 0) { tprecise=rprecise+lprecise; if(tprecise == lprecise)dpreallow=0; if(dpreallow > 0) { dpreactr=dpreactr+1; rprecise=rprecise*udprebase; if(dpreactr >= dprelimit)dpreallow=0; } } return dpreactr; } int dewey::ifromdub() { dewey ifdleft,ifdzero,ifdone,ifdtwo,ifdhalf; int ifdres,ifdctr,ifdneg,ifdexp; ifdleft=(*this).dewcopyr(); ifdctr=1; ifdone.dfromint(ifdctr); ifdctr=2; ifdtwo.dfromint(ifdctr); ifdctr=0; ifdzero.dfromint(ifdctr); ifdres=0; ifdneg=0; if(ifdleft < ifdzero) { ifdleft = ifdzero - ifdleft; ifdneg = 1; } ifdhalf = ifdone / ifdtwo; if(ifdleft >= ifdone) { ifdexp=0; while(ifdleft >= ifdone) { ifdleft=ifdleft*ifdhalf; ifdexp=ifdexp+1; } ifdctr=0; while(ifdctr < ifdexp) { ifdres=ifdres+ifdres; if(ifdleft >= ifdhalf) { ifdres=ifdres+1; ifdleft=ifdleft-ifdhalf; } ifdleft=ifdleft+ifdleft; ifdctr=ifdctr+1; } } if(ifdneg != 0) { ifdres=0-ifdres; } return ifdres; } // bottom of ifromdub int dewey::decompwr(int dwpoz,int dwmax,int dwbase, char* dwyyc) { dewey womplef,wompexp,wompbase,womptra; int ddwctr,ddwexp,ddiscard; int ddleftwid,adwpoz; ddwctr=1; womptra.dfromint(ddwctr); womplef = (*this).dewcopyr(); ddleftwid = womptra.dprecise(dwbase,(dwmax-14))-2; adwpoz=dwpoz; ddwexp=womplef.fcsimpwr(adwpoz,ddleftwid,dwbase,dwyyc); ddwctr=0-2; ddiscard=ddwctr+ddleftwid+adwpoz; if(ddiscard > 7) { *(dwyyc + ddiscard) = 'E'; } wompexp.dfromint(ddwexp); wompbase.dfromint(dwbase); ddwctr=1; womptra.dfromint(ddwctr); while(ddwctr < 6) { womptra = womptra * wompbase; ddwctr = ddwctr + 1; } if(ddwexp >= 0)wompexp = wompexp + womptra; if(ddwexp < 0)wompexp = wompexp - womptra; ddiscard = wompexp.fcsimpwr(adwpoz+ddleftwid,7, dwbase,dwyyc); ddwctr=0; while(ddwctr < 8) { ddiscard=ddwctr+ddleftwid+adwpoz; *(dwyyc + ddiscard) = *(dwyyc + (ddiscard + 3)); ddwctr=ddwctr+1; } ddwctr=0; ddiscard=ddwctr+ddleftwid+adwpoz-1; *(dwyyc + (ddiscard + 5)) = '\40'; *(dwyyc + (ddiscard + 6)) = '\40'; *(dwyyc + (ddiscard + 7)) = '\40'; *(dwyyc + (ddiscard + 8)) = '\40'; *(dwyyc + (ddiscard + 9)) = '\40'; *(dwyyc + (ddiscard + 10)) = '\40'; *(dwyyc + (ddiscard + 11)) = '\0'; ddwctr=ddiscard+10; return ddwctr; } void dewey::shove(dewey* dove,int jove) { dewey shovel; int jovectr; shovel=(*this).dewcopyr(); jovectr=jove-1; while(jovectr >= 0) { *(dove + (jovectr + 1)) = *(dove + jovectr); jovectr = jovectr -1; } jovectr=0; *(dove + jovectr) = shovel; } void dewey::unshove(dewey* undove,int uove) { dewey unovel; int uovectr; uovectr=0; unovel = *(undove + uovectr); while(uovectr < (uove + 1)) { *(undove + uovectr) = *(undove + (uovectr + 1)); uovectr = uovectr + 1; } (*this).dewcopyo(unovel); } // end of class dewey // end of class dewey // test class dewey // begin diagnostic section // void mackhw(char* margret,int pegwid,void* peggyv) // { // char* peggyc; // int mackhctr,mackwsub,mare,mackictr; // int mackrem,mackout; // peggyc = reinterpret_cast<char*>(peggyv); // mackhctr=0; // mackwsub=0; // while(mackhctr < pegwid) // { // mare = (int)(*(peggyc + mackhctr)); // if(mare < 0)mare=mare+256; // mackictr=1; // while(mackictr >= 0) // { // mackrem = mare & 15; // mackout = mackrem + 48; // if(mackrem > 9)mackout = mackrem + 55; // *(margret + (mackwsub + mackictr)) = (char)mackout; // mare = mare >> 4; // mackictr = mackictr - 1; // } // mackwsub = mackwsub + 2; // mackhctr = mackhctr + 1; // } // *(margret + mackwsub) = '\0'; // } // end of mackhw. mackhw was previously used for debugging. int main() { char lionel[80]; // char ruth[80]; // zackint arnold; dewey zove[20]; dewey dele,deri,alge,dezero; int otallow,dtallow,otchk,zize; otchk=0; dezero.dfromint(otchk); zize=20; otallow=1; while(otallow > 0) { dtallow=0; while(dtallow < 80) { lionel[dtallow]=0; dtallow = dtallow + 1; } dtallow=1; std::cout << "Test Reverse polish calulator" << std::endl; std::cout << "for class dewey. number enter" << std::endl; std::cout << "number enter operation enter" << std::endl; std::cout << " + - * / q quit" << std::endl; std::cout << " & change sign " << std::endl; std::cout << "this cs085dww.cpp was written by Eric Matteson"; std::cout << std::endl; while(dtallow > 0) { std::cout << "enter number or operation" << std::endl; std::cin.get(lionel,80); std::cin.ignore(80,'\n'); otchk = (int)lionel[0]; if((otchk < 48)||(otchk > 57))dtallow=0; if((otchk >= 46)&&(otchk <= 46))dtallow=1; if(dtallow > 0) { otchk=alge.fcpread(1,70,10,lionel); alge.shove(zove,zize); } } otchk=(int)lionel[0]; if(otchk == (int)'q')otallow=0; if(otallow > 0) { // check operation type if(otchk == (int)'+') { deri.unshove(zove,zize); dele.unshove(zove,zize); alge = dele + deri; alge.shove(zove,zize); } // right // check operation type if(otchk == (int)'-') { deri.unshove(zove,zize); dele.unshove(zove,zize); alge = dele - deri; alge.shove(zove,zize); } // right if(otchk == (int)'&') { dele.unshove(zove,zize); alge = dezero - dele; alge.shove(zove,zize); } // check operation type if(otchk == (int)'*') { deri.unshove(zove,zize); dele.unshove(zove,zize); alge = dele * deri; alge.shove(zove,zize); } // right // check operation type if(otchk == (int)'/') { deri.unshove(zove,zize); dele.unshove(zove,zize); alge = dele / deri; alge.shove(zove,zize); } // right dtallow=0; while(dtallow < 80) { lionel[dtallow]='\0'; dtallow=dtallow+1; } otchk=alge.decompwr(1,78,10,lionel); std::cout << lionel; std::cout << std::endl; } } return 0; } // end of program /* after end of program Tag: Easy question on numeric limits Tag: 295387
  - 7
    - Passing an array of structuresfrom a pointer? Hello, I have a two part question. I do understand the principle of an array like so: int xxx[10]; where xxx holds the address of the first item. But, when using this analogy with arrays of structures, I feel a little unsure! Anyways, here is my problem! -[Q#1]- I am trying to pass the address of an array of structures, that have already been declared in one of my headers. I don't know... but I feel like I am drawing a blank ! I know I did similar stuff like this but not from the same exact perspective and the fact that I need to pass the address of an array of strucutes somehow confuses me! Basically I need to pass the address of the array of structures to a fiunction that will then store this address in another structure for later use. Here's what I am trying to accomplish! In a header file I will be declaring several strucutures that have several members. Furthermore, an array for every one of these structures will be declared. For example, as shown below, the "ddlbColl1" structure will have an array of itself declared as "DDLB_COLL1". In my example below I only declared one (1) such strucure called "ddlbColl1", but someday I could end up having 2, 3 or even 5 of them! The thing is, that in my main calling function I want to configure all the necessary components for my drop down list box by calling the ULC_DDLB_config_ddlb ( ) function which will setup informations in the "ddListBox" structure . One of the parameters of this function will be the desired array of strucutres collections such as the DDLB_ARY[] ! Then once back in main I would call the: ULC_DDLB_ddlb (obj_DDLB); function and pass the pointer to a ddListBox structure and manipulate its data, such as the data from the DDLB_COLL1 array within the function's implementation code. -[Q#2]- When I am in the void ULC_DDLB_ddlb () function, will I be able to reference data from the DDLB_COLL1 array? I believe that to retreive or modify the data in the DDLB_COLL1 array I can do it like this: right? DDLB_COLL1[1]. LIST_NUMBER; I could be wrong, this is the first time I am passing arrays around this way!!!! Here's the code sample, I really am unsure of the lines flagged with "[***]" !!! All help is appreciated! ============================================== typedef struct ddListBox{ int x; long y; struct ddlbColl1 DDLB_ARY[]; //Store address of DDLB_COLL1 here ! [***] } DDLB; DDLB *obj_DDLB; struct ddlbColl1{ long LINK_ID; int LIST_NUMBER; } DDLB_COLL1[] = { 200, 4, 201, 8}; DDLB* ULC_DDLB_config_ddlb ( struct ddlbColl1 DDLB_ARY[], int x, int y) { obj_DDLB = NULL; obj_DDLB = (DDLB*) malloc (sizeof (struct ddListBox)); obj_DDLB->x = x; obj_DDLB->y = y; obj_DDLB->DDLB_ARRAY = DDLB_ARY; //Store address of DDLB_COLL1 ! [***] return obj_DDLB; } void ULC_DDLB_ddlb ( DDLB *obj_DDLB) { // This is the implementation file where I would like to use the informations from // the DDLB_COLL1 array! } int main() { obj_DDLB = ULC_DDLB_config_ddlb ( DDLB_COLL1, 10, 5); ULC_DDLB_ddlb (obj_DDLB); return 0; } ==================================================== If this is too complicated I can simply find an easier more ineffecient way, but I really find this would be a good fit for my project since all I have to do, is set my drop down list box once with the ULC_DDLB_config_ddlb() function and then just fetch the ULC_DDLB_ddlb(obj_DDLB) function and access all the info's via the obj_DDLB pointer using the points to operator "->". I really would appreciate any help on how I can store the address of an array of structures in the respective places in the above example flagged by [***] ! Sincerely, thanking all in advance! PS. PLease note, I haven't shown the code for any testing of pointers returned by malloc or any free()'s in order to keep the code short and simple! -- Best regards Roberto Tag: Easy question on numeric limits Tag: 295385
  - 8
    - GetProcAddress failed? Hello everyone, Here is my below code for the DLL consumer and the DLL itself's code. I am using LoadLibrary and GetProcAddress to get the function pointer and then make a call. But statement " StoreDataFp StoreData = (StoreDataFp)GetProcAddress(hMod,TEXT("StoreData"))" always return 0 for StoreData. What is the root cause? DLL Consumer code: [Code] // The process code #include <windows.h> #include <stdio.h> #define THREADCOUNT 1 #define DLL_NAME TEXT("testdll.dll") VOID ErrorExit(LPSTR); typedef BOOL(*StoreDataFp)(DWORD); typedef BOOL (*GetDataFp)(DWORD*); DWORD WINAPI ThreadFunc(VOID) { int i; HMODULE hMod; hMod = LoadLibrary(TEXT("testdll.dll")); StoreDataFp StoreData = (StoreDataFp)GetProcAddress(hMod,TEXT("StoreData")); GetDataFp GetData = (GetDataFp)GetProcAddress(hMod, TEXT("GetData")); if(!StoreData(GetCurrentThreadId())) ErrorExit("StoreData error"); for(i=0; i<5; i++) { DWORD dwOut; if(!GetData(&dwOut)) ErrorExit("GetData error"); if( dwOut != GetCurrentThreadId()) printf("thread %d: data is incorrect (%d)\n", GetCurrentThreadId(), dwOut); else printf("thread %d: data is correct\n", GetCurrentThreadId()); Sleep(0); } return 0; } int main(VOID) { DWORD IDThread; HANDLE hThread[THREADCOUNT]; int i; // Load the DLL LoadLibrary(DLL_NAME); // Create multiple threads. for (i = 0; i < THREADCOUNT; i++) { hThread[i] = CreateThread(NULL, // default security attributes 0, // use default stack size (LPTHREAD_START_ROUTINE) ThreadFunc, // thread function NULL, // no thread function argument 0, // use default creation flags &IDThread); // returns thread identifier // Check the return value for success. if (hThread[i] == NULL) ErrorExit("CreateThread error\n"); } WaitForMultipleObjects(THREADCOUNT, hThread, TRUE, INFINITE); return 0; } VOID ErrorExit (LPSTR lpszMessage) { fprintf(stderr, "%s\n", lpszMessage); ExitProcess(0); } [/Code] DLL code: [Code] // The DLL code #include <windows.h> static DWORD dwTlsIndex; // address of shared memory // DllMain() is the entry-point function for this DLL. BOOL WINAPI DllMain(HINSTANCE hinstDLL, // DLL module handle DWORD fdwReason, // reason called LPVOID lpvReserved) // reserved { LPVOID lpvData; BOOL fIgnore; switch (fdwReason) { // The DLL is loading due to process // initialization or a call to LoadLibrary. case DLL_PROCESS_ATTACH: // Allocate a TLS index. if ((dwTlsIndex = TlsAlloc()) == TLS_OUT_OF_INDEXES) return FALSE; // No break: Initialize the index for first thread. // The attached process creates a new thread. case DLL_THREAD_ATTACH: // Initialize the TLS index for this thread. lpvData = (LPVOID) LocalAlloc(LPTR, 256); if (lpvData != NULL) fIgnore = TlsSetValue(dwTlsIndex, lpvData); break; // The thread of the attached process terminates. case DLL_THREAD_DETACH: // Release the allocated memory for this thread. lpvData = TlsGetValue(dwTlsIndex); if (lpvData != NULL) LocalFree((HLOCAL) lpvData); break; // DLL unload due to process termination or FreeLibrary. case DLL_PROCESS_DETACH: // Release the allocated memory for this thread. lpvData = TlsGetValue(dwTlsIndex); if (lpvData != NULL) LocalFree((HLOCAL) lpvData); // Release the TLS index. TlsFree(dwTlsIndex); break; default: break; } return TRUE; UNREFERENCED_PARAMETER(hinstDLL); UNREFERENCED_PARAMETER(lpvReserved); } extern "C" __declspec(dllexport)BOOL WINAPI StoreData(DWORD dw) { LPVOID lpvData; DWORD * pData; // The stored memory pointer // Retrieve a data pointer for the current thread lpvData = TlsGetValue(dwTlsIndex); // If NULL, allocate memory for the TLS slot for this thread if (lpvData == NULL) { lpvData = (LPVOID) LocalAlloc(LPTR, 256); if (lpvData == NULL) return FALSE; if (!TlsSetValue(dwTlsIndex, lpvData)) return FALSE; } pData = (DWORD *) lpvData; // Cast to my data type. // In this example, it is only a pointer to a DWORD // but it can be a structure pointer to contain more complicated data. (*pData) = dw; return TRUE; } extern "C" __declspec(dllexport) BOOL WINAPI GetData(DWORD *pdw) { LPVOID lpvData; DWORD * pData; // The stored memory pointer lpvData = TlsGetValue(dwTlsIndex); if (lpvData == NULL) return FALSE; pData = (DWORD *) lpvData; (*pdw) = (*pData); return TRUE; } [/Code] thanks in advance, George Tag: Easy question on numeric limits Tag: 295376
  - 9
    - Soliciting user input in an edit box Hello, I have an application in which a dialog contains an object that does some processing, during which it calls a callback function in the parent dialog. I would like to prompt the user to input something in an edit box in the dialog and suspend processing until the user input is complete (either the user hits <enter> or clicks a button, etc.). Then the callback function will return the user input. I could do this with a modal dialog that pops up. However, I don't want a separate dialog just for the input. Rather, I would like the edit box to behave like a modal component. Is there a way to suspend processing and wait on input in this way? Any help would be greatly appreciated. Regards, Dan Tag: Easy question on numeric limits Tag: 295373
  - 10
    - question about cin Hello everyone, I posted the source codes related with assembly code for a very simple program below. I have inlined my questions. Let me repeat my question here, 1. ebp is not used in function main? only esp is used to represent variables on stack? 2. what is the purpose of rdx in statement "lea rdx,[rsp+30h]", store the 2nd parameter to operator <<, which is address of b (address of an integer)? [Code] #include <iostream> using namespace std; int foo (int a) { int b; cin >> b; a = a * b; return a; } int main() { 0000000140001000 sub rsp,28h // ebp is not used? cout << foo (10); 0000000140001004 mov rcx,qword ptr [__imp_std::cin (140002080h)] 000000014000100B lea rdx,[rsp+30h] // rcx as the 1st parameter for function operator >> // what is the purpose of rdx here, store the 2nd parameter to operator >>, which is address of b (address of an integer)? 0000000140001010 call qword ptr [__imp_std::basic_istream<char,std::char_traits<char> >::operator>> (140002078h)] 0000000140001016 mov r11d,dword ptr [rsp+30h] 000000014000101B mov rcx,qword ptr [__imp_std::cout (140002090h)] 0000000140001022 lea edx,[r11+r11*4] 0000000140001026 add edx,edx 0000000140001028 call qword ptr [__imp_std::basic_ostream<char,std::char_traits<char> >::operator<< (140002088h)] return 0; 000000014000102E xor eax,eax } [/Code] thanks in advance, George Tag: Easy question on numeric limits Tag: 295370
  - 11
    - queryperformance i understand how queryperformancecounter/-frequency works, and how to use it for my own code. however: i don't understand how to use and convert the 'LARGE_INTEGER' union to an integer. please help me out Starglider_4 Tag: Easy question on numeric limits Tag: 295353
  - 12
    - clock() hi everybody, i use clock() to synchronize all functions in my 3d game: clock_t d_t=clock()-previous_t; in clock() 1 millisec is the smallest measurable unit. my question: is there a timer function available with more precicion, say 0.1 millisec? regards, marc Tag: Easy question on numeric limits Tag: 295327
  - 13
    - Calling managed function from unmanaged code Hi to all! I am writing my first .NET app and I already have to subclass .NET window and work with unmanged code to get notification about any content being copied into clipboard. Here's a relevant part of my code: #pragma unamanged #include <windows.h> namespace UnManagedCode { void* lpFunc; // Pointer to managed function to be called void RegisterManagedNotificationFunc(void* lpFunc) { lpManagedFunc = lpFunc; } extern "C" long __cdecl UnManagedWinProc(HWND hWnd, unsigned int uMsg, unsigned int wParam, long lParam) { // Check to see if we have been notified of new data on the clipboard if(uMsg == WM_DRAWCLIPBOARD) { // Call a managed func __asm{call lpManagedFunc}; // Pass this message to another registered viewer window behind // TODO } // Pass control and data to default window procedure and return its return value to OS return CallWindowProc((WNDPROC)lngPrevWinProc, hWnd, uMsg, wParam, lParam); } } #pragma managed namespace ClipBoardApp { ................ private: System::Void SystemClipBoard_ContentsChanged() { // Called by unmanaged part as a notification event for managed part // TODO } } Now how will the unmanaged part of the code be able to detemine the pointer to managed function 'Void SystemClipBoard_ContentsChanged'? Some details on how managed function calls are different from calling unmanaged functions will also be helpful. Thanks, Sanjeev Tag: Easy question on numeric limits Tag: 295324
  - 14
    - Making DLL work with VB and VC++ I have a DLL that I've written in C++ and it works fine when called from other C++ projects. But VB programs crash & burn when I try to call any functions in it. Is there something I can do to make a DLL work with both languages? The export declarations I use are like this: extern "C" __declspec(dllexport) int DLL_Init(int port1,int port2 = NULL) (This one should accept either 1 or 2 int arguments... can VB do that too?) Thanks Tag: Easy question on numeric limits Tag: 295319
  - 15
    - include file for a COM client (consumer) Hello everyone, When writing an ATL COM server, MIDL compiler will automatically generated file _i.c and _i.h from IDL definition. My question is whether my below usage for the two files are correct when writing a COM client (consumer)? 1. I put _i.h file into the "Header files" of the client project, and using it to make the build for client project; 2. I put the _i.c into the "Source files" of the client project, and using it to make the build for client project. thanks in advance, George Tag: Easy question on numeric limits Tag: 295310
  - 16
    - T-SQL type application? Hello everyone, When using debugger to attach to a process written by me, its type is T-SQL, x64 and Managed. What does T-SQL mean? (my codes is using ADO.Net, does it mean all application which uses ADO.Net objects are type of T-SQL?) thanks in advance, George Tag: Easy question on numeric limits Tag: 295309
  - 17
    - value of uninitialized variables This is my code: // cl2.h class cl2{ int numero; public: int Aggiungi() { return numero; } }; //main.cpp #include <iostream> #include "cl2.h" using namespace std; int main(...) { cl2 c ; cl2 d ; cl2 e ; cout << c.Aggiungi() << endl; cout << d.Aggiungi() << endl; cout << e.Aggiungi() << endl; return 0; } I see on the screen always the value -858993460 every time I rum the program in debug mode and 2018779649 in release mode. Is there a reason for this? With others compiler I see a similar behaviour with different numbers. I already know that I can fix the problem with contructors, but here I just want to understand why. Giuseppe Tag: Easy question on numeric limits Tag: 295308
  - 18
    - VC++ 2005/2008 failed to open a file whose filename encoded in MBCS? Hi All, I've run across a problem of failing to open a file whose filename is encoded in traditional Chinese (BIG5). The example program is shown below. I've conducted a test using VC++ 2008 express, VC++ 2005 and VC++ 6.0, all on Windows XP-sp2 and found the programs compiled by VC++ 2008 and 2005 failed to open the file and hence to access the content. However, the program compiled vc++ 6.0 succeeded to do it. Almost the program settings are using the default ones except the following three items: 1) debug mode; 2) using MBCS; 3) no pre-compiled header. The file, =AD=BA=AD=B6.txt, is composed of a string of "=AD=BA=AD=B6.txt" a= nd the file, a.txt, is also composed of a string of "This is a string". I'm wondering whether this is a bug introduced by VC++ 2005 and 2008. If it is NOT, would you please show me the way of how to attack this problem. Thanks, JSK ////////////////// program ////////////////// #include <fstream> #include <iostream> void OpenAndPrintFile(const char *chFile); int main() { OpenAndPrintFile("=AD=BA=AD=B6.txt"); cout << endl; OpenAndPrintFile("a.txt"); return 0; } void OpenAndPrintFile(const char *chFile) { cout << "start to read " << chFile << endl; ifstream hFile; hFile.open(chFile); if (!hFile.fail()) { char chTmp[2049]; chTmp[2048] =3D '\0'; while (!hFile.eof() ) { hFile.getline(chTmp, 2048); cout << chTmp << endl; } } else cerr << "failed to open " << chFile << endl; } /////////////////////////////////////////////////// Tag: Easy question on numeric limits Tag: 295307
  - 19
    - Global Mutex on Vista How to create a Global Mutex on Windows Vista which can be accessed across all windows user sessions. I created a named mutex using "Global\\MyMutexName" but it is not working on Vista while on XP SP2 it is working fine. Regards, Manish Agarwal Tag: Easy question on numeric limits Tag: 295303
  - 20
    - WAB (Windows Address Book) I'm trying to enumerate the WAB, which has names and email addresses. There seem to be two ways to do this, the old APIs load and call wab32.dll's WABOpen function and new (Vista-only) APIs that use the IContactManager COM interface. I have some code working for both of these ways, but in both cases it only sees the address book entries from Outlook Express/Windows Mail. It does not get any address book entries from Outlook. I'm wondering if I'm missing something in these APIs, or if Outlook does not make its (local contacts) list available via the APIs? -- Scott McPhillips [VC++ MVP] Tag: Easy question on numeric limits Tag: 295296
  - 21
    - Can LoadIcon be used to load the icon of another process Hello LoadIcon take HInstance as one of its parameters. We usually use LoadIcon to load the icon of the current process instance: LoadIcon(hInstance, MAKEINTRESOURCE(...)) Can it be used to load the icon resource of another process? Suppose I can call OpenProcess to get the process's handle. What parameter should I set for MAKEINTRESOURCE? THANKS Tag: Easy question on numeric limits Tag: 295276
  - 22
    - How to get the array size of TCHAR* Hello I get an arrary of TCHAR* as parameter TCHAR* env[] How can I get the entry count in this array? No param tells me the count value. Tag: Easy question on numeric limits Tag: 295272
  - 23
    - VkKeyScan problem Hello, How can i use VkKeyScan to get the scan code of keys like 'F1', 'Pg Up', 'Num 0', etc.. Regards and Thanks, John. Tag: Easy question on numeric limits Tag: 295255
  - 24
    - VkKeyScanCode Hi, How can i use VkKeyScan to return the values of Pg Up,Pg Down,F1,F2, etc Thanks in advance. Regards, John. Tag: Easy question on numeric limits Tag: 295254
  - 25
    - Forward declaration of managed class: howto? Hi, I'm using VC++ 2008 Express. I have this managed class: ref class C1 { private: UINT i; public: C1() { //code }; void MyMethod() { //code }; } How am I to forward declare this class and it's public members? What do I have to put in the header file? I'd like to include the header file in another class, create an instance of C1 and access it's public members. However, if I only write ref class C1; it complains when trying to create an instance (using gcnew keyword) because it doesn't find the appropriate constructor. Now, if I do add it in the forward declaration this way ref class C1 { C1(); }; then it works but it complains at the definition of the class: error C2011...class redefinition. But the first should only be the declaration, not the definition. So, how to declare only but not define in the header file? I've searched for "forward declaration" a lot but didn't find anything satisfying. All I found was forward declaring the class but without mentioning it's members. Armin Tag: Easy question on numeric limits Tag: 295237

Hi,

I've mostly only coded in stl c++ before despite using Visual studio
for a long time for C#. So this is a new one on me.

In stl c++ I could do the following

std::numeric_limits<float>::max()

and thus get the maximum value a float could hold on a machine. This
is a bit non-critical at the moment, as I can just SET the value since
I'm only developing for one platform, but here's my problem.

In VS2008, it sees that as an implementation of _max(int a, int b),
and won't compile.

I'm clearly missing something. I'm including <limits>, is there some
other include that I'm missing?

Top

Re: Easy question on numeric limits by Victor

Victor
Thu Oct 09 09:03:40 CDT 2008

Scoots wrote:
> I've mostly only coded in stl c++ before despite using Visual studio
> for a long time for C#. So this is a new one on me.
>
> In stl c++ I could do the following
>
> std::numeric_limits<float>::max()
>
> and thus get the maximum value a float could hold on a machine. This
> is a bit non-critical at the moment, as I can just SET the value since
> I'm only developing for one platform, but here's my problem.
>
> In VS2008, it sees that as an implementation of _max(int a, int b),
> and won't compile.
>
> I'm clearly missing something. I'm including <limits>, is there some
> other include that I'm missing?

No, it's the macro 'max' from <windows.h> that's probably interfering
with your work. Try

#define NOMINMAX

at the top of your source code.

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask

Top

Re: Easy question on numeric limits by Scoots

Scoots
Thu Oct 09 09:12:16 CDT 2008

On Oct 9, 10:03=A0am, Victor Bazarov <v.Abaza...@comAcast.net> wrote:
> No, it's the macro 'max' from <windows.h> that's probably interfering
> with your work. =A0Try
>
> =A0 =A0#define NOMINMAX
>
> at the top of your source code.
>
> V
> --
> Please remove capital 'A's when replying by e-mail
> I do not respond to top-posted replies, please don't ask- Hide quoted tex=
t -
>
> - Show quoted text -

Yup, that did it! No wonder I'd never run into it before, and you are
quite right in that windows.h was included in the stdafx.h.

Thanks Victor,
~Scoots

Top