Reg Query Problem

Hey Guys

Can anyone see a problem with this code its meant to read a value out of a string in teh registry and there is definately a value there. All this does though is display 3 random characters. The dword returns the right length of the string though.

Thanks for the help
Peter

Code:

---

        TCHAR sInstallDirectory[MAX_PATH];
        BYTE sInstallDirectoryUNI[MAX_PATH];
        DWORD dwLengthOfInstallDirectory = NULL;

        HKEY hUninstallInfo;
        RegOpenKeyEx(HKEY_LOCAL_MACHINE, "SOFTWARE\\RCAS\\UninstallInfo", 0, KEY_ALL_ACCESS, &hUninstallInfo);

        if(RegQueryValueEx(hUninstallInfo, "InstallDirectory", NULL, NULL, &sInstallDirectoryUNI[0], &dwLengthOfInstallDirectory) == ERROR_SUCCESS)
        {
                TCHAR sTest[10];
                ultoa(dwLengthOfInstallDirectory, sTest, 10);
                AfxMessageBox(sTest);

                for(int i = 0; i <= dwLengthOfInstallDirectory; i++)
                {
                        sInstallDirectory[i] = sInstallDirectoryUNI[i];
                }
                sInstallDirectory[i++] = NULL;
       
                AfxMessageBox(sInstallDirectory);
        }
        else
        {
                AfxMessageBox("Read Reg Entry Failed");
                CErrorHandler Error1;

                DWORD dwLastError;
                dwLastError = GetLastError();
                Error1.ResolveError(dwLastError);
        }

---

Your loop doesn't translate unicode, if in fact, it just copies
it. sIntallDirectory has a lot of 'junk' bytes in it. Unicode is a two-byte character. This is the reason you're getting garbage. At the moment, I don't remember how to convert unicode to standard ASCII.

Anyone else know?

Code:

---

for(int i = 0; i <= dwLengthOfInstallDirectory; i++)
                {
                        sInstallDirectory[i] = sInstallDirectoryUNI[i];
                }
                sInstallDirectory[i++] = NULL;

---

I went brain dead.

Convert from wide to ASCII with:

size_t wcstombs( char *mbstr, const wchar_t *wcstr, size_t count );

That actually converts to MBCS.

As of today, I still haven't found any conversion from Unicode to straight ASCII.

Parksie -
do you know about USES_CONVERSION, part of atl?
There are a bunch of macros. A2W, W2A, etc.

Code:

---

#include <atlbase.h>

USES_CONVERSION;

LPCWSTR strWide = L"unicode";
LPCSTR strASCII = W2A(strWide);

---

Yes but that only applies to MSVC.

Don't you need to bring COM in in order to use ATL?

I think so.

Anyway, it would add a lot of complexity to what the pskboy is trying to do. Those macros are the first thing I think about when I run into wide chars. wcstombs is at the bottom of the list.

You've confused me.

You said it would add a lot of complexity but it's the first thing on your list :confused:

I wrote a couple of operators for the strings (i.e. you could implicitly create a string from a wstring and vice versa) but I forgot which bit of code I put them in. The main problem with that is that it's not pure Unicode->ASCII - it does MBCS since that's all that I could find that was supported.