[2/2] oleacc: implemented GetRoleText[A/W] with tests

[Nikolay Sivov]

Dmitry Timoshkov wrote:
> "Nikolay Sivov" <bunglehead at gmail.com> wrote:
>
>> +UINT WINAPI GetRoleTextA(DWORD role, LPSTR lpRole, UINT rolemax)
>> +{
>> +    UINT length;
>> +    WCHAR *roletextW;
>> +    CHAR *roletextA;
>> +
>> +    TRACE("%u %p %u\n", role, lpRole, rolemax);
>> +
>> +    length = GetRoleTextW(role, NULL, 0);
>> +    if(length == 0)
>> +        return 0;
>> +
>> +    roletextW = HeapAlloc(GetProcessHeap(), 0, (length + 
>> 1)*sizeof(WCHAR));
>> +    if(!roletextW)
>> +        return 0;
>> +
>> +    GetRoleTextW(role, roletextW, length + 1);
>> +
>> +    length = WideCharToMultiByte( CP_ACP, 0, roletextW, -1, NULL, 0, 
>> NULL, NULL );
>> +
>> +    if(!lpRole){
>> +        HeapFree(GetProcessHeap(), 0, roletextW);
>> +        return length - 1;
>> +    }
>> +
>> +    roletextA = HeapAlloc(GetProcessHeap(), 0, length);
>> +    WideCharToMultiByte( CP_ACP, 0, roletextW, -1, roletextA, 
>> length, NULL, NULL );
>> +
>> +    if(rolemax < length)
>> +        length = rolemax;
>> +
>> +    memcpy(lpRole, roletextA, length);
>> +    lpRole[length-1] = '\0';
>> +
>> +    HeapFree(GetProcessHeap(), 0, roletextA);
>> +    HeapFree(GetProcessHeap(), 0, roletextW);
>> +
>> +    return length - 1;
>> +}
>
> Why do you need to allocate an intermediate roletextA?
>
WideCharToMultiByte requires to have a sufficient buffer. I think it's 
the easiest way to allocate necessary buffer instead of using local 
buffer of some hardcoded length. What do you think?