"Fred Zwarts" <> wrote in message
news:imcbim$6mu$...
"Paul" <> wrote in message
news:826ip.79934$2...
>
> "Fred Zwarts" <> wrote in message
> news:imaitg$ei$...
>
>>>>> If you use the union as a float the memory location does not contain
>>>>> an array.
>>>>>
>>>>> --According to your reasoning,
>>>>> --IntPointer would be also a pointer to a float,
>>>>> --and to a float array, etc. etc.
>>>>>
>>>>> No, when the union chnages to represent a new data-type the pointer to
>>>>> the old data-type is no longer valid IMO.
>>>>
>>>>Your confusion leads you to wrong conclusions.
>>>>The pointer can still be used to assign an int value to x.Integer.
>>>>*IntPointer = 3;
>>>>So, it remains a valid pointer to int, even if the object at the memory
>>>>location is a float.
>>>>
>>> I'm not confused at all by your scenraio.
>>
>> --That is right. Your wrong conclusion does not come from my scenario,
>> --but from your ideas about pointers that are incompatible with the C++
>> --language.
>>
>> If you think p does not point to an array then I think you'll find that
>> it's you who is wrong.
>>
>> int*p = new int[3];
>
> --If you think that the pointer to the old data-type is no longer valid,
> --it is you who is wrong.
> --*IntPointer = 3;
>
> I didn't say the pointer was invalid. I said the pointer points-to the
> union no matter what type the pointer was.
> And you have just proved this is correct.
>
--First Paul said (See above):
>>>>> No, when the union chnages to represent a new data-type the pointer to
>>>>> the old data-type is no longer valid IMO.
--Now Paul summarizes:
> I didn't say the pointer was invalid.
I was obviously talking about using a pointer to access data of the old
data-type.
--Sorry, I'm unable to find the logic in this language.
--If "no longer valid" is the same as "not invalid", anything can be true.
--This makes it impossible for me to help him any further.
Good I don't need nor do I wnat your help TYVM, Goodbye and GL
..
|
Similar Threads
