Different types with auto

[Jacob Carlborg]

On 2011-03-17 21:35, Nick Sabalausky wrote:
> "bearophile"<bearophileHUGS at lycos.com>  wrote in message
> news:iltdqr$1rd7$1 at digitalmars.com...
>>
>> D disallows bug-prone C syntax like this (C style guides strongly suggest
>> to declare only each variable in a distinct statement and line of code):
>>
>> int a = 1, *b = null;
>>
>> D accepts code like:
>>
>>> auto a = 1, b = null;
>>
>> This seems against the D rule of not allowing different types to be
>> initialized in the same statement. In my opinion on this design detail D
>> is worse than C++0x. As an example, if you write a line of code like this,
>> meaning it to initialize six double variables, you have a bug:
>>
>> auto x1=1., x2=2., x3=3., x4=4., x5=5, x6=6.;
>>
>
> I'm not sure how I feel about D's auto creating multiple types in one
> statement. But at the very least, I think your code above is yet another
> good example of why "1." and ".1" float-literal syntax needs to die, die,
> die. It does what? Saves one character just so it can be less readable and
> cause syntax problems for new potential features? Bah.

Yes, it must die. It also conflicts with uniform function call syntax.

>>
>> There are only few (one?) other thing(s) that I think C++0x gets better
>> than D, like a more strict enum, I don't understand why D doesn't follows
>> C++0x design on this other detail:
>> http://d.puremagic.com/issues/show_bug.cgi?id=3999
>>
>
> I agree that would be nice. Weak typing like that is one of the reasons I
> abandoned C/C++, but D still holds on to it in this particular case.
>
>
>


-- 
/Jacob Carlborg