Re: [Vala] Memory safety

[Michele Dionisio <michele dionisio gmail com>]

do you think that  in a code like:



void func() {

 uint64[] output = new uint64[1];  // return callback

 GLib.Timeout.add( 1000,
                            () => {
                                     output[0]=1;
                                     return false;
                                  }
                );
}


I have to check if output is still valid?

regards


2014-05-15 10:22 GMT+02:00 Luca Bruno <lethalman88 gmail com>:

> On 15/05/2014 10:18, Florian Weimer wrote:
>
> > Hi,
> >
> > am I correct in assuming that Vala is supposed to inherit the lack of
> > memory safety from C?  Or put differently: If there is Vala code that does
> > not guard callers from memory safety violations but it could do so, this is
> > not a problem because callers are generally expected to fulfill such
> > (implicit) preconditions.
> >
> >  Vala inherits the problem of C, yes. You could certainly do some evil
> casts like in C. Vala however is certainly safer than C in many aspects, in
> other aspects however you have to know what you are doing and how Vala
> compiles down to C in certain cases.
> For example:
> 1. func1(): create a local array
> 2. Call async func2(array) without waiting the function to finish
> 3. Return from func1
> 4. func2 runs with a dangling pointer to the array
>
> Vala is not free of rough edges, it makes your life easier. But since a
> Vala array is a standard C array, it can't be referenced counted and that's
> why you get the above (unwanted from a newbie perspective) behavior.
>
> Best regards
>
> _______________________________________________
> vala-list mailing list
> vala-list gnome org
> https://mail.gnome.org/mailman/listinfo/vala-list