[RFC] Throwing an exception with null pointers

[Dave P.]

On Thursday, 17 April 2025 at 05:31:48 UTC, Walter Bright wrote:
> On 4/16/2025 2:49 PM, Dave P. wrote:
>> [...]
> I've long since given up on memorizing the documentation of 
> each parameter. Too many functions! I just google it, it just 
> takes a sec.

It’s easier to goto-definition and see the annotations. Fancier 
editors can even show it inline.

> [...]
> I don't see how a runtime detector can work better than a seg 
> fault with stack trace.

The source of a null pointer can be very far from its eventual 
dereference, especially when stored in a structure.

The nullability sanitizer also doesn’t kill your program by 
default, it just logs when a null pointer is stored in a nonnull 
variable. So you can see where the first non null store is made 
and see how it eventually moves to when it is derefenced.

---

It’s funny though, when I first started using this kind of thing 
it caught loads of bugs. But after years of use, I realized it 
hadn’t caught any bugs in a long time. I had internalized the 
rules.