On Friday, 2 May 2025 at 22:19:53 UTC, Andy Valencia wrote:
In the following code, two questions. First, is there any
difference between "x[] = y" and "x[] = y[]"? It appears not.
`y` is already a slice `int[]`, so slicing it does not change the
type. Slicing without indices selects all elem
On 5/2/25 2:44 PM, Python wrote:
> On Friday, 2 May 2025 at 14:33:57 UTC, Richard (Rikki) Andrew Cattermole
> wrote:
>> Have you tried filter?
>>
>> https://dlang.org/phobos/std_algorithm_iteration.html#.filter
>
> Thx. Anyway, what find returns aftrr first occurrence is unexpected.
Yes, understa
On Saturday, 3 May 2025 at 11:18:00 UTC, Nick Treleaven wrote:
Second, in assigning from arrays of differing sizes, Phobos
causes an illegal instruction, rather than the sort of
exception I'd have expected. I'm curious why they stepped
away from D's exception architecture?
It throws a RangeErr
You link with release version of druntime, try to link with debug
version.
But yeah, core.internal.util.array is wrong design, it should be
boolean function, and the compiler should generate
assert(areTypedArraysConformable()); at the caller side.
On Saturday, May 3, 2025 8:23:00 AM Mountain Daylight Time Andy Valencia via
Digitalmars-d-learn wrote:
> At least on 1.40.1 of the ldc2 distro for x86-64, uses the
> "illegal instruction" instruction.
That sounds like an ldc bug then. With dmd, your program gives
[2, 2, 2, 2, 1, 1, 1, 1]
[2, 2,
On Saturday, May 3, 2025 12:08:45 PM Mountain Daylight Time Ali Çehreli via
Digitalmars-d-learn wrote:
> On 5/2/25 2:44 PM, Python wrote:
> > On Friday, 2 May 2025 at 14:33:57 UTC, Richard (Rikki) Andrew Cattermole
> > wrote:
> >> Have you tried filter?
> >>
> >> https://dlang.org/phobos/std_
On Friday, 2 May 2025 at 16:53:05 UTC, monkyyy wrote:
This is still a continuation of aliases are for types, enums
for literals
```d
alias F=(i)=>i+1;
```d
pragma(msg, is(F)); // false, F is not a type
```
Even if we give `i` a type:
```d
alias F=(int i)=>i+1;
pragma(msg, is(F)); // still fa
