the question is how should the safe stuff look

R

Redhacker2OP•8/12/22, 12:25 AM

like you could have multiple "safe" overloads

R

Redhacker2OP•8/12/22, 12:25 AM

which should you generate

A

Aqua•8/12/22, 12:26 AM

Only thing we need now is

using static Silk.NET.INSANITY;

public class MyClass {

    public void thing() {
        glGenBuffer(GL_VERTEX_BUFFER);
    }

}

using static Silk.NET.INSANITY;

public class MyClass {

    public void thing() {
        glGenBuffer(GL_VERTEX_BUFFER);
    }

}

using static Silk.NET.INSANITY;

public class MyClass {

    public void thing() {
        glGenBuffer(GL_VERTEX_BUFFER);
    }

}

using static Silk.NET.INSANITY;

public class MyClass {

    public void thing() {
        glGenBuffer(GL_VERTEX_BUFFER);
    }

}

Etc

A

Aqua•8/12/22, 12:26 AM

I screwed that up

RRedhacker2 the question is how should the safe stuff look

P

Perksey•8/12/22, 12:26 AM

well that's why i was thinking of the

PointerLike

PointerLike

PointerLike

PointerLike stuff

P

Perksey•8/12/22, 12:27 AM

one parameter type that multiple types implicitly convert to

P

Perksey•8/12/22, 12:27 AM

the main issue i can already see is ref lifetime though and implicit ref conversions

PPerksey well that's why i was thinking of the `PointerLike` stuff

R

Redhacker2OP•8/12/22, 12:27 AM

hmm

R

Redhacker2OP•8/12/22, 12:27 AM

interesting

P

Perksey•8/12/22, 12:27 AM

actually ref lifetimes would be easy

P

Perksey•8/12/22, 12:27 AM

but the implicit conversion of a ref to a type is i think impossible

P

Perksey•8/12/22, 12:27 AM

so it'd have to be

new(ref x)

new(ref x)

new(ref x)

new(ref x) instead of

ref x

ref x

ref x

ref x

A

Aqua•8/12/22, 12:28 AM

Lifetimes? I thought that was only a rust thing

AAqua Lifetimes? I thought that was only a rust thing

R

Redhacker2OP•8/12/22, 12:28 AM

most certainly not

P

Perksey•8/12/22, 12:28 AM

rust is just the thing that makes lifetimes explicit

P

Perksey•8/12/22, 12:28 AM

you have implicit lifetimes in C#

P

Perksey•8/12/22, 12:28 AM

you also have scary things called "GC holes" as well

R

Redhacker2OP•8/12/22, 12:28 AM

it is a thing in all languages, some do better at hiding it

A

Aqua•8/12/22, 12:28 AM

I see

R

Redhacker2OP•8/12/22, 12:29 AM

RAII for instance in C++

P

Perksey•8/12/22, 12:29 AM

int* x = default;
{
    int y;
    x = &y;
}
// y is out of scope thus x's "lifetime" has been exceeded

int* x = default;
{
    int y;
    x = &y;
}
// y is out of scope thus x's "lifetime" has been exceeded

int* x = default;
{
    int y;
    x = &y;
}
// y is out of scope thus x's "lifetime" has been exceeded

int* x = default;
{
    int y;
    x = &y;
}
// y is out of scope thus x's "lifetime" has been exceeded

P

Perksey•8/12/22, 12:30 AM

same with

Span<int> x = default;
{
    int y = default;
    x = MemoryMarshal.AsSpan(ref y, 1);
}
// y is out of scope thus x's "lifetime" has been exceeded

Span<int> x = default;
{
    int y = default;
    x = MemoryMarshal.AsSpan(ref y, 1);
}
// y is out of scope thus x's "lifetime" has been exceeded

Span<int> x = default;
{
    int y = default;
    x = MemoryMarshal.AsSpan(ref y, 1);
}
// y is out of scope thus x's "lifetime" has been exceeded

Span<int> x = default;
{
    int y = default;
    x = MemoryMarshal.AsSpan(ref y, 1);
}
// y is out of scope thus x's "lifetime" has been exceeded

P

Perksey•8/12/22, 12:30 AM

though i realised the easy solution would be just make

PointerLike

PointerLike

PointerLike

PointerLike itself a

ref struct

ref struct

ref struct

ref struct that can track its own refs

P

Perksey•8/12/22, 12:35 AM

ah yep it's illegal https://sharplab.io/#v2:EYLgxg9gTgpgtADwGwBYA0AXEBDAzgWwB8ABAJgEYBYAKGIAYACY8gOgCUBXAOwwEt8YLAJI8YUCAAcAymIBuvMDFwBuGvSatOPfoIDCEfBN4AbMTKjzFKmmoDMDWADMGuDFA5gMDAAoReoqAAZXgBrGAAeABUAPgYAdwALMRgGSJAGbnxsLmwAcxgAExoAbxoGcoYJKF5ZbAwUqQlsqNiAfVwmrlVqCqZ7ZiQGGAQJYwVeL0kxOugfPwDgsJaACidUhgQASjLeioBeWK4YOLn/eqDQiJiGYoZ2zoY9hgBZGHxoAE9n7ChcBOxjCxdLA6jBGtlVjBnAg0AxyJsGABfbqImhAA===

SharpLab

C#/VB/F# compiler playground.

KKai Please use an data source instead of the foor loops and use a XUnit theory

R

Redhacker2OP•8/12/22, 12:58 AM

question, do you want me to write an IEnumerable<(int,int,int, int)>, if so is the behavior to be like the previous for loop, and I imagine you would want both tests to be done in such a manner?

L

Lyris the Kitori•8/12/22, 1:56 AM

okay got everything written down for live backend switching
https://github.com/orgs/Furball-Engine/projects/1/views/1
so hopefully without needing a restart i can switch between OpenGL, GLES, Veldrid, DX11, etc.

L

Lyris the Kitori•8/12/22, 1:56 AM

gonna chew through this one by one

L

Lyris the Kitori•8/12/22, 3:45 AM

ah great version number https://i.beyleyisnot.moe/68oAM2g.png

L

Lyris the Kitori•8/12/22, 4:55 AM

yup nothing wrong here https://i.beyleyisnot.moe/k8IpBpE.png

L

Lyris the Kitori•8/12/22, 4:59 AM

ignoring that mishap, heres quad https://i.beyleyisnot.moe/GlZqz0g.png

L

Lyris the Kitori•8/12/22, 4:59 AM

d3d9

L

Lyris the Kitori•8/12/22, 6:57 AM

dxvk + d3d9 is actually a really good way to get renderdoc debugging

RRedhacker2 question, do you want me to write an IEnumerable<(int,int,int, int)>, if so is t...

K

Kai•8/12/22, 7:25 AM

Not sure I understand what you're saying

P

Perksey•8/12/22, 6:10 PM

L

Lyris the Kitori•8/12/22, 6:11 PM

i can't wait for the day when someone comes in here asking for help cus they tried to use that with something that expects UTF16 :^)

P

Perksey•8/12/22, 6:12 PM

well in that case you have

char* name = "POS"

char* name = "POS"

char* name = "POS"

char* name = "POS"

L

Lyris the Kitori•8/12/22, 6:12 PM

what about ShiftJIS

P

Perksey•8/12/22, 6:12 PM

no idea

L

Lyris the Kitori•8/12/22, 6:13 PM

old japanese c libraries intensify

K

Kai•8/12/22, 7:12 PM

oh god no please

K

Kai•8/12/22, 7:12 PM

I'd rather think about UTF128

K

Kai•8/12/22, 7:14 PM

Since no sentient life has been found on Mars, the code
points that would have been allocated for Martian imojis are hereby
allocated for this private use. These addresses are thus called
"Martians", also known as "Bogons" due to them being bogus.

this is a joke, right?

L

Lyris the Kitori•8/12/22, 7:15 PM

poor martians

K

Kai•8/12/22, 7:15 PM

(That's from RFC-8369, "Internationalizing IPv6 Using 128-Bit Unicode")

K

Kai•8/12/22, 7:16 PM

Either the person that wrote this has taken many drugs or this is a joke

K

Kai•8/12/22, 7:16 PM

I'm leaning towards joke, but I'm not 100% convinced

L

Lyris the Kitori•8/12/22, 7:16 PM

128 bit unicode scares me

K

Kai•8/12/22, 7:16 PM

IPv6 addresses no longer need to be typed as hex values --
instead, the glyph for the script character, symbol, emoji, or
imoji representing that address can be input by the user, and it
will be displayed by the application as the graphic itself. From
the user's perspective, this will also be more compact than the
representation described in RFC 1924 [RFC1924].

I mean that seems like it's a joke, right?

L

Lyris the Kitori•8/12/22, 7:16 PM

def is

K

Kai•8/12/22, 7:17 PM

Feels like it's fairly obvious that this makes no sense, but it's also the kind of thing I've heard people that don't really know what they are talking about suggest

the question is how should the safe stuff look

Similar Threads