11:55 <forgot-password> Calling @typeInfo on a slice gives me back a TypeInfo.Pointer with a size value of 'Slice', correct?

12:09 <Hejsil> Yes, forgot-password

12:11 <forgot-password> So if I were to create a function that could handle slices *and* arrays I had to check via @typeInfo to get a pointer to the actual data, right? Because &<slice> gives me a pointer to the slice struct instead of its contents for which I need <slice>.ptr

12:12 <Hejsil> I'm not sure exactly what you're trying to do. Why do you need to know that it's an array or a slice

12:12 <Hejsil> These types support most of the same operations

12:13 <forgot-password> Wait, I'm gonna put together a little sample. I want to pass the data to OpenGL

12:13 <Hejsil> Alright

12:13 <Hejsil> I guess you need a ?[*]T correct?

12:14 <Hejsil> You could do: arr_or_slice[0..].ptr

12:17 <forgot-password> I need a ?*const c_void

12:18 <forgot-password> And my function takes a parameter "geometry: var"

12:18 <forgot-password> To that I need a pointer

12:18 <forgot-password> &geometry works fine for arrays, but it fails for structs. Presumably to the aforementioned reason

12:19 <Hejsil> This should work too: &arr_or_slice[0]

12:20 <Hejsil> I would think

12:20 <Hejsil> Hmm, unless your slice/array is len 0

12:21 <forgot-password> At least currently it fails if I pass an array of len 0

12:21 <Hejsil> This is my best hack: &arr_or_slice[0..].ptr[0]

12:22 <Hejsil> There should probably be a unified way for arrays and slices to get the underlying data

12:38 <forgot-password> Thanks, &arr_or_slice[0] worked perfectly fine

12:46 <forgot-password> Though I agree with you about the unified way to access the underlying data. It was rather confusing the first time I tried to pass &slice to glBufferData

19:42 <MajorLag> RE: #1503, It kinda feels like an issue you'd get when dealing with padded data buffers and forgetting to adjust for the padding. I'm not sure what that would mean in this context though.

19:42 <MajorLag> by padded, I mean aligned. Like how BMP aligns rows of pixels to 4-byte boundaries.

23:07 <forgot-password> Hey, I was just wondering if the parameters are flipped intentionally between std.math.atan2 and its underlying functions atan2_32/64.

23:09 <andrewrk> forgot-password, good catch. the names are incorrect in std.math.atan2. the first arg is y and second is x. I'll push a fix

23:13 <andrewrk> pushed

23:17 <forgot-password> andrewrk: Nice, thank you

23:18 <forgot-password> I was playing around with custom format functions earlier today, and I thought that the fmt parameter to said function is just what's inbetween the braces, but that doesn't seem to be the case, right?

23:20 <andrewrk> I'm not sure, I haven't looked at the details of that code recently. But keep in mind there's an API overhaul planned: https://github.com/ziglang/zig/issues/1358