p5.Camera.slerp() behaves strangely

[inaridarkfox4231]

Most appropriate sub-area of p5.js?

• Accessibility
• Color
• Core/Environment/Rendering
• Data
• DOM
• Events
• Image
• IO
• Math
• Typography
• Utilities
• WebGL
• Build Process
• Unit Testing
• Internalization
• Friendly Errors
• Other (specify if possible)

p5.js version

1.8.0

Web browser and version

Chrome

Operating System

Windows11

Steps to reproduce this

Steps:

1. Performing a reset on the p5.Camera.slerp() demo page
2. The camera behaves strangely while trying to return to its original position.

Snippet:

SLERP_BUG

let mycam, lastCam, initialCam;
let countForReset = 30;
// This sample uses orbitControl() to move the camera.
// Double-clicking the canvas restores the camera to its initial state.
function setup() {
  createCanvas(400, 400, WEBGL);
  strokeWeight(3);

  mycam = createCamera(); // main camera
  lastCam = createCamera(); // Camera for recording loc info before reset
  initialCam = createCamera(); // Camera for recording the initial state

  setCamera(mycam); // set main camera
}

function draw() {
  if (countForReset < 30) {
    // if the reset count is less than 30,
    // it will move closer to the original camera as it increases.
    countForReset++;
    mycam.slerp(lastCam, initialCam, countForReset / 30);
  } else {
    // if the count is 30,
    // you can freely move the main camera with orbitControl().
    orbitControl();
  }

  background(255);
  box(160);
}
// A double-click sets countForReset to 0 and initiates a reset.
function doubleClicked() {
  if (countForReset === 30) {
    countForReset = 0;
    lastCam.set(mycam);
  }
}

2023-10-29.22-00-05.mp4

In addition, we have confirmed that slerp behaves strangely in some sketches.

I know the cause.
#6287
This is because the specifications of row() and column() were changed in this commit, calling it a bug.
I was the one who created this specification, but from the perspective of looking at the elements of a matrix from left to right, this is not a bug.

However,

It no longer matters what is a bug or what is not. I'm tired of repeating pointless discussions about it.
I have no connection with the circumstances or validity of this specification change, and I have no interest in it.
This problem has already been resolved, so I'll just state my conclusion, and create a pull request, and solve it. That's it.

CAMERA_SLERP_BUG

  // BUG IS HERE.

  var front0 = rotMat0.column(2);
  var front1 = rotMat1.column(2);
  var up0 = rotMat0.column(1);
  var up1 = rotMat1.column(1);

  // FIX THE BUG


  // The specifications of column() and row() have been reversed, so column() must be rewritten as row().
  var front0 = rotMat0.row(2);
  var front1 = rotMat1.row(2);
  var up0 = rotMat0.row(1);
  var up1 = rotMat1.row(1); // prepare new vectors.

2023-10-29.22-10-43.mp4

That's all.
It may affect unit tests, but since this behavior is correct, it makes sense to adapt to it.