[p5.js 2.0 Bug Report] : v.div(2, 4, 6) divides everything by first arg only, v.mult(2, 3, 4) does nothing

[avinxshKD]

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
• Internationalization
• Friendly errors
• Other (specify if possible)

p5.js version

2.2.1-rc.0

Web browser and version

Any (reproduced on Chrome 132)

Operating system

Windows 11

Steps to reproduce this

div() with multiple number args uses args[0] for all components instead of args[i]. So v.div(2, 4, 6) on (10, 20, 30) gives (5, 10, 15) instead of (5, 5, 5)

One char typo p5.Vector.js line 1241:

this._values = this._values.map((val, i) => val / args[0]); // should be args[i]

Also mult() straight up has no branch for multiple number args. v.mult(2, 3, 4) silently returns the vector unchanged. No warning, nothing. Docs say both should work

add() and sub() handle this fine btw

function setup() {
  let v1 = createVector(10, 20, 30);
  v1.div(2, 4, 6);
  console.log(v1.toString()); // (5, 10, 15) — wrong, should be (5, 5, 5)

  let v2 = createVector(10, 20, 30);
  v2.mult(2, 3, 4);
  console.log(v2.toString()); // (10, 20, 30) — unchanged, should be (20, 60, 120)
}

[avinxshKD]

What do you think of this?? ready to go, just let me know @davepagurek @perminder-17

[ksen0]

Hi @avinxshKD thanks so much for flagging this, it's definitely a regressing in 2.x and should be fixed. However, right now I'm neck-deep in a refactor of the p5Vector class (resulting from an earlier community discussion of 2.x vector behavior), so to avoid merge conflict and duplicate work, I would prefer to keep this open but it's not ready for work.

I added the following to the test suite:

    suite('with arglist', function () {
      test('divide the x, y, z with the scalar', function () {
        v.div(2, 3, 4);
        expect(v.x).to.be.closeTo(0.5, 0.01);
        expect(v.y).to.be.closeTo(0.333, 0.01);
        expect(v.z).to.be.closeTo(0.25, 0.01);
      });
    });


    suite('with arglist', function () {
      test('multiply the x, y, z with the scalar', function () {
        v.mult(2, 3, 4);
        expect(v.x).to.eql(2);
        expect(v.y).to.eql(3);
        expect(v.z).to.eql(4);
      });
    });

Please let me know if there are other tests that could be added. I am hopefully very close to a PR on the ongoing work, so review on that would also be super helpful as well, when it's ready. Thanks again for spotting and reporting.

[avinxshKD]

Hey thanks for checking in @ksen0 makes total sense, happy to wait for the refactor PR.

For extra test cases, a few edge cases worth covering:

• v.div(2, 0, 4) div-by-zero in one component, currently silently gives Infinity
• v.mult(2, 3) / v.div(2, 3) on a 3D vector fewer args than dimensions, what should z do?

And happy to review the PR when it's up 🌸

[avinxshKD]

@ksen0 @davepagurek and Also if you get a chance, would love a review on #8517, been sitting for a couple weeks.