Solves issue #4562

[Garima3110]

Solves issue #4562
Changes:
I have introduced a separate variable for pre-erase blend mode, which can maintain clarity in the code and avoid conflicts between the optimization goal and erasing logic. I hope this should make the blend state management more straightforward and easier to maintain.
I would love to have the suggestions of the maintainers to this solution, thanks a lot!

• npm run lint passes
• Inline documentation is included / updated
• Unit tests are included / updated

[davepagurek]

Thanks for taking this on! It looks like npm run test is currently failing, so it seems like right now this accidentally changes behaviour. That might just be from a potential typo in the code that I commented on, could you see if that fixes it or if there are still test failures?

[davepagurek]

Should this be curBlendMode without the underscore?

[Garima3110]

Thanks for taking this on! It looks like npm run test is currently failing, so it seems like right now this accidentally changes behaviour. That might just be from a potential typo in the code that I commented on, could you see if that fixes it or if there are still test failures?

Thanks a lot @davepagurek for looking into this!
The tests still fail, I am trying to find out the exact cause , If you have any suggestions please do let me know.! Thanks:-)

[davepagurek]

The error output says that this test is failing, in test/unit/color/setting.js line 94:

test('should cache renderer blend', function() {
  my3D.blendMode(my3D.SCREEN);
  my3D.erase();
  assert.deepEqual(my3D._renderer._cachedBlendMode, my3D.SCREEN);
});

I think this failure is just because you renamed that property, so I believe you'll need to update that test to refer to it under its new name.

[Garima3110]

I think this failure is just because you renamed that property, so I believe you'll need to update that test to refer to it under its new name.

I have made the changes. Please have a look and let me know if some more changes are to be made . Thanku!

[davepagurek]

Thinking this through a bit more, I think this might be slightly mixing the logic a bit. I think this method's intention is to only change the blend settings if the mode has changed from what we were last using, regardless of whether or not that was used for erasing. So I believe we need two stored properties: a preEraseBlend like you have, but also a cachedBlendMode, referring to the last value that _applyBlendMode was called with. So everything in this method would be using a cachedBlendMode, and the erase/noErase ones would refer to preEraseBlend.

[Garima3110]

Thanks @davepagurek for the insights. I'll just look into this and get back to you soon ..!

[Garima3110]

I've made the suggested changes . Please have a look!

[davepagurek]

Thanks, I think this looks good now!

[Garima3110]

Thanks, I think this looks good now!

Thanks @davepagurek for your help on this . Got more insights on the codebase. All thanks to you..! :-)