[Feature Request] Bracket Pair Colorization: Force unique opening color

[Stanzilla]

This would be porting over said feature from Bracket Pair Colorizer 2

"bracket-pair-colorizer-2.forceUniqueOpeningColor"

[vscodebot]

(Experimental duplicate detection)
Thanks for submitting this issue. Please also check if it is already covered by an existing one, like:

• Investigate into the performance issues of the bracket pair colorizer extension (#128465)
• Bracket Pair Colorization: Getting Started View (#130557)
• Bracket Pair Colorization: specify where to configure colors (#130993)

[hediet]

How does this feature work? Why "opening"? Do you have an example that demonstrates why it is useful?
What about ()[]()?

[Stanzilla]

Basically it makes sure that each bracket pair is a different color instead of only having a color switch when they start nesting.

[NatoBoram]

It would be nice to have it along with "bracket-pair-colorizer-2.forceIterationColorCycle".

[hediet]

What's the difference between forceUniqueOpeningColor and forceIterationColorCycle?
Are really both settings needed?

[Stanzilla]

I'm used to opening color but I guess could get used to color cycle

[hediet]

So forceUniqueOpeningColor just cycles between the first and second color, while forceIterationColorCycle cycles between all colors?
It seems to me they both solve the same issue - thus I only would like to implement one of them.

[Nahor]

IMHO, forceIteration has the drawback that the last nested pair might be the same color of the outer pair. With forceUnique, there is a way to ensure that the inner and outer pairs have different colors (I can't remember if it was implemented that way in the original extension though).

[hediet]

I still don't understand the behavior of the extension with forceUniqueOpeningColor:

---

I think a setting editor.bracketPairColorization.distinguishAdjacentPairs: boolean might be cleaner.
It would ensure that no adjacent pairs get the same color (which is not the case for the screenshot).

Unfortunately, this could happen then:

( () () () )  -  brackets
r gg bb rr r  -  computed colors

The last two closing brackets have the same color.
We could just avoid the parent color, but we have to see how that works with independentColorPoolPerBracketType.
Probably a bracket pair should then avoid the color of the closest ancestor with the same type.

[Nahor]

The original extension was a bit buggy and some corner cases were broken. I don't remember the exact scenario but I do remember that in some cases, consecutive brackets could have the same colors, although my memory was between a parent and its last child (...or maybe I was using forceIteration rather than forceUnique...?🤔).

We could just avoid the parent color

+1 on that, although I would even like it better if the pair would take the same color has the parent when the parent has no other characters but that pair (😛, I'm still hoping it will happen one day).

[clari7744]

I'd like to bring attention back to this issue since this is the only thing I've found in my searching and it seems to have fallen by the wayside a bit, and just wanted to add in that I would also like to see the forceIteration setting come to the native feature.