[Slider] Add new tick visibility modes

[pubiqq]

• Resolves [Slider] Hide dots instead of limiting the number #1233 (taking into account the wishes of [Slider] Hide dots instead of limiting the number #1233 (comment))
• Partially resolves [Slider] Stop indicators don't fulfil their purpose when the slider has too many acceptable values #4237

Highlights

• New tick visibility API: Slider.getTickVisibilityMode(), Slider.setTickVisibilityMode(int), and app:tickVisibilityMode.
• New tick visibility modes:
  • visibleAll - All tick marks will be drawn, even if they are spaced too densely and overlap each other.
  • autoLimit - (Default) All tick marks will be drawn if they are not spaced too densely. Otherwise, the maximum allowed number of tick marks will be drawn.
  • autoHide - All tick marks will be drawn if they are not spaced too densely. Otherwise, the tick marks will not be drawn.
  • hidden - Tick marks will not be drawn at all.
• Slider.setTickVisible(boolean) is deprecated in favor of Slider.setTickVisibilityMode(int).
• app:tickVisible is deprecated in favor of app:tickVisibilityMode.

Migration

• Slider.setTickVisible(true) -> Slider.setTickVisibilityMode(TICK_VISIBILITY_AUTO_LIMIT)
• Slider.setTickVisible(false) -> Slider.setTickVisibilityMode(TICK_VISIBILITY_HIDDEN)
• app:tickVisible="true" -> app:tickVisibilityMode="autoLimit"
• app:tickVisible="false" -> app:tickVisibilityMode="hidden"

[drchen]

Thank you so much for making the change!! Sorry that I have been pretty occupied recently and forgot about the change.

[pubiqq]

No problem. Thanks a lot!

[drchen]

Sorry for the delay. I'm working on this now. : )

[drchen]

I'm running into very weird issues when trying to merge the PR. Somehow the change blocks the argument validation from being done when running unit tests.

I don't really have time to debug it yet. If you can help check that on your end it will also be very helpful.

[pubiqq]

Checked, fixed, and rebased.

[drchen]

Hey thanks for bumping this!! Sorry I missed your fix during my vacation.

Sending this for internal review now.

[drchen]

Hi, during the internal review, people raised that we are not sure what is the use scenario of autoHide mode?

Do you have an example of that?

[pubiqq]

Do you have an example of that?

#1233

As for me, I don't like the current default behavior either, because in the "dense" case ticks stop pointing to the position you can drag the slider to.

[drchen]

Dan, can you provide some insights here? I think the latest conversation we had regarding this is we probably need designers' opinion. ^^;

[pubiqq]

Just in case, I want to make it clear that this PR doesn't change the default behavior of the slider, but only adds two new strategies - autoHide (reason #1, reason #2) and visibleAll (I know this approach is not recommended by M3 specs, it's added mostly for symmetry with hidden).

[paulfthomas]

@drchen any update on this? I talked with our designers and they are not keen on adding this change, especially visibleAll, maybe removing it would be better.

[pubiqq]

If you want, I can delete visibleAll. I already explained why it was added, but I don't need it either and I don't insist on keeping it.

However, the autoHide should remain, that's what this PR was created for (explanation). Moreover, I believe the new behavior should be made the default one because the current behavior is not just illogical and confusing but directly contradicts the description of stop indicators.

[paulfthomas]

Hey, if visibleAll is removed and the default remains as the current behavior I should be able to have this merged.

[pubiqq]

Ok, done.

P.S. It would also be great if someone would look at my other PRs as well. I realize that adding features is more fun than fixing bugs, refactoring, and reviewing other people's code, but I believe it still needs to be done.

[paulfthomas]

I'll take a look and try to merge this PR very soon, I know the delay is frustrating but we have a lot of things to consider internally so it's not just clicking the merge button unfortunately...